Wavewizard v0.54b

[johnman]

UPDATE:
old link doesnt work (http://www.needfulthings.webhop.org/) the newest release.
New Download link (http://www.rarewares.org/wavewiz/wavewizardv0.54b.zip)
Look @ this post (http://forum.doom9.org/showthread.php?p=696063#post696063) for info on the most recent release.
An interesting thread (http://forum.doom9.org/showthread.php?t=98131) about ssrc, resampling, using twopass and stuff like that
Some general info about ww in this (http://forum.doom9.org/showpost.php?p=662842&postcount=5) post
------

I made a little program which can do some cool stuf:
-upsampling
-downsampling
-dithering
-channel mixing
-channel swapping
-sampel conversion
-normalising (file/list)
and some other stuff which you may see for yourself

I uploaded it on daphy's site (http://www.needfulthings.webhop.org) where everyone can download it.

If you have any comments, or feature request, post them in this thread. This program is far from finished. New features will be added, and manny bugs have to be fixed.

Currently only standard wav is supported.

It is the first release, and since my english isnt very good, there is no manual. If you want to make one..... contact me.

If you have a question about the program, ask it in this thread so others might benefit from it.

Also i would like to thank Daphy for hosting my program :thanks:

John man

[daphy]

download Wavewizard EFB.NU Freeware Board (http://www.essential-freebies.de/board/viewtopic.php?f=21&t=9228)

[daphy]

-dead link-

[Kurtnoise]

It sounds great...I'll try it asap.

[johnman]

Since im bored im using this old post to describe some of the options of wavewizard (=ww).

Well lets start with the channel mapping feature. ww makes it possible to go from a x channel wav to a y channel wav by "simply" creating a matrix. This can be done on 2 different but very similair ways:
moving and mixing.

Channel moving will just move the channels from one "location" to an other "location".
To illustrate this here is an example (NB the first channel is numbered as 0).
Lets say you have a normal stereo file which you want to convert to a 4 channel wav. The 2 channel wav is an interleaved stream like this:

L1R1 L2R2 L3R3 L4R4 L5R5
This means the wav has first a sample for the left channel (ch0) and then a sample for the right channel (ch1) and then the second sample for the left channel etc....


If you now make a 2 to 4 channel mapping like this:

http://www.rarewares.org/wavewiz/example1a.png

And fill it in like this:
http://www.rarewares.org/wavewiz/example1b.png

It wil convert a 2 channel file to 4 channels by placing channel 0 at channel 0 and 2 of the new file, and channel 1 goes to channel 1 and 3 of the new file. So what you get is this:

L1R1L1R1 L2R2L2R2 L3R3L3R3

If you enable this mapping, and insert 2 channel files, all 2 channel files will be converted to 4 channel files. All other files with a different number of channels are unaffected (unless you enable also other mappings)

Now lets say you dont to just move the channels, but you want to add a third channel by mixing left and right. Well for this you would need to use an other type of mapping: mix

Make a new 2->3 channel mapping like this:
http://www.rarewares.org/wavewiz/example2a.png

and configure it like this:
http://www.rarewares.org/wavewiz/example2b.png

Channel 0 and 1 remain the same, but channel 2 is created by taking the sum of 0.5* ch0 and 0.5*ch1. So this will create a 3 channel file with the third channel a mix from the original left and right channel.


When you leave a channel empty when configuring a mapping, it will be filled with silence. You can also use negative values when configuring a mix mapping.

Whenever you have enabled a channel mapping from i.e. 2->3 channels, all 2 channel files in the list are considered as 3 channel files. So If you want to use cb to convert it to an other format, you must make sure that format supports 3 channels. But dont wory to much about this, since ww will give an error if the formats doesnt support the selected configuration. Since the file is processed as a 3 channel file, this also means the resampling takes longer since it has to resample an extra channel.

If you have created a couple of mappings which you think are usefull, you can save the "mappings.ini" file to a save place. This is a normal textfile which contains the data for all the mappings. The graphical interface only allows the creation of mappings for 1..10 -> 1..10 channels. But if you wanted a channelmapping for more channels there is an other option. Manually edit the mappings.ini to suit your needs. This way you can go from 1..1000->1..1000 channels. They are accepted by ww, but cant be viewed or edited graphicaly.

If you want to merge an old mapping with a new one, you must close ww, add a new mappings file with the name "mappingsNEW.ini" to the directory and start up ww. If ww has read the file it will rename it to mappingsIMPORTED.ini.

When you are merging files, the file order will be the same as the order in the list. If you want to change the order, you can simply select a file (or multiple) and drag them up or down to change the order.

Well that about all i can say about channelmapping.
If im bored again ill describe the other features (below this post).

----

Edit a new day has arrived.... And this time i'll describe the preferences.

- The first option is to select the output directory. Well nothing to say about that, except that whenever temp files are used (see normalization) the output directory selected will also be used to store the temp files.

- Process delay
When you add a file with the name "somefileDELAY 1500.wav"
ww will add a delay of 1500 msec to the file when this option is enabled. When using negative values it wil cut away samples from the begining.

- Maintain length
If you want to add a delay but dont want to change the length of a file check this option. It will add silence or remove samples from the end of the file depending on if you added a negative or positive delay

- 32bit float CE compatible
For some reason somebody had the bright idea to make files with a 32 bit int wav header and then to fill it with floats. Since there is no way to tell from the header if its a 32bit int or float, ww can never be sure if its dealing with a 32 bit int or float. If you enable this option it will analyse the wav data to see what kind of file it is. Since this is just guessing, you should only enable this option when your conversion comes out wrong. If you disable this option all 32 bit wav pcm files will be treated as 32 bit int wav files.

- ignore size in header
If you add a wav file with an invallid size in the header, the file will not be recognized by ww. To ignore the size in the header enable this option. If you do this all data after the data chunk will be used as sampledata. When you create files bigger then 4gb ww will set the size (in the wavheader) to 0 since the wav header cant store the real size for such large files. So when you want to work such large files you should enable this option.


- enable channelmapping
This will enable or disable all channel mappings. This same option is also in the channelmapping configuration dialog.

- Stream manipulation
* Mono streams
will convert all mutlichannel files to monostreams

* Stitch files
Will stitch all file together in the order as they appear in the main dialog. All files must have the same sampletype, samplerate and number of channels. It is possible to use a channelmaping, convert the sampletype or rate to let all the files have the same properties. If you add i.e. 5 stereo files into the list and stich them together, you get a single stereo file which has the length off all the files combined.

* Merge files
This option will merge all files in the main dialog to a single multichannel file with the length of the longest file, and with the number of channels equal to the total number of channels from all files in the dialog. So selecting merge on 6 mono files will create a single 6 channel file with the length of the longest file in the list.

- Output format
Not really much to say about. Select wavePCM for maximum compatibility.

- Convert sampletype
...

- Convert samplerate
...


- Sound level control
There are 3 different ways to influence the sound level:

* amplify
Will increase (or decrease) the volume on all files with the number of dB entered in the dB box.

* Normalize
Will first convert find the highest peak, and adjust the volume so the peak will get the desired level. Normalising can be done by file,list or whenever clipping is detected.

* Normalize with temp file
This wil function exactly the same as the normalizing without temp file. But because the result from the first pass is saved (in 64 bit resolution) it only has to convert the sample type in the second pass. This can greatly increase the speed when normalizing and resampling. One exeception to this is the normalise when clipping. Since ww cant know in advance which file will clip, nromalizing with temp file + on clipping will always do two passes. If after the first pass no clipping is detected, it will convert the temp file normaly to the desired output format. If clipping is detected, it will lower the volume.


* dithering
Will dither the output file. If you dont know what it is, try google. It not something i can explain in 2 sentences :rolleyes: . Currently only the dithering amplitude can be changed. A dithering amp of 1 means it will add diher with a max peak of +/- 1 LSB to the file. So for 8 bit int this is 1/128. Dithering will only be applied when
a) sampels are converted to floating point and
b) if the output format is of int type. This is because dithering doesnt make any sence on floatig point.

Well thats all for now :) .

[ursamtl]

John, have you considered implementing ssrc's two-pass mode? My understanding is that it is designed to prevent such clipping by attenuating the level after conversion to prevent potential clipping.

[johnman]

John, have you considered implementing ssrc's two-pass mode? My understanding is that it is designed to prevent such clipping by attenuating the level after conversion to prevent potential clipping.

I had such a feature, but since the peak is shown, and since there is a normalisation option (which obviously prevents clipping) i thought the feature was not really be usefull. But i can always re-insert it. Was the clipping caused by upsampling/downsampling? What was the peak of the original file?

[ursamtl]

The original file had a peak level of 0.97723. Converting it to 48kHz resulted in a peak level of 1.02103. Providing the normalization occurs before the peaks are clipped, it should be ok.

[johnman]

Providing the normalization occurs before the peaks are clipped, it should be ok.

And since this is the case, it IS ok :).

Has anyone tried the channel mapping/mixing features.... i wonder if its clear how they should be used.

EDIT:
Has anyone tried the program at all :(

[johnman]

UPDATE v0.43b

It has been a while since the last release, but now there is a new version. This one can convert to 9 different formats. The only limitations are the limitations of each format itself. So it is i.e. possible to create a 100 channel ogg if you want. Because of the way wavewizard works, it can now also convert i.e. 10 stereo files to 10 dts quadraphonic files using surcode without the need for making 6-channel wavs first. And this can be done without editing a textfile or changing a commandline. Just configure wavewizard once and let it work on all the files you have added to the program.

The conversion to the other formats are done with the help of external programs. But normaly you dont have to edit a single commandline, or even look at those dos programs running. Everything is done from the GUI.

Although its not very hard to use, the usage is a little different compared to other programs. So im again hoping someone wants to make a manual for this program. If you want to do this, just send me a PM. Ill email the program so you can start immediately :).

EDIT: Someone responded to my request so no manual writer is currently needed.

Wavewizard now also supports wav's > 4gb and there have been a couple of minor bugfixes.

The programs which are currently supported are :

surcode cd (dts)
surcode dvd (dts)
twolame (mpeg 2)
ape (monkey's audio)
flac
wavpack
nero fends (frontend for nero aac)
MPC
lame (mp3)
Ogg

And as a bonus, all freely distributable programs (everything except surcode) can be downloaded from within the program, so no need to search and download the correct version with a browser.

Current issues:
- i tested this version and it does not run on winme. So i dont think it will run on win98 either. The good news is i already fixed the problem, and the next version should run normaly on winme (and possibly win98).

- The settings for nero fends in ww are completely ignored by cb.

- Mpc settings dont work correctly

- Athough ww accepts mpeg audio files (like mp3) it can absolutely NOT decode them.

[Zyphon]

Sounds like a neat app, I will have to test it when when I get back from my holiday.

[johnman]

Sounds like a neat app, I will have to test it when when I get back from my holiday.

Great, i hope you give some feedback when you tried it.


Something else.....
I added a screenshot but it still has the text "Attachments Pending Approval" above it. Do i need to notify a mod or something, or do they automaticaly see when a file is waiting for approval?

(EDIT maybe ww will soon also have support for the improved version of ac3enc.dll. So this might be the 11th supported "program")

[Zyphon]

(EDIT maybe ww will soon also have support for the improved version of ac3enc.dll. So this might be the 11th supported "program")

This sounds good, so will it be able to make a nice ac3 file from a wav file then?

Good luck with the updates. :)

[johnman]

This sounds good, so will it be able to make a nice ac3 file from a wav file then?

Yep :) (and maybe the 12th app supported will be a commercial ac3 encoder)

But currently im looking into vst support.

[Zyphon]

Yep :) (and maybe the 12th app supported will be a commercial ac3 encoder)

But currently im looking into vst support.

Cool can't wait to see this version. :)

[tebasuna51]

This is a verified channelmapping for ac3:
6->6#move#inactive#ac3 order for SoftEncode or ac3enc
0 2 1 4 5 3

And this other for aacenc32.exe, not necesary for Nero plugin or Faac:
6->6#move#active#aac order for aacenc32.exe (Not for Nero plugin or Faac)
2 0 1 4 5 3

[johnman]

This is a verified channelmapping for ac3:
6->6#move#inactive#ac3 order for SoftEncode or ac3enc
0 2 1 4 5 3

And this other for aacenc32.exe, not necesary for Nero plugin or Faac:
6->6#move#active#aac order for aacenc32.exe (Not for Nero plugin or Faac)
2 0 1 4 5 3


thanks for the mappings, ill add them in the next release :)

One question about the 2nd mapping, my program supports nero Nero Frontends (fends), how is the mapping needed for this program in relation to the second mapping you gave. (im not really into aac encoding )

[tebasuna51]

Sorry, I don't know Nero Frontends.
I test Belight (BeSweet, interface bsn.dll, and aac.dll, aacenc32.dll from Nero) and don't need channelmapping, the normal wav order (L,R,C,LFE,SL,SR) make a correct aac.

[KillaByte]

Dude, your program rocks :thanks:

I thought there would be no way to fix the channel order of my ripped DVD-A files with free tools - until I found your program. Can't wait to see the new version.

One thing that might be useful to add in a future version is WAVE_EX support. I fiddled around a bit with the Multi-Channel ToolKit (http://www.bath.ac.uk/~masrwd/mctools.html) and wavewizard was unable to read the resulting files.

Keep up the good work :)

[johnman]

Dude, your program rocks :thanks:

I thought there would be no way to fix the channel order of my ripped DVD-A files with free tools - until I found your program. Can't wait to see the new version.

One thing that might be useful to add in a future version is WAVE_EX support. I fiddled around a bit with the Multi-Channel ToolKit (http://www.bath.ac.uk/~masrwd/mctools.html) and wavewizard was unable to read the resulting files.

Keep up the good work :)

Thanks for the compliment :)

I know i should support WAVE_EX, and i looked at it before (look @ the disabled option in the preferences) but when i was busy with this format i didnt have any vallid wave_ex files to test with, and since then i kind of forgotten it. But since it is called wavewizard it is high on my priority list again. And when im working on it, ill make sure my program will swallow everything mctools can produce :).

BTW if you think your mapping(s) might be usefull for others, you might consider posting it in this thread. The mappings are just plain text in the file "mappings.ini".

---

If anyone is interested, i have played a little with vst's and i succesfully called some vst functions from several vst's. So wavewizard might become a VST host sometime in the near future.

[KillaByte]

BTW if you think your mapping(s) might be usefull for others, you might consider posting it in this thread. The mappings are just plain text in the file "mappings.ini".Sure. The mapping for converting "LF, RF, LS, RS, C, LFE" (as used for DVD-Audio, for example) to "LF, RF, C, LFE, LS, RS" (WAV multichannel standard) is "0 1 4 5 2 3"

6->6#move#active#LF RF LS RS C LFE to LF RF C LFE LS RS
0 1 4 5 2 3

For more information about the WAVE_EX format, have a look at this link (http://www.microsoft.com/whdc/device/audio/multichaud.mspx).

[johnman]

Sure. The mapping for converting "LF, RF, LS, RS, C, LFE" (as used for DVD-Audio, for example) to "LF, RF, C, LFE, LS, RS" (WAV multichannel standard) is "0 1 4 5 2 3"

6->6#move#active#LF RF LS RS C LFE to LF RF C LFE LS RS
0 1 4 5 2 3

For more information about the WAVE_EX format, have a look at this link (http://www.microsoft.com/whdc/device/audio/multichaud.mspx).
Thank you, ill add it to the next release under a different name if you dont mind:

6->6#move#active#DVD-A to multichannel wav
0 1 4 5 2 3

About that link from msdn, i already have been on that page many many times :). But thx anyway.

Edit and i am happy to report ww can now read WAVE_EX files. It still cant write them, and i dont see any real reason to do so. If anyone knows a good reason, ill add that to.

[KillaByte]

Edit and i am happy to report ww can now read WAVE_EX files.Groovy ;)
It still cant write them, and i dont see any real reason to do so. If anyone knows a good reason, ill add that to.One reason I could think of is that with WAVE_EX multichannel files, applications don't have to guess the channel order - they can rely on the information stored inside the WAV. WavPack for example uses the channel information stored inside WAVE_EX files to decide about the channel grouping.

Another cool feature would be a metadata viewer/editor (it would be very interesting to be able to see what applications store in addtional chunks within the WAV)

[johnman]

Groovy ;)
One reason I could think of is that with WAVE_EX multichannel files, applications don't have to guess the channel order - they can rely on the information stored inside the WAV. WavPack for example uses the channel information stored inside WAVE_EX files to decide about the channel grouping.

:goodpost: I didnt know that. But i was thinking.... if people made a wavEX out of a normal wav, ww has no clue about the channels which have been used, so im thinking to only enable conversions from wavEX -> wavEX when no channel-mappings are enabled. Or i could just set the channel mask to 0 to indicate i dont know what channels are available, but i dont know if that is allowed, and i think some programs might crash if i do that...... im still not really sure how im going to solve this..... maybe ill have a look at other programs that make wavEX and get some inspiration :cool: .


Another cool feature would be a metadata viewer/editor (it would be very interesting to be able to see what applications store in addtional chunks within the WAV)


Well that is indeed a splendid idea. I was thinking about this before (there is a repair soundfile menu item IIRC in v0.2b), but i postponed it because my wav parser wasnt 100% according to the specifications. It accepted almost all regulars wav's, but not all. Although i havent found a wav (until i used mctoolkit) that wasnt accepted by ww, i knew some legal wavs might actually not be recognised. Since i now also have updated the wav parser, i already decided that i finaly could make some kind of editor to edit some of the data in the wavheader.

BTW in my tests i noticed almost all programs fail on some wavs that are perfectly legal according to the specifications. Even an advanced program like foobar2k makes at least one mistake when reading wavs. So doing to much crazy stuff with the chunks might not be a very wise thing to do.

[daphy]

New application: Wavewizard (new version: v0.4b3)

--------------------------------------------------------------------------------

UPDATE:
now it can convert to 10 different formats. Look @ post at the bottom of the page. NB the newer version is not available at this moment, but you can take a look at the screenshot .


is now online at EFB.NU Freeware Board (http://www.essential-freebies.de/board/viewtopic.php?f=21&t=9228) :)

[johnman]

is now online at needfulthings (http://www.needfulthings.webhop.org) :)

Great, thanks daphy :).

NB this version is a couple of days old and still cant read wavEX files.

[daphy]

@ johnmann

plz check PM :rolleyes:

[johnman]

@ johnmann

plz check PM

Reply send.

---

To anyone that has trouble letting Configuration Batcher find surcode.... try runnning surcode and THEN press find... CB gets the exe path by searching for a running instance of surcode.

(and im working on an option to manual point to the executable now)
edit Done :)

---
And to know if the conversions are done correctly... CB creates a log which shows some basic info.

[tebasuna51]

I can't send jobs to Conversion Batcher. How do it?

[johnman]

I can't send jobs to Conversion Batcher. How do it?

I dont exactly understand what's the problem, since your description is a little vague. You cant send jobs to conversion batcher yourself. This can be done only by ww. Just add some files to ww, configure the cb settings in ww and press convert. Then the jobs should show up in cb.

Do you get an error message (please post it), or do the jobs dont show up in cb?

There is an option called: "send jobs to batcher". Have you enabled it?

Does cb start when you press the "convert" button in ww?

[tebasuna51]

After my first post i have a error message (can't find msvcrt71.dll). I copied this file in the bin folder and after I have 5 copies of my job. It run now.

I test the aac encode and confirm you: need the aac mapping :

6->6#move#active#multichannel wav to aac
2 0 1 4 5 3

[johnman]

After my first post i have a error message (can't find msvcrt71.dll). I copied this file in the bin folder and after I have 5 copies of my job. It run now.

I test the aac encode and confirm you: need the aac mapping :

6->6#move#active#multichannel wav to aac
2 0 1 4 5 3


ww and cb dont need any external dll's on a normal winxp instalation. I checked the other programs which are run by cb and i found out aacenc32.exe needs MSVCR71.dll. It might be that other programs also need extra dll's.

And :thanks: for the confirmation , ill add it in the next release.

---

A general hint to everyone using this program: cb does not check if the conversions have been performed succesfully. It just runs the jobs, and thats all. But normally ww only sends jobs when they really can be preformed. And since ww creates the wav files which are converted, there isnt anything that can go wrong once everything is correctly configured. But there is no check if the programs really run, so if a dll is missing you dont get an error from cb or ww.

But the file "logCB.txt" is created to give some feedback.

[KillaByte]

But i was thinking.... if people made a wavEX out of a normal wav, ww has no clue about the channels which have been used, so im thinking to only enable conversions from wavEX -> wavEX when no channel-mappings are enabled. Or i could just set the channel mask to 0 to indicate i dont know what channels are available, but i dont know if that is allowed, and i think some programs might crash if i do that...... im still not really sure how im going to solve this..... maybe ill have a look at other programs that make wavEX and get some inspiration :cool: .
Using a channel mask of 0 seems to be perfectly legal. If you create a WAVE_EX using interlx from the mctools package the resulting file will have such a channel mask until you put it through copysfx and assign one.

One possibility would be to use an advanced channel mapping dialog when converting to WAVE_EX where one can assign the logical channel names like C, LFE, etc. to the numbered channels of the source WAV. Wavewizard could then map them accordingly and assign the correct channel mask. When converting from WAVE_EX to WAVE_EX channel mapping could be even more comfortable since one could work exclusively with logical channel names instead of channel numbers.

BTW in my tests i noticed almost all programs fail on some wavs that are perfectly legal according to the specifications. Even an advanced program like foobar2k makes at least one mistake when reading wavs. So doing to much crazy stuff with the chunks might not be a very wise thing to do.
Yeah - but it would be nice if one could do it anyway :D

[tebasuna51]

One reason to use WAVE_EX.

First a precision: dwChannelMask don't inform about "the channel order" but the channel identity. The order is always (mulchaud doc):

1. Front Left - FL
2. Front Right - FR
3. Front Center - FC
4. Low Frequency - LF
5. Back Left - BL
6. Back Right - BR
7. Front Left of Center - FLC
8. Front Right of Center - FRC
9. Back Center - BC
10. Side Left - SL
11. Side Right - SR
12. Top Center - TC
13. Top Front Left - TFL
14. Top Front Center - TFC
15. Top Front Right - TFR
16. Top Back Left - TBL
17. Top Back Center - TBC
18. Top Back Right - TBR

Then dwChannelMask inform what channels are present.

Now a posible use:
When you demux a multichannel wav in monowaves (Stream manipulation-Mono streams, for edit purpose, for instance) your names are suffixed with ch0, ...
If you have read a WAVE_EX you can safely suffix with FL, FR, ... And if you write WAVE_EX you can identify in dwChannelMask each monowav.
After, when you need remux (Stream manipulation-Merge files) you don't need keep in mind the channel order, the dwChannelMask of each monowave inform to your soft.

But for me is enough ww can read WAVE_EX >4 GB (with "Ignore invalid wav size", I suppose) to demux big 6wav files generated with faad.

"and i am happy to report ww can now read WAVE_EX files."
I am happy also and wait for your new beta.

Thanks.

[johnman]

Using a channel mask of 0 seems to be perfectly legal. If you create a WAVE_EX using interlx from the mctools package the resulting file will have such a channel mask until you put it through copysfx and assign one.

Great, thats one thing less to wory about.


One possibility would be to use an advanced channel mapping dialog when converting to WAVE_EX where one can assign the logical channel names like C, LFE, etc. to the numbered channels of the source WAV. Wavewizard could then map them accordingly and assign the correct channel mask. When converting from WAVE_EX to WAVE_EX channel mapping could be even more comfortable since one could work exclusively with logical channel names instead of channel numbers.
Good suggestion, ill use it in the next release :)


Yeah - but it would be nice if one could do it anyway :D
I think so to so im already working on it. Maybe the next version will already have a readonly window which shows all the chunks in a wav.

[johnman]

One reason to use WAVE_EX.

First a precision: dwChannelMask don't inform about "the channel order" but the channel identity. The order is always (mulchaud doc):

My english isnt as good as i would like it to be. But you're absolutely correct.


Then dwChannelMask inform what channels are present.

Now a posible use:
When you demux a multichannel wav in monowaves (Stream manipulation-Mono streams, for edit purpose, for instance) your names are suffixed with ch0, ...
If you have read a WAVE_EX you can safely suffix with FL, FR, ... And if you write WAVE_EX you can identify in dwChannelMask each monowav.
After, when you need remux (Stream manipulation-Merge files) you don't need keep in mind the channel order, the dwChannelMask of each monowave inform to your soft.

But for me is enough ww can read WAVE_EX >4 GB (with "Ignore invalid wav size", I suppose) to demux big 6wav files generated with faad.

"and i am happy to report ww can now read WAVE_EX files."
I am happy also and wait for your new beta.

Thanks.
Good idea:) . I already wanted to give correct naming before, not something like ch0, ch1 etc and waveEx makes that possible :).

The next version may still take a couple of days. I have noticed a couple of problems which im trying to solve. The most important one i found out is that the settings for nero fends in ww are ignored by cb.

[johnman]

Before releasing i would like to test my program a little with wavEX files and 64 bits wavs. I already used mc tools and got some files from this (http://www.tsp.ece.mcgill.ca/MMSP/Documents/AudioFormats/WAVE/WAVE.html) page.

Maybe somebody knows a program or a site to get some more files to test with?

[johnman]

UPDATE v0.5xb


The programs which are currently supported are :

surcode cd (dts)
surcode dvd (dts)
twolame (mpeg 2)
ape (monkey's audio)
flac
wavpack
nero fends (frontend for nero aac)
MPC
lame (mp3)
Ogg
sonic foundry soft encode (ac3) new
DA's ac3enc new

The new version includes a couple of new features (and probabely new bugs since i have rewritten a lot of code (again) ):

- logging option in cb. This shows the commandlines which are used and some other interesting stuff. This can also be usefull if someone wants to make his own program for sending jobs to cb.

- supports now the ac3enc.dll from Dark Avenger which should be better then the old one. ( :thanks: DA for your help)

- support for sonic foundry ac3 encoder

- support for non PCM and wav ex.

- if a channelmask is available in the wave ex and a wav is splitted to monostreams, it now gives the correct channelnames.

- an option to automaticaly import new channelmapping. If you add a fille "mappingsNEW.ini" to the ww dir it will be be imported and renamed to "mappingsIMPORTED.ini". The new ww comes with a mappingsNEW.ini in the dir so you can combine it with the one you have. Remember, ONLY the mappings file can be used in different version, DONT copy the preferences from old to new version.

- option to avoid clipping

- fast switching between the different config dialogs is now possible by using F2,F3,F4 and escape (for cancel).

- Option to create a tempfile when normalising. This is usefull when resampling, since the result will be saved and the second pass it uses the tempfile to normalise. But this option might create huges files.... so becaurefull when using it. And when you are doing a "normalize list" with tempfile creation, all filles will first be converted to a tempfile, so make sure you got enought diskspace when using it !

- some other other stuff i dont remember.

If daphy's site is online again ill upload a new version.

Current issues:

- support for sonic foundry is available, but not reliable. Somehow sonic doesnt read the settings correct from the .ini file which is created.

- in v0.5b there was a problem with the detection of surcode dvd. Thx to daphy this has been fixed in v0.52b.

-in v0.52b there was a problem when converting big files. Thx to tebasuna51 this problem is fixed. (look in the forum for details)

-there is a bug in sonic foundry softencode which will disable LFE when it is set to On en it is enabled when it is set to Off. Thx to tebasuna51 this problem got a workaround by adding an option in CB. (look in this forum for details)

- The "save output" in preferences should not be unchecked. I had removed this option, but when going back to an older backup version of my prog, i forgot to remove it again. When this option is disabled, the program might not work properly. (removed in v0.53b)

- Not every setting gives correct results when using da's ac3enc, and sometimes it doesnt run 100% stable.

- When cb is used on winme and a dos program is canceled by pressing stop, winme asks if it should terminate the program. You should press "Yes" manualy. If the window is hidden, you dont see this message, which means you have to check "show programs" to see it. I think i can fix this, but i doubt if this is really usefull since i dont think many people are using winme anyway.

-when resampling a file with more then 100 channels the memory consumptions becomes very significant (500MB+) when resampling a file with more then 500 channels ww might crash.

Things im considering to implement (in random order)
- vst-hosting
- replaygain
- settings the channelmask manually (for wav-ex files)
- show/edit the chunks inside a wav (showing part is done)
- decent documentation
- dynamic amplification
- equalizer
- decoding for all formats which cb can create
- cue sheets

Changelog

from v0.53 -> v0.54
- changed the option invert LFE in cb to invert specific settings. Now 5 settings will be inverted if this is checked

- Added chunk browser. This also works on avi's

- changed the location from where programlist.txt is downloaded in cb.

- some other minor fixes.

[tebasuna51]

Thanks for your job.
I wait for daphy's site to test it.

[Mug Funky]

If anyone is interested, i have played a little with vst's and i succesfully called some vst functions from several vst's. So wavewizard might become a VST host sometime in the near future.

cool. there's a big need for free VST hosts. "bidule" is a nice one, but you have to register that and all... i like the graphedit style interface though.

[ursamtl]

Mug, if you're just doing stereo stuff with VSTs, there are a couple of options already around for free VST hosts. One of the nicest I've seen is Kristal Audio Engine, which is like a Cubase clone with two VST slots per audio track and then two VST master slots at the end. There are a couple of other free host progs around too if you check on www.kvr-vst.com. I also look forward to John's implementing VST hosting since WW is rapidly turning into an app that ROCKS!! :)

[Zyphon]

@johnman

Thanks for the new update and the new support for sonic foundry soft encode much aprpeciated. :)

Keep up the great work. :)

[johnman]

I have uploaded it to daphy's site.... i dont think it will take long before it will be available.

@ ursamtl as i said in the other post, i added an option to do twopass resampling. (use normalize using tempfile + check when clipping)

@ Mug Funky
vst hosting is somthing i realy want to do, but i think it will take a while. I want to make a solid design, or ill have a hard time maintaining it.

@Zyphon
Im glad you like it :).

@all i hope you enjoy the new release, and if you find a serious bug, please post it and ill make a fix asap (probabely within 2 days).

And i would like to ask if some of you would like to check if the commandlines used by cb are correct. For this purpose i made a logging window in CB (not ww) so you can see immediately if a setting changes in ww, what the resulting command line will be. You can also look at it in the tiny textbox in ww, but this is obviously not as easy since its very very small.
Im not an expert on the usage of the supported commandline programs, so im not 100% sure i did everything correct.

To do this:
- start up ww and cb.
- got to the cb settings in ww and select configure i.e. lame
- enable the logging window in cb
- check the "message log" radio button on the logging window
- make the logging window as wide as possible so you can see the full commandline without horizontal scrolling
- now just change the settings in ww and look at the commandline in cb to see if they match up.

[daphy]

uploaded ;)

Questions:

-which ac3enc.dll do you use? the one from besweet or the one out of the DA package (I guess this one)
if DA, it´s not avaible as single download - maybe we should at it on needfulthings for automated download or generally put inside the package ;)
- I am not sure but plz check the finding routine on surcode DVD :confused:

[johnman]

uploaded ;)

Questions:

-which ac3enc.dll do you use? the one from besweet or the one out of the DA package (I guess this one)
if DA, it´s not avaible as single download - maybe we should at it on needfulthings for automated download or generally put inside the package ;)
- I am not sure but plz check the finding routine on surcode DVD :confused:

thx for the upload.

For ac3enc.dll look at the current issues for this version.

The routine for finding surcode is not changed, and it works 100% for me. Did you maybe chang the directory name/executable name to something else?
(and there is always the option to find it manually)

[daphy]

hm, the main problem is for me to find the mentioned file (ac3enc.dll with that date) at all (on the German board I´ve only found one from former times, maybe I missed something :confused: )-> I guess most of those guys will try ww with the one from besweet, so no serious bugreport can be expected :eek:

[johnman]

Ok, i couldnt find it either, but DA said it was on the german board in one of the packages. But im not to good in reading german so i might have missed it. Im going to post it in a sec, But i must explicitly mention it is a BETA version.

The one i used, i got by mail, so i dont know where else it is available.

Thx for bringing this to my attention :)

[daphy]

okay

as part of the package or as seperate download (manually or via list)?

[johnman]

Here is a link to a beta ac3encoder dll which will be removed when the attachment is enabled.

edit: removed
edit2: newer version (the 12.9 KB one):

[tebasuna51]

Good job Johnman. I make any test with WW v0.52b:

a) wav6 -> aac (aacenc32.exe Nero Fends)

- Channel Mapping: Enable (multichannel wav to aac)
- Stream manipulation: Disable
- Output Format: Wave PCM (Don't work with wav_ex)
- Sample Type: Work only with: 16bit int

b) wav6 -> ac3 (ac3enc.dll)

- Channel Mapping: Enable (ac3 order for SoftEncode or ac3enc)
- Stream manipulation: Disable
- Output Format: Wave PCM (Don't work with wav_ex)
- Sample Type: Work only with: 32bit float IEEE
- Notes:
Don't work with ac3enc.dll from HeadAC3he v0.24-a13 2004-12-15.
Work with ac3enc.dll 2005-08-09 in your previous post.

c) wav6 -> ac3 (Soft Encode)

- Channel Mapping: Disable (See Configure notes)
- Stream manipulation: Enable (Mono Streams)
- Output Format: Wave PCM (Don't work with wav_ex)
- Sample Type: Work at least with: 16bit int, 32bit int, 32bit float IEEE
- Configure notes:
1) Channel config: with L-0, R-1, C-2, LFE-3, LS-4, LR-5 is not necessary enable the ww Channel Mapping (only necesary when you make a re-ordered wav6 to be open directly in SoftEncode)
2) LFE Enable (necessary for 5.1). Then you insert, in SonnicAc3.ini, "UseLFE=On". SoftEncode make a 5.0 ac3, because a bug or a bad chosen name need "UseLFE=Off" to make a 5.1 ac3. I use:
Sonic Foundry Soft Encode Version 1.0 (Build 19)
Dolby Digital Encoder Version 6.2.2 December 2, 1997

d) wav_ex 6chan <-> 6 mono wav_ex

Correct split in 6 mono wav_ex preserving the ChannelMask.
Incorrect Merge of 6 mono wav_ex ignoring the ChannelMask.

e) wav > 4GB

Opened when enable "Ignore invalid wav size". Length limited to 4GB. I think, in this case, the length must be calculated with the file size. I know the problem with the extrachunks at the end of file, but is better to have little clicks at the end, than lose MB of correct data.
When split in 6 mono wav only the first 4 GB are converted, the rest are ignored.

[johnman]

Thx for your testing report. Its very detailed, which is very helpfull to narrow down those nasty bugs.

Since a and b seem to work im going straigth to c.



c) wav6 -> ac3 (Soft Encode)

- Channel Mapping: Disable (See Configure notes)
- Stream manipulation: Enable (Mono Streams)
- Output Format: Wave PCM (Don't work with wav_ex)
- Sample Type: Work at least with: 16bit int, 32bit int, 32bit float IEEE
- Configure notes:
1) Channel config: with L-0, R-1, C-2, LFE-3, LS-4, LR-5 is not necessary enable the ww Channel Mapping (only necesary when you make a re-ordered wav6 to be open directly in SoftEncode)
2) LFE Enable (necessary for 5.1). Then you insert, in SonnicAc3.ini, "UseLFE=On". SoftEncode make a 5.0 ac3, because a bug or a bad chosen name need "UseLFE=Off" to make a 5.1 ac3. I use:
Sonic Foundry Soft Encode Version 1.0 (Build 19)
Dolby Digital Encoder Version 6.2.2 December 2, 1997


1)The channelmappings supplied are from other users, they are not tested by me. I added them since they might be usefull to others.
2)If i understand you correctly, you are saying i have to set the UseLFE to off to enable it? Im checking it out myself to see whats going on.


d) wav_ex 6chan <-> 6 mono wav_ex

Correct split in 6 mono wav_ex preserving the ChannelMask.
Incorrect Merge of 6 mono wav_ex ignoring the ChannelMask.


Only the splitting preservers the mask, when merging or stitching the mask is set to 0. This is done intentionaly, because what should ww do if 2 files with the same mask were added. So to quickly solve this is just set the mask to 0.
I am thinking about an option to manualy set the mask, and give a hint based on the existing masks.



e) wav > 4GB

Opened when enable "Ignore invalid wav size". Length limited to 4GB. I think, in this case, the length must be calculated with the file size.

Whenever the size in the header is ignored, everything behind the start of the datachunk is treated as sampledata. So it should read until the end of the file.


I know the problem with the extrachunks at the end of file, but is better to have little clicks at the end, than lose MB of correct data.
When split in 6 mono wav only the first 4 GB are converted, the rest are ignored.
If not all data is converted, then something is wrong. ww doesnt cut away a couple of MB's just to be sure you dont get a pop. I just created a 6 channel file of 6 gb and splitted it into 6 monostreams.. Alls of them have the correct length :confused:
Are you sure you had enough diskspace?

[tebasuna51]

1)The channelmappings supplied are from other users, they are not tested by me. I added them since they might be usefull to others.
This ww channelmapping is supplied for me, before know the channelmapping inside the Configure of Soft Encode. Is only to explain when must be used or not.
2)If i understand you correctly, you are saying i have to set the UseLFE to off to enable it? Im checking it out myself to see whats going on.
Yes. I tested it. Maybe a bug of my version of Soft Encode, Version 1.0 (Build 19)
Whenever the size in the header is ignored, everything behind the start of the datachunk is treated as sampledata. So it should read until the end of the file.

If not all data is converted, then something is wrong. ww doesnt cut away a couple of MB's just to be sure you dont get a pop. I just created a 6 channel file of 6 gb and splitted it into 6 monostreams.. Alls of them have the correct length :confused:
Are you sure you had enough diskspace?
Yes, I work in a NTFS partition with more than 20 GB of free space.
The same file is separated by BeSplit with correct length (but with a error in the BlockAlign field, keeping the wav_ex header with a incorrect Channelmask corresponding to the wav_ex 6 chan.).
The file is a wav_ex generated by Faad with only 4.2 GB (130 min., 48 KHz) and the mono wav extracted have exactly 4/6 = 0.666 GB (124 m. 16 sec., 48 KHz).
The two erroneous fields in the wav_ex 6chan (RIFF_chunk_size and data_chunk_size) have values corresponding to the excess over 4 GB (aprox. 6 min., 48 KHz). Without extrachunks at the end of file.

The only check options in ww Preferences are:
- Ignore invalid wav size
- Save output
- Analyse files inmediately
- Stream manipulation -> Mono streams
- Output format -> Wave PCM

Thanks.

[johnman]

Im just guessing here, but it might be that the size is vallid. If the size is set to 4 GB ww will not use the full file. That is also why ww sets the size to 0. A way to test this is to unselect ignore invallid size, and then to drop the file. If it accepts it, then the size is vallid.

Im going to solve this "bug" by adding the options
- always ignore the size and
- automaticaly ignore the size when its larger then 2gb/4gb

If this is not the problem, it might be handy if you could post the first 100 bytes of the wav. Then ill can experiment with it myself.
it would look something like this:

52 49 46 46 44 CB 36 83 57 41 56 45 66 6D 74 20
10 00 00 00 01 00 06 00 44 AC 00 00 30 13 08 00
0C 00 10 00 64 61 74 61 20 CB 36 83 01 00 01 00
01 00 01 00 01 00 01 00 FF FF FF FF FF FF FF FF
FF FF FF FF 01 00 01 00 01 00 01 00 01 00 01 00
FF FF FF FF FF FF FF FF FF FF FF FF 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00


EDIT and i can confirm the bug in sonic surcode. I have to set the option to off to enable it :confused: I think i will add an option to cb to invert the flag.

EDIT 2 and i got the same verion of surcode as you

[johnman]

I've uploaded ww v0.53 to needfullthings.

No real changes, only a couple of bugfixes
- option to invert includeLFE flag in cb (default is to switch the flag)
- new options to ignore size in header
- and 3 other minor issues.
- the channelmask is still set to 0 when merging/stitching file. This will be changed when i add an config option for settig the channelmask on all wavs created.

please check the new version tebasuna51

[tebasuna51]

Sorry. I mistake my precedent comment. WW don't stop at 4 GB, stop at the invalid length.
I decode a 130m12s.aac with Faad v2.1b to 6faad_ex.wav, and with Foobar2000 v8.3 to 6Foobar.wav.
The two wav are open like Type: Unknown with Ignore invalid wav size Disabled.
With Ignore invalid wav size Enabled can split them in mono wav.
I report the data in WW (I add the first line, aac, to compare length), and the wav headers of two wav 6 channel.


Filename Size Length Channels SampleRate Type
130m12s.aac 112.028 KB 2:10:12.459 6 48000 aac
6faad_ex.wav 4.394.508 KB 2:04:16.540 6 48000 Wav Ext.
6faad_ex_ch1_R.wav 699.051 KB 2:04:16.540 1 48000 Wave PCM
6Foobar.wav 4.394.508 KB 05:55.918 6 48000 Wave PCM
6Foobar_ch1.wav 33.367 KB 05:55.918 1 48000 Wave PCM
-------------------------------------------------------------------------
Offset 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 6faad_ex.wav
------------
0000 52 49 46 46 00 FF FF FF 57 41 56 45 66 6D 74 20 RIFF.ÿÿÿWAVEfmt
0016 28 00 00 00 FE FF 06 00 80 BB 00 00 00 CA 08 00 (...þÿ.._»...Ê..
0032 0C 00 10 00 16 00 10 00 3F 00 00 00 01 00 00 00 ........?.......
0048 00 00 10 00 80 00 00 AA 00 38 9B 71 64 61 74 61 ...._..ª.8>qdata
0064 00 FF FF FF B9 FF 00 00 00 00 00 00 00 00 00 00 .ÿÿÿ¹ÿ..........

FileSize: 4.499.976.260 = 0x10C383044
RIFF_chunk_size = FileSize - 8 = 0x10C38303C Invalid: 0xFFFFFF00
data_chunk_size = FileSize - 68 = 0x10C383000 Invalid: 0xFFFFFF00
-------------------------------------------------------------------------
Offset 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 6Foobar.wav
-----------
0000 52 49 46 46 24 30 38 0C 57 41 56 45 66 6D 74 20 RIFF$08.WAVEfmt
0016 10 00 00 00 01 00 06 00 80 BB 00 00 00 CA 08 00 ........_»...Ê..
0032 0C 00 10 00 64 61 74 61 00 30 38 0C B9 FF 00 00 ....data.08.¹ÿ..

FileSize: 4.499.976.236 = 0x10C38302C
RIFF_chunk_size = FileSize - 8 = 0x10C383024 Invalid: 0x0C383024
data_chunk_size = FileSize - 44 = 0x10C383000 Invalid: 0x0C383000


Maybe WW take like valid anything distinct of 0xFFFFFFFF or 0x00000000, but you can see there are two programs than fill data_chunk_size with different values.
I think if FileSize > RIFF_chunk_size + 8, the Length can be calculated with FileSize - 8 - Offset_data_chunk (ignoring possible extrachunks at the end of the file).

Thanks for your interest.

[johnman]

Thx again for your detailed report. I managed to pinpoint the exact problem, and im pretty sure it is fixed in v0.53. The problem was that there are multiple chunks, which each have their own size . The RIFF chunk is the filesize -8, and because the riffchunk was wrong ww did not accept the file. But the data chunk size was "correct". It is to small, but only a size that's to big, or a size of 0 would be rejected. Since ww sets the size to 0 if an sizeoverflow occures, it will reject the size, but other programs just set the size to an invallid number, so the size was not rejected.

Anyway, if you use the new version and select "ignore the size in the header" you're ok.

And to check if the problem is fixed, you only need to drop the wav into ww. If the length is displayed correct (2:10:12.459) , then it should run fine.

[johnman]

Although i doubt im going to get any help, im gonna ask it anyway :rolleyes:

I just finished a chunk analyser to show all the chunks in the wav, and i would like to test it a little, so again i need some testfiles. If anyone know a site like this (http://www.tsp.ece.mcgill.ca/MMSP/Documents/AudioFormats/WAVE/Samples.html) with some example files i would like to get a link.

And here is a little screenshot:

http://www.rarewares.org/wavewiz/chunks.png

And since im posting anyway, im thinking about letting ww also show all the data of other RIFF formats like avi. Both avi and wav's have the same basic RIFF structure, so it wouldnt be that much works (i think). But this doesnt really fits into ww so im only doing this if there isnt any other tool that can do this. So.... anyone wants to comment about this? Should ww be able to analyse avi's?

[tebasuna51]

I think is enough with the examples in this site. Of course can have many types with obsolete compression and different kinds of parameters and extrachunks, but for me is not necessary.

About the avi I think is a bit more complex. Do you know abcAVI Tag Editor? http://abcavi.tk/

[johnman]

I think is enough with the examples in this site. Of course can have many types with obsolete compression and different kinds of parameters and extrachunks, but for me is not necessary.

About the avi I think is a bit more complex. Do you know abcAVI Tag Editor? http://abcavi.tk/


I decided to support it, since im also thinking about letting ww demux whatever audio is present in the avi (although ww can only use the audio itself if its wav ofcourse)

[Zyphon]

I decided to support it, since im also thinking about letting ww demux whatever audio is present in the avi (although ww can only use the audio itself if its wav ofcourse)

This is good news johnman. I wish you luck implementing this and I hope the coding goes well. :)

[bobcat56458]

Hello johnman: Thanks for the great program, I’m using version 0.53. It works well converting a 5.1 Wave file into a DD5.1 AC3 file using the linked to ac3enc.dll, but when I try to do the same using Soft Encode version 1.0 (Build 19) I get the error message [CB error on: “name of 5.1 wave file.wav”. The configuration of conversion batcher requires 6 monostreams(s).] Now I’m able to load this same 5.1 Wave file directly into Soft Encode and it is able to convert it to a DD5.1 AC3 file. I could convert this 5.1 Wave to 6 mono files using BeSweet, but it would be much better to do all of the job in Wavewizard & its Batch Converter. Is what I’m doing possible using your program, and using Sonic Foundry Softencode as the Output type?

[johnman]

Hello johnman: Thanks for the great program, I’m using version 0.53. It works well converting a 5.1 Wave file into a DD5.1 AC3 file using the linked to ac3enc.dll, but when I try to do the same using Soft Encode version 1.0 (Build 19) I get the error message [CB error on: “name of 5.1 wave file.wav”. The configuration of conversion batcher requires 6 monostreams(s).] Now I’m able to load this same 5.1 Wave file directly into Soft Encode and it is able to convert it to a DD5.1 AC3 file. I could convert this 5.1 Wave to 6 mono files using BeSweet, but it would be much better to do all of the job in Wavewizard & its Batch Converter. Is what I’m doing possible using your program, and using Sonic Foundry Softencode as the Output type?

Im sorry the error message is not clear enough.

You have to go to options and check stream manipulation + monostreams. Then you get monostreams. If you want to use sonic, you must always convert the files to monostreams (unless it has only 1 channel already) and the same goes for surcode. This should fix it, but if it does not, post a msg again.

[daphy]

Hi Johnman,

I played a little with ww these days. The possibiltlies are exciting! :)
I also played with the interface to surcode. One tiny bug always accured see screenshot:

http://img291.imageshack.us/img291/2719/wwbug2wr.jpg (http://imageshack.us)

evertime I pressed convert I run into this error, after closing the ConversationBatcher manually and again pressing the convert button everything went fine :rolleyes:
As you mentioned in your last posting - I would suggest to add a routine which sets (if necessary for the encoder) the output automatically to mono files.
Next thing which makes me wonder: I did a DTS-WAV but the file ended up with a DTS extention - file itself was correct, only the the extension has to be renamed.
I guess that is because of the source file has the same name 123.WAV (PCM 6 channel) => 123.DTS (DTSWAV) should be a WAV means two times the same file name in one folder - thats not possible!
What do you thing about a ww prefix or something like: 123(DTSWAV-16-44.1).WAV?
Another question - I played around with some CUE sheet tools these days - is it possible to add some functions of them in ww, too? This is a general question, I guess we could collect some ideas what CUE function would be nice to add here on the board later :)

that´s it folks ;)

[johnman]

Hi Johnman,

I played a little with ww these days. The possibiltlies are exciting! :)
I also played with the interface to surcode. One tiny bug always accured see screenshot:


evertime I pressed convert I run into this error, after closing the ConversationBatcher manually and again pressing the convert button everything went fine :rolleyes:


You dont have to close it, pressing convert again also works. The problem is that ww starts cb and then waits 100 msec to see if it really started. If it didnt, it gives the error. I also noticed 100msec is to short. When windows has to read the file from the disk it often takes more then 100 msec. After starting it once, windows caches it (i think) and then 100msec does work. Ill fix this for the next release.



As you mentioned in your last posting - I would suggest to add a routine which sets (if necessary for the encoder) the output automatically to mono files.


I thought about this, but should ww also set the sample rate correct when its wrong, and the sampletype? I dont think it would be a good idea to do to much automaticaly, since the setting for cb might be wrong itself, and not the preferences.



Next thing which makes me wonder: I did a DTS-WAV but the file ended up with a DTS extention - file itself was correct, only the the extension has to be renamed.
I guess that is because of the source file has the same name 123.WAV (PCM 6 channel) => 123.DTS (DTSWAV) should be a WAV means two times the same file name in one folder - thats not possible!
What do you thing about a ww prefix or something like: 123(DTSWAV-16-44.1).WAV?

I never thought about this :). Cb just adds .dts since it is dts (or dts-wav).
ill change it for the next release. And its not a problem when the file already axcists. It just adds a (1) to the end of the file, just like ww. If everything is correect, ww and cb never overwrites any file.


Another question - I played around with some CUE sheet tools these days - is it possible to add some functions of them in ww, too? This is a general question, I guess we could collect some ideas what CUE function would be nice to add here on the board later :)
that´s it folks ;)

Well i havent ever used it myself, but from what i know about it, its pretty simple. Ill look into this.

Thx for all the suggestions and bug report's daphy. I'll fix them asap.

EDIT: I fixed this stuff, but since i dont regard the mentioned bugs as critical, i wont release a nre ww it immediately.

[tebasuna51]

This post is for Jonhman and all users of Command Line Operation of Sonic Foundry Soft Encode (used for WaveWizard).

This mode uses a Parameters.ini to set all options of the encoder (defined in an Appendix on the Soft Encode help).
All parameters with numerical value runs ok, but all parameters On/Off works on the contrary. For instance:

-UseLFE=Off -> insert the LFE channel in the ac3 (like I say in previous post).
-Deemphasis=Off -> use the filter and high frequencies are attenuated.
-DCfilter=On -> don't use the filter DC.
...
-TimeStamp=Off -> use the frame type SMPTE (16 bytes before the normal header). Played ok, but rejected from GraphEdit and DVD Authoring software. (default in WaveWizard).
-NonIntel=On -> make a file with Intel byte order rejected always (frame type Intel). If present TimeStamp=Off, frame type SMPTE override frame Intel.

The only parameter don't tested is BandWidthFilter (I don't know how), all others are verified.

I use v1.0 (build 19) of Sonic Foundry Soft Encode.

And i can confirm the bug in sonic surcode. I have to set the option to off to enable it :confused: I think i will add an option to cb to invert the flag.

I think the option to invert the flags must be general not only for UseLFE. (Soft Encode, not Surcode).

A final note for Jonhman: when I set -20 in Dialog Normalization (Configuration of Soft Encode) you put DlgNorm=19 in ini file and must be 20.

[johnman]

This post is for Jonhman and all users of Command Line Operation of Sonic Foundry Soft Encode (used for WaveWizard).

This mode uses a Parameters.ini to set all options of the encoder (defined in an Appendix on the Soft Encode help).
All parameters with numerical value runs ok, but all parameters On/Off works on the contrary. For instance:

-UseLFE=Off -> insert the LFE channel in the ac3 (like I say in previous post).
-Deemphasis=Off -> use the filter and high frequencies are attenuated.
-DCfilter=On -> don't use the filter DC.
...
-TimeStamp=Off -> use the frame type SMPTE (16 bytes before the normal header). Played ok, but rejected from GraphEdit and DVD Authoring software. (default in WaveWizard).
-NonIntel=On -> make a file with Intel byte order rejected always (frame type Intel). If present TimeStamp=Off, frame type SMPTE override frame Intel.

The only parameter don't tested is BandWidthFilter (I don't know how), all others are verified.

I use v1.0 (build 19) of Sonic Foundry Soft Encode.

I think the option to invert the flags must be general not only for UseLFE. (Soft Encode, not Surcode).

A final note for Jonhman: when I set -20 in Dialog Normalization (Configuration of Soft Encode) you put DlgNorm=19 in ini file and must be 20.

Thx for the bugreport. Ill release a new version later today, which will fix these problems.

EDIT: te Dialog Normalization is correct if i interpret the manual correctly :

DlgNorm 0 to 31 (-1 dB to -32 dB)

But taken into account all the ON's that should be OFF's and reverse, im not to sure about this either :(.

[daphy]

Hey guys,

what do you think is a useful function for CUE support?

some ideas
- for added (drag´n´dropped) single files a CUE sheet should be created (options -> gap between tracks, Albumname, Interpret, ... )
- for added (drag´n´dropped) playlists (m3u f.e.) a CUE sheet should be created including all infos from the playlist + the possible options mentioned above ...
(meanwhile I found some simular functions in the latest foobar2000 0.9b6+ version)

do you think these functions will be senseful?

or are there other/better ideas? :rolleyes:

[johnman]

Hey guys,

what do you think is a useful function for CUE support?

some ideas
- for added (drag´n´dropped) single files a CUE sheet should be created (options -> gap between tracks, Albumname, Interpret, ... )
- for added (drag´n´dropped) playlists (m3u f.e.) a CUE sheet should be created including all infos from the playlist + the possible options mentioned above ...

do you think these functions will be senseful?

or are there other/better ideas? ::


I sended you a pm about this but you havent answered it :(

I think its a good idea BTW :)

If someone else has any ideas about this, dont hesitate to share it with the rest of us

[daphy]

sorry my fault - I was a little stressed because of business :(

(but maybe collecting ideas on the board won´t be a bad idea, don´t you think? :) ) <- obsolete you´ve edited your posting :D

I played around with CUE Splitter (see needfulthings (http://www.needfulthings.webhop.org)), it offers a nice GUI for most of it´s functions - I don´t know wether that much function are needed in ww, but some won´t be bad.
As I mentioned at PM there are some further information possible to add inside MP3 (f.e. ALBW) or WAV (Audition) - as basis for a CUE file if possible? :rolleyes:

[johnman]

As I mentioned at PM there are some further information possible to add inside MP3 (f.e. ALBW) or WAV (Audition) - as basis for a CUE file if possible? :rolleyes:

Im still in doubt about the functionality of cue files, so you can imagine i dont know anything about the "album format". But from wht you described, its a handy feature, so i might add it someday.

Currently i want to implement cue functionality and vst hosting. Yesterday i spended a couple of hours trying to get a GUI from VI.dll, but the only thing i figured out is that the steinberg (= "inventor" of vst) documentation is really bad.

I "fixed" the bugs from tebasuna51. The problem is that the settings are not really reliable. For now i have added a setting to invert uselfe,ephasis,DC filter and timestamp and i disabled the intel byte order, so sonic will use the default byteorder (is this correct tebasuna51?).

But i still think the supports is a little shaky since i dont got much confidence sonic is using the .ini file correct. I have to look into this, and i possibly will
a) remove support for sonic
b) use something more reliable then the .ini file

For now i did what tebasuna51 suggested

I will upload ww v0.54b to needfull things in a minute

EDIT
i also tried to get bitidentical output from sonic in normal mode, and sonic through cb, and somehow i never get the same output. I tried to change different settings, but i never got identical files. This means i cant compare the settings from sonic+ cb with sonic in normal mode which again means, the settings are not 100% reliable.

[tebasuna51]

The Dialog Normalization is correct if i interpret the manual correctly :
DlgNorm 0 to 31 (-1 dB to -32 dB)

From a52 doc:
"Valid values are 1–31. The value of 0 is reserved. The values of 1 to 31 are interpreted as -1 dB to -31 dB with respect to digital 100 percent. If the reserved value of 0 is received, the decoder shall use –31 dB. The value of dialnorm shall affect the sound reproduction level."

There are other errors in Soft Encode help:
CodeMode 1 to 8 (1/0 to 3/2)
may be:
CodeMode 0 to 7 (Ch1 + Ch2, 1/0 to 3/2)
From a52 doc:
"If acmod is 0, then two completely independent program channels (dual mono) are encoded into the bit stream, and are referenced as Ch1, Ch2."

And:
MixLevel 0 to 32 must be 0 to 31 (5 bits field, max 31)

The settings not captured in your Configuration window, taken the default values (not the actual settings in Soft Encode):
BSMode=0 (Complete Main)
CopyRight=1 (set)
Original=1 (set)
RoomType=2 (Small Room, flat monitor)
MixLevel=105 (dB SPL)


But taken into account all the ON's that should be OFF's and reverse, im not to sure about this either.
I'm sure, after many, many test, but you are free to believe me.
May be others users can test this, and post any confirmation.

Another test using 0/1 instead Off/On don't work, the parameter are ignored and use the default.

[daphy]

uploaded :cool:
John, can plz check the programlist, I think it must be outdated?!? ;)

[johnman]

From a52 doc:
"Valid values are 1–31. The value of 0 is reserved. The values of 1 to 31 are interpreted as -1 dB to -31 dB with respect to digital 100 percent. If the reserved value of 0 is received, the decoder shall use –31 dB. The value of dialnorm shall affect the sound reproduction level."

There are other errors in Soft Encode help:
CodeMode 1 to 8 (1/0 to 3/2)
may be:
CodeMode 0 to 7 (Ch1 + Ch2, 1/0 to 3/2)
From a52 doc:
"If acmod is 0, then two completely independent program channels (dual mono) are encoded into the bit stream, and are referenced as Ch1, Ch2."

And:
MixLevel 0 to 32 must be 0 to 31 (5 bits field, max 31)

The settings not captured in your Configuration window, taken the default values (not the actual settings in Soft Encode):
BSMode=0 (Complete Main)
CopyRight=1 (set)
Original=1 (set)
RoomType=2 (Small Room, flat monitor)
MixLevel=105 (dB SPL)


I'm sure, after many, many test, but you are free to believe me.
May be others users can test this, and post any confirmation.

Another test using 0/1 instead Off/On don't work, the parameter are ignored and use the default.

I believe you, i dont really need confirmation. The problem is that i used the manual of sonic as a reference (i thought that someone who can make an encoder surely can make a proper manual), and as you mentioned, it has a couple off errors. I know i dont support all settings (like copyright) but i thought i supported the most important ones.
Anyway, im thinking about dont using that .ini file anymore and use something else. I cant seem to verify all the settings reliable , and seeing there are already enough issues with it, im hoping i can use an other method to transport the settings into sonic.



uploaded
John, can plz check the programlist, I think it must be outdated?!?


If you mean that thing we talked about earlier (version checking), it slipped my mind :(.

[tebasuna51]

Use the left column for SonicAc3.ini and right column for Sonic in normal mode. The files generated are bitidentical (for me).

[Parameters] [Soft Encode Settings]
DataRate=448 Data Rate = 448
SampleRate=0 Sample Rate = 48 KHz
CodeMode=7 Audio Coding Mode = 3/2 (L,C,R,l,r)
UseLFE=Off LFE enable = Checked
Bit stream mode = ...Complete Main
DlgNorm=20 Dialog Normalization = -20 dB
NonIntel=Off Save fr. in Intel byte order = Unchecked
CMix=0 Center Mix level = -3 dB
SurMix=0 Surround Mix level = -3 dB
DolbySurround=0 Dolby Surround mode = Not indicated
Copyright bit = Checked
Original bit stream = Checked
Info exists = Checked
Mix levels = 105 dB SPL
Room type = Small room, flat monitor
Deemphasis=On Digital deemphasis = Unchecked
DCFilter=Off DC high-pass Filter = Checked
BandWidthFilter=Off BandWidth low-pass Filter = Checked
LFEFilter=Off LFE low-pass Filter = Checked
Phase90=Off 90 Deegre Phase shift = Checked
Cut3dB=On 3 dB attenuation = Unchecked
CompressionType=0 Compression characteristic = None
RFCmpInUse=On RF overmodulation protection = Unchecked
TimeStamp=On Add Time Stamp = Unchecked

Sonic uses the defaults for not supplied parameters, then you need always:
Bit stream mode = ...Complete Main
Copyright bit = Checked
Original bit stream = Checked
Info exists = Checked
Mix levels = 105 dB SPL
Room type = Small room, flat monitor

[johnman]

Ahhh thanks tebasuna51, i see i made a mistake with a couple of defaults. I unchecked them while they should be checked. I have to say your feedback is very helpfull, and detailed. It really helps me to fix the bugs faster :). :thanks:


Im was/am working on a written tentamination (some really heavy stuff), and doing wavewizard devlopment together . Since that doesnt go to well together (it gave me a little headache :eek: ), im now doing only the tentamination. I've got 5 days for it, so until it is finished, i dont have to much time to spare.

When i get some free time again, im going to triple check all settings for surcode to make sure they are ALL correct.

Although everyone still can send bug reports, im afraid i cant fix them immediately.

EDIT
Has anyone checked if the settings in sonic in normal usage (without cb) are correctly applied in the resulting file?
If i would compare the output from sonic + cb with the output from sonic, and the output from sonic itself is incorrect, its still useless.

[tebasuna51]

Testing Wavewizard 0.54 for Soft Encode

[Normal] [Inverting]
UseLFE=On UseLFE=Off Ok
DlgNorm=19 DlgNorm=19 I set 20
RFCmpInUse=Off RFCmpInUse=Off ?
Deemphasis=Off Deemphasis=On Ok
DCFilter=On DCFilter=Off Ok
BandWidthFilter=On BandWidthFilter=On ?
LFEFilter=On LFEFilter=On ?
Phase90=On Phase90=On ?
Cut3dB=Off Cut3dB=Off ?
TimeStamp=Off TimeStamp=On Ok

You believe in me only 40%.
But, always, thanks for your job.

Edit: Before send this post I read your new post, well, you believe in me a little more. But I want the 100%.
You say:
Has anyone checked if the settings in sonic in normal usage (without cb) are correctly applied in the resulting file?
If i would compare the output from sonic + cb with the output from sonic, and the output from sonic itself is incorrect, its still useless.
And me:
"The only parameter don't tested is BandWidthFilter (I don't know how), all others are verified."
This include the effects in file output.
Let me more time for this last test (BandWidthFilter) and post later.

[johnman]

i believe you 100%. If you say its ok... it IS ok.

Im sorry, but i didnt read you post carefully enough. I missed theat part about you verified the other settings. Like i said, got much on my mind, and my tentamination is going kinda slow...

What i would like to know is
a) the bandwith filter
b) you said in an earlier post something about that intel byte order. I didnt fully understand it, and i thought its better to just disable it and use the default setting. But if you can clarify it, i would very much appreciate it :).
c) the other settings, are they correct now, or dont you know if they are correct> You got all those '?' next to them.

And thx for the effort in trying to get sonic work reliable

[tebasuna51]

a)Test for BandWidthFilter
Using a test wav 6 chan 48000 Samples/sec. (24 seconds tone test from 0 Hz to 24 KHz)

Encoded at 448 kbps (Nominal BandWidth = 20.3 KHz)
BandWidthFilter = On -> Progressive cut from 19.4 to 19.8 KHz
BandWidthFilter = Off -> Irregular cut from 20.3 to 20.4 KHz

Encoded at 320 kbps (Nominal BandWidth = 15.8 KHz)
BandWidthFilter = On -> Progressive cut from 15.0 to 15.5 KHz
BandWidthFilter = Off -> Irregular cut from 15.8 to 15.9 KHz

Then the switch work, another question is recommend it.
Before the test I think not see so clearly the differences.

b) Just like other switch you must invert On -> Off
This switch is ignored when you check TimeStamp. In Soft Encode Help:
"Time stamped frames are automatically not stored in Intel byte order."

Soft Encode uses 3 frame types:
-Non-Intel, First byte 0x0B -> run Ok in all soft. Default or with:
Add Time Stamp = Unchecked
Save frame in Intel byte order = Unchecked

-SMPTE, First byte 0x01, byte 17 0x0B (rest of frame like Non-Intel) -> Are played ok (ignoring the first 16 bytes), rejected from GraphEdit and DVD Authoring software. Activated with:
Add Time Stamp = Checked

-Intel, First byte 0x77 -> Only know the proper Soft Encode to open it. Activated with:
Save frame in Intel byte order = Checked
Add Time Stamp = Unchecked

I don't know the use of Time Stamp or Intel byte order.

c) Switches with ? aren't inverted by the switch in cb Configure, then don't work properly.
All switches must be inverted. And the Dialog Norm must be the same value than selected, now is value - 1.

EDIT: (ignoring the first 16 bit) must be (ignoring the first 16 bytes),

[johnman]

Thx tebasuna, this is some info i can use immediately in ww. The next release will have correct support for sonic softencode thanks to you :D.

[johnman]

Im trying to let ww also decode ac3 and dts so anyone can transcode them to any other format in only a single step. Since the code of tranzcode is freely available, im thinking about changing it a little so it can work together with ww, but since this is not the case with azid, im thinking about using the azid.exe from the azid1.9 package. The problem is that i havent used it very much, and i remember from a long time ago the azid.dll is sometimes better then the azid.exe (or was it the other way arround?). Does anybody know if there really is a difference between azid.dll vs azid.exe? Or should i maybe use an other ac3 decoder? Any comment/info is welcome :)

[jm duchenne]

Hi Johnman,

First, thank you for your work. It is really a must have tool for everyone who does multichannel sound.

I've read in the previous pages that you are working on a VST host feature ???

I make multichannel plugins (http://acousmodules.free.fr) and I wonder how it could be wonderful to use them inside Wavewizard !

Have you some beta version to test ?

[johnman]

Hi Johnman,

First, thank you for your work. It is really a must have tool for everyone who does multichannel sound.

I've read in the previous pages that you are working on a VST host feature ???

I make multichannel plugins (http://acousmodules.free.fr) and I wonder how it could be wonderful to use them inside Wavewizard !

Have you some beta version to test ?

I havent got a beta with vst hosting unfortunately, but it would be indeed a wonderfull idea to use your stuff it in ww. I looked at your site and its really interesting.
Have you some indepth knowledge about making vst's ? Im having a little problem with the implementation (of the host), and i could use some tips/pointers sometimes.

(@ all Although im also busy with my study (currently i have to make a sort of a simple graphic card driver) ww is still being developed, but a little slower then before.)

[jm duchenne]

Have you some indepth knowledge about making vst's ?
I'm sorry, but no. :(

I use Synthedit or other graphical environments, so I have not to deal with the basic codes...

But perhaps could you see with Herman Seib : http://www.hermannseib.com/english/vsthost.htm
or Tobias Fleischer (Toby Bear) : http://www.tobybear.de/developers.html
They have free hosts and I'm sure that they can share some resources or tips.

Perhaps could you send me an email (sonart@free.fr) when you will have made some progress with the VST host implementation ?
Thanks !

[johnman]

I'm sorry, but no. :(

I use Synthedit or other graphical environments, so I have not to deal with the basic codes...

But perhaps could you see with Herman Seib : http://www.hermannseib.com/english/vsthost.htm
or Tobias Fleischer (Toby Bear) : http://www.tobybear.de/developers.html
They have free hosts and I'm sure that they can share some resources or tips.

Perhaps could you send me an email (sonart@free.fr) when you will have made some progress with the VST host implementation ?
Thanks !

I already tried to get some info on vst on a number of ways, but so far wasnt very succesfull. I just have to find out everything myself :(.

And when i got a working beta host ill send you a mail. Im developing the host and ww seperate for now, so the first beta will probabely be only a hosting program. Later ill merge them together.

[johnman]

Im experimenting a little with recoring from the soundcard and i noticed that somehow all regular samplerates can be used to record from the soundcard. Even my onboard soundcard can record at 192khz :confused: . Does this mean my soundcard can really support this samplerate natively, or is windows resampling. How can i find out what the real hardware suported samplerates are? (BTW cooledit also shows my onboard soundcard is capable of recording @ 192khz)

And here is a little screenshot of ww recording in simplemode if anyone is interested:

screenie (http://www.rarewares.org/wavewiz/recording.png)

[daphy]

And here is a little screenshot of ww recording in simplemode if anyone is interested
Does the enhanced mode support multichannel recordings?
-> There meanwhile lots of soundcards on the market which support multichannel input, so let´s feed them :D :D :D

[johnman]

Does the enhanced mode support multichannel recordings?
-> There meanwhile lots of soundcards on the market which support multichannel input, so let´s feed them :D :D :D

It was already on my mind, but i think i need asio stuff for this. Normal windows recording only supports up to 2 channels (i think). I dont know exactly how hard/easy asio is to support, but if it is as poorly documented as the VST-interface (both are from steinberg if im not mistaken) it will take while :(.

My soundcard has an option for 6 channel recording (and i already captured 6 channel audio from powerdvd (digitally) using krystal audio not to long ago) and i intend to support this in ww :). Maybe i eventually create some sort of "virtual soundcard", so the output from ANY program can be recorded digitally using ww.

But how and when the support comes, im not sure. Im trying to gather some good info on recording, but this isnt easy. The more support i get, the faster the development will go :D .

[daphy]

you already know this (http://www.asio4all.com/) one?

[johnman]

Unfortunately i still can not release a new version, but to the ppl that are interested in ww: Im pretty busy with other stuff, but i still i work on ww from time to time. Currently I've finished another rewrite on ww. I've totally changed the internal structure of ww. Now i just create a "sample-pipeline" @ runtime by simply connecting together several functions. What this means is that its possible to do i.e.:

upsample -> dither -> downsample -> vst -> downsample -> normalize -> amplify->channelmap->vst-> etc (not really usefull, but still possible)

Previously the "sample-pipeline" was very static, and to support unlimited vst's was kind of difficult, so thats why the rewrite.

Currently ive not made a GUI to configure it, so configuring is currently hardcoded in. But if i would make a GUI, ww will get similair functionality as bidule :).

This internal stuff will only be visible in the next release by the number of vst's which can be linked together (virtual unlimited), but the change is really much bigger then that. Any function or vst can be seperately configure and used as often as needed.

Thats all for now :).....

[ursamtl]

john this sounds really promising. Take your time and I'm sure ww will be great once you get this all together.

[johnman]

Perhaps could you send me an email (sonart@free.fr) when you will have made some progress with the VST host implementation ?
Thanks !

I sended you an email, but i still havent got any responce. Could you please check your mailbox to see if they arent blocked?

[johnman]

Here is a screenshot of ww hosting a vst. NB notice the option to save as many configurations as you want for every vst. In the VI thread an idea was mentioned to add saving options to VI. When ww is used as a host, the settings can already be saved (and exchanged if needed).

screenshot (http://www.rarewares.org/wavewiz/vsthost.png)

[daphy]

Hi Johnman,

looks promising :D I´m looking forward to the next release!

[ursamtl]

Hi Johnman,

looks promising :D I´m looking forward to the next release!

Me too! John, tell me, would it be possible to add realtime file playback to wavewizard or some sort of preview function? It would be useful for making adjustments. For example, I have a 5.1 soundcard/speaker setup, so when I'm using V.I in Bidule, I'm able to immediately hear the results of any slider adjustments I make. Without monitoring capability, it's very much a "hit and miss" or trial and error process. Ideally, one could preview what the output file would sound like, and then switch to something equivalent to Bidule's "offline" mode to do the actualy processing as fast as the CPU allows.

Regardless of the monitoring issue, this latest screen grab looks very good!

Regards,
Steve.

[johnman]

Multichannel playback (and recording) will be supported :). This is indeed a must have for ww if vst's are used. But i wont promise its already available in the next release .

johny

[ursamtl]

Great news! No need for promises. You're already doing some great work!

[daphy]

Hiho Johnman,

I tried something simple with conversation batcher simply: want to convert an existant WAV to MP3 with out any changes on the source file. Hmm -> I have to let wavewizard (at least) copy the file to another place, is it possible to feed cb directly or did I missed something http://smilies.neo101.nl/smilies/2160.gif (http://www.maxthon.com/)

[johnman]

sry daphy, this cant be done...

when you feed wavs to ww and ww accepts it you know you will get a vallid mp3. Its a bit more reliable, but also more time consuming...

but i agree its not pretty, but what i actually want to do is to integrate cb in ww and just select output format in ww. Dont know how, but that would be a big improvent i think

[daphy]

-obsolete-

[johnman]

ty daphy :)

[bcn_246]

Do you think it would be possible to implement a function where you could set exactly the output time, by adding silence at the end/cropping the end audio?

I allways find this very useful when encoding to DVD to make sure all the streams are exactly the same lengh. I guess it would be best accurate to the millisecond.

Great tool by the way, keep it up.

Ben :)

[johnman]

I believe this should be possible to see upto the milisecond the legth of a file, but im not at home so i take a look at it this afternoon ...

[johnman]

It is possible to set the exact length of a file...
if you set the DELAY option on an change the filename you can change the length of the file.

[raquete]

If you have a question about the program, ask it in this thread so others might benefit from it.

is possible to encode one group of files? something like batch file?

thanks for this great encoder!

[raquete]

...one month later...

maybe the question was answered before but i can't find ...please,send me in the right direction or answer my doubt because i'm using Wavewizard following the "Steve's guide" and i'm encoding lots of AC-3 (one by one).

thanks.

[tebasuna51]

is possible to encode one group of files? something like batch file?
Is not possible with WaveWizard
...please,send me in the right direction or answer my doubt because i'm using Wavewizard following the "Steve's guide" and i'm encoding lots of AC-3 (one by one).
I don't know the "Steve's guide", but if you explain how use WaveWizard to encode ac3 (what kind of wav input, ac3enc.dll or SoftEncode) maybe there are another solution.

[raquete]

is not possible? (oh no)

here is the "Steve's guide" (Ursamtl):
http://forum.doom9.org/showthread.php?t=105684

read "Part 2: Merging or splitting files" that use Wavewizard.

:thanks: tebasuna51.

[tebasuna51]

here is the "Steve's guide" (Ursamtl):
http://forum.doom9.org/showthread.php?t=105684

read "Part 2: Merging or splitting files" that use Wavewizard.
I know this thread, just I post:
http://forum.doom9.org/showthread.php?p=792462#post792462
to modify the "Part 2: Merging or splitting files" because with the guide (first post) you obtain an ac3 with wrong channelmapping.
EDIT: Now the guide is correct. Sorry.

Really is difficult automatize this process (3 stereo input wav 32 bit float to encode to ac3 5.1), maybe with AviSynth and Bepipe ...

[ursamtl]

Actually that's no longer accurate. As I mentioned in my March 5 reply to you in the thread, I edited that section of the guide to correct the problem.

[tebasuna51]

Actually that's no longer accurate. As I mentioned in my March 5 reply to you in the thread, I edited that section of the guide to correct the problem.
Sorry ursamtl, is my fault:(
Your first post is ok now.
Bad memory and bad re-read :stupid:

[slothsquid]

Hi Johnman

Thanks for the wavewizard. It helps me a lot in transcoding wavs to DTS. I have a couple of questions

1. Do you any plan to integrate AF's tranzcode with ww?
2 Is there any way I can use WW in CLI?
3. It is more helpful, if you include the function to delete intermediate files in batch mode. As 6Ch wavs create huge files, harddisk filled so quickly.

Thanks
SS

[raquete]

slothsquid, (first post after register in Apr 2003...)
good "wish list".

@ tebasuna51
Really is difficult automatize this process (3 stereo input wav 32 bit float to encode to ac3 5.1), maybe with AviSynth and Bepipe ...
can you (maybe) do that? (please. :p )
now we have Aften to encode AC3-5.1 using the 6 channels(merged) from WW.

[tebasuna51]

can you (maybe) do that?
Now with Aften only need one step in WaveWizard (remmaping not needed).

There are also a double conversion not recommendable:
float -> Int 16 (AviSynth) -> float (Aften).

But if you have many conversions to do and/or don't want create the intermediate wav6 files ...

To automatize using AviSynth you need create .avs files like:
a = BassAudioSource("X:\Xpath\x_fLfR.wav")
fl = GetChannel(a, 1)
fr = GetChannel(a, 2)
b = BassAudioSource("X:\Xpath\x_CLFE.wav")
c = GetChannel(b, 1)
lfe = GetChannel(b, 2)
d = BassAudioSource("X:\Xpath\x_slsr.wav")
sl = GetChannel(d, 1)
sr = GetChannel(d, 2)
MergeChannels(fl, fr, c, lfe, sl, sr)

And after execute batch files like:
bepipe --script "Import(^X:\Xpath\x1.avs^)" | aften -b 448000 - x1.ac3
bepipe --script "Import(^X:\Xpath\x2.avs^)" | aften -b 448000 - x2.ac3
...

The problem is only write paths and names of the files. To help this job I write 3wav1ac3.vbs
' This file (3wav1ac3.vbs) must be
' in the same folder than bepipe.exe and aften.exe
'
' You can Drag & Drop the *fLfR.wav file/s over this file or
' over a ShortCut created on Desktop.

' Autor: tebasuna
'
Dim wsh, args, fso, fe, fe2, fe3, fg, fa
title="3wav1ac3 Script"
' get input files
Set wsh = WScript.CreateObject("WScript.Shell")
Set args = WScript.Arguments
If args.Count = 0 Then
MsgBox "Special Aften ac3 encoder script." & vbCR & _
"Use drag & drop to specify the *fLfR.wav sources.", vbInformation, title
WScript.Quit
End If
' check path
Set fso = CreateObject("Scripting.FileSystemObject")
' Search aften.exe at same folder than 3wav1ac3.vbs
aften = WScript.ScriptFullName
camino = left(aften, len(aften) - len(WScript.ScriptName))
aften = camino & "aften.exe"
If Not fso.FileExists(aften) Then
MsgBox "Don't found aften!" & vbCR & "(must be at '" & aften & "')", vbCritical, title
WScript.Quit
End If
' Search bepipe.exe at same folder than 3wav1ac3.vbs
bepipe = camino & "bepipe.exe"
If Not fso.FileExists(bepipe) Then
MsgBox "Don't found bepipe!" & vbCR & "(must be at '" & bepipe & "')", vbCritical, title
WScript.Quit
End If
' Options to encode. Here we can put defaults or capture other parameters.
msg = "Bitrate to encode."
msg = msg & chr(13) & "Accepted values:"
msg = msg & chr(13) & "320, 384, 448, 512, 576 or 640 Kb/s"
p = InputBox(msg, "Bitrate", "448")
opts = " -b " & p & "000 - "
'
aften = Chr(34) & aften & CHR(34) & Chr(32)
bepipe = Chr(34) & bepipe & CHR(34)
bat = 0 ' bat not open
For i = 0 To args.Count-1
fe = args(i)
' check input file
If fso.FileExists(fe) Then
pre = Left(fe, Len(fe) - 8)
suf = LCase(Right(fe, 8))
If suf = "flfr.wav" then
fe2 = pre & "CLFE.wav"
fe3 = pre & "slsr.wav"
If fso.FileExists(fe2) Then
If fso.FileExists(fe3) Then
' write .avs file
Set fa = fso.CreateTextFile(pre & ".avs", True)
fa.WriteLine "a = BassAudioSource(" & Chr(34) & fe & Chr(34) & ")"
fa.WriteLine "fl = GetChannel(a, 1)"
fa.WriteLine "fr = GetChannel(a, 2)"
fa.WriteLine "b = BassAudioSource(" & Chr(34) & fe2 & Chr(34) & ")"
fa.WriteLine "c = GetChannel(b, 1)"
fa.WriteLine "lfe = GetChannel(b, 2)"
fa.WriteLine "d = BassAudioSource(" & Chr(34) & fe3 & Chr(34) & ")"
fa.WriteLine "sl = GetChannel(d, 1)"
fa.WriteLine "sr = GetChannel(d, 2)"
fa.WriteLine "MergeChannels(fl, fr, c, lfe, sl, sr)"
fa.Close
' write lines in batch file
If bat = 0 then
bepipe = bepipe & " --script " & Chr(34) & "Import(^"
Set fg = fso.CreateTextFile(pre & ".bat", True)
End if
bat = bat + 1
bepip2 = bepipe & pre & ".avs^)" & Chr(34) & " | "
bepip2 = bepip2 & aften & opts & Chr(34) & pre & ".ac3" & Chr(34)
fg.WriteLine bepip2
bepip2 = "del " & Chr(34) & pre & ".avs" & Chr(34)
fg.WriteLine bepip2
End If
End If
End If
End If
Next
if bat > 0 then
fg.Close
MsgBox "Finish. Run the batch file.", vbInformation, title
else
MsgBox "Don't found sources.", vbInformation, title
End If
WScript.Quit
Edit: Added line:
bepipe = Chr(34) & bepipe & CHR(34)

[raquete]

tebasuna51 (cool helper)
:helpful: :goodpost:

thank you.
;)

[raquete]

tebasuna51,
running the batch file (i add "pause" in the end) give this error:


C:\Arquivos de programas\AftenGUI-1.1>bepipe --script "Import(^e:\Xpath\script.avs^)" | aften -b 448000 - x1.ac3

Aften: A/52 audio encoder
(c) 2006 Justin Ruggles

***************************************
BePipe by dimzon
***************************************
Script used:
# BEGIN
Import("e:\Xpath\script.avs")
# END


Scanning for Audio Stream...
ERROR: Can't find audio stream!
invalid wav file: -b

C:\Arquivos de programas\AftenGUI-1.1>pause
Pressione qualquer tecla para continuar. . ."

as the sources,paths,scripts are ok,the reason of the error is because my .waves are 48k-32bit?
:confused:

[tebasuna51]

To open in AviSynth 2.56 32 bit waves we need BassAudioSource, if you install the last BeHappy
shon3i package (http://forum.doom9.org/showthread.php?p=829754#post829754) you have BassAudioSource ready.

Sorry, I forget 1 line in 3wav1ac3.vbs (to work with paths with spaces):

'
aften = Chr(34) & aften & CHR(34) & Chr(32)
bepipe = Chr(34) & bepipe & CHR(34)
...
Edited in precedent post.

Now my log:
D:\Internet\AudioTest\usrmtl>"C:\Arquivos de programas\AftenGUI-1.1\bepipe.exe" --script "Import(^D:\Internet\AudioTest\usrmtl\Bueno_.avs^)" | "C:\Arquivos de programas\AftenGUI-1.1\aften.exe" -b 448000 - "D:\Internet\AudioTest\usrmtl\Bueno_.ac3"

Aften: A/52 audio encoder
(c) 2006 Justin Ruggles

***************************************
BePipe by dimzon
***************************************
Script used:
# BEGIN
Import("D:\Internet\AudioTest\usrmtl\Bueno_.avs")
# END


Scanning for Audio Stream...
Found Audio Stream
Channels=6, BitsPerSample=16, SampleRate=48000Hz (1)
Writing Header...
Writing Data...
Signed 16-bit 48000 Hz 6-channel
Done! 0 | bw: 60.0 | bitrate: 448.0 kbps


D:\Internet\AudioTest\usrmtl>pause
Presione una tecla para continuar . . .
(1) Original wav are 32 bit float, after BassAudio-AviSynth 16 bit Int.

In your log:
"invalid wav file: -b"

Maybe you need the last aften 0.3 or _dev from Kurtnoise (the parameters order are inverted).

[raquete]

oh boy....oh boy!!! :cool:

i had the aften 0.3 _dev in the AftenGUI folder,was only needed the "BassAudio.extension".
my "default bitrate" to encode AC3(music) is 512.
i edit the "batch" and the "3wav1ac3.vbs" to this bitrate.

look:
"C:\Arquivos de programas\AftenGUI-1.1>bepipe.exe --script "Import(^e:\Xpath\scri
pt.avs^)" | aften.exe -b 512000 - "e:\Xpath\x1.ac3"

Aften: A/52 audio encoder
(c) 2006 Justin Ruggles

***************************************
BePipe by dimzon
***************************************
Script used:
# BEGIN
Import("e:\Xpath\script.avs")
# END


Scanning for Audio Stream...
Found Audio Stream
Channels=6, BitsPerSample=16, SampleRate=48000Hz
Writing Header...
Signed 16-bit 48000 Hz 6-channel
Writing Data...
Done! 7 | bw: 27.3 | bitrate: 512.0 kbps


C:\Arquivos de programas\AftenGUI-1.1>pause
Pressione qualquer tecla para continuar. . ."


was not needed "drag and drop" over "3wav1ac3.vbs",
running only the "batch" give the result working perfectly.

thank you so much (again) tebasuna51
:cool: :cool:

[UVLaser]

Hi Johnman

Thanks for the wavewizard. It helps me a lot in transcoding wavs to DTS. I have a couple of questions

1. Do you any plan to integrate AF's tranzcode with ww?
2 Is there any way I can use WW in CLI?
3. It is more helpful, if you include the function to delete intermediate files in batch mode. As 6Ch wavs create huge files, harddisk filled so quickly.

Thanks
SS

:helpful: Good list. I am interested in #2 & 3. very much. I am having same problem. The batcher is not really useful me as I have small harddisk. I hope Johnman is still reading the posts.

:thanks:
UVLaser

[ReinerSchweinlin]

Hm, the download-link in post 1 is broken. Sorry for asking - I am lazy, but where can I download this tool?

[raquete]

don't be lazy. :p
here in this same page have a link to the guide where you'll find WW:
http://forum.doom9.org/showpost.php?p=841747&postcount=107

[ReinerSchweinlin]

Thanx :)

[raquete]

anyone can help me to extract 6 mono channels-16bit from 5.1 dtswav with WW? ( tebasuna51,are you around? )
with tranzcodegui is easy but only for 32 bits waves.
i read the johnman "general info" http://forum.doom9.org/showpost.php?p=662842&postcount=5
but i can't put it working.(maps,etc..i'm lost)

thanks so much.

[Austin Forgotten]

anyone can help me to extract 6 mono channels-16bit from 5.1 dtswav with WW? ( tebasuna51,are you around? )
with tranzcodegui is easy but only for 32 bits waves.


Hi there, since I might not complete Tranzcode anytime soon since I've been working on a gui version which hopefully will be able to display DD/DTS decoded channel waveforms, where you would (again hopefully) be able to select section bounded by frames and cut/paste/trim/split visually and hope to add the playback as well. What was my point, oh ya since I ceased work on the console ver (called beta 4) in March (which still lacks the 'Two-Pass', and stereo down mixing), as the next step would be to create a dll from it. Anyway I just posted the beta 4 ver (March), on my little website. You can access the file here, use this link:

http://tranzcode.byethost22.com/files/Tranzcode_v0.40(beta).rar

I still need to document some added commands. In short, it will (last I tested way back then), decode .ac3 and ddwav as well as .dts, .cpt and dtswav, and you can specify the output bit-depth (8,16,24,32 bit integer, and 32 & 64 bit floating pt), as well as resample to either 44100, 48000 and 96000 Hz. It will also do batch file processing (use the '*' character as the file delimiter (also use double quotes for filenames containing spaces-of course), and also a means to list channel files if you want to mux a multichannel wav. It can also strip channels from multichannel wavs. and you can specify the speaker settings since it will create in wave extensible format. I haven't done this for awhile muxing, err.. actually used/tested it since March, so bare with me. Oh ya, it will produce an accurate info on the DD/DTS files using hmm.. I forgot, think it's the /i (file info calculated)i, and /im (file info measured) commands if I recall correctly and also produce an info file .tzf (I think :) ) which I would later use for editing using the gui. So I suppose I'll have to write a short guide on some of the new commands.

This may still work with TranzGUI but may only work as before which will be quite limited as to its capabilities.

Copy 'Tranzcode.exe into the folder where the file you want to decode is in (or copy it into your i.e. 'C:\WINDOWS\system32' dir)

So to decode to 16 bit, open a 'DOS' window (Start/Accessories/Command Prompt), Change dir 'cd' to the dir and issue the following command (example)


tranzcode "my file.dts" /16

or if you want to resample the output as well..

tranzcode "my file.dts" /16 /44 [edited: thats WRONG! :) it sould be /44100 or /48000 or /96000

corrected: tranzcode "my file.dts" /16 /44100

Good luck, hope not to many bugz :P


Ok I finally ran some testing (err guess I should have done that before posting it). No matter there were some log messages which you can ignore for now. I'll repack a new compile to get rid of any troubleshooting msgs and clean up any garbage I left in it, and add a list of some new commands with parameters.

Also I'll add those commands here and totally hijack this thread completely :P (sorry John).

BTW I would also like to mention that I thank Johnmann for giving me the inspiration, along with some pointers on how to incorporate SSRC (Shibatch Tools) into Tranzcode for resampling/bit depth conversion, dithering etc. Thanks guy!

[EDIT]

Ok it didn't take long to find bugs right away (I found 2 already) - which weren't there on preveous testing, so I messed something up.

bug 1) When Upsampling & Bit Depth Converting together, it fails (at least on DTSWav and DDWav)

bug 2) When decoding DDWav the progress display stays at 1% but actually decoded fine.

Argh.. it may take awhile since I haven't look at the code so long ago. Sorry folks never expected this... Really! Already tarnished my bad reputation :(

Ok it should be ok for now, uploaded a slightly older working version.

Seems the link is not working, so just go to the website and download it from there:

http://tranzcode.byethost22.com

Cheers

-AF

[tebasuna51]

anyone can help me to extract 6 mono channels-16bit from 5.1 dtswav with WW?
WaveWizard can't decode dts, then a dtswav can't be used like input for WaveWizard.

You can use WaveWizard to convert the TranzGUI output to 16 bit int, or DTSParser to convert dtswav to dts an after decode to 6 mono wav 16 int with BeHappy.

Sorry I don't know dtswav -> 6 mono wav 16 bit int in only 1 step.

Edit: I don't see the Austin Forgotten post. We can try with Tranzcode v0.40

[johnman]

hello ppl, guess who is back :)
as you might have seen development of ww sort of halted last couple of months but i just wanted to drop by to see what has happened. Who knows what the future brings :). iirc the last thing i was working on was vst chaining and multichannel recording but to be honest i am not really sure :D.

[raquete]

@ johnman,welcome back.

@ tebasuna51 & Austin Forgotten

:thanks: so much!!

[johnman]

I have taken a look at the code and i was busy trying to chain vst's together. I wonder, is anyone still using the program ? even needfull things seems unreachable so i wonder if it would be worth the effort to atleast finish the vst chaining + multichannel recording?

It has been a while and it takes me a lot of effort to understand my 1000's lines of code :sly:

And to be honest, the program is to complex to really grasp. I should have focussed more on functionality because now its a sort mess with the mapping and the vst chaning and merging of files.

[raquete]

I wonder, is anyone still using the program ?
i can be sure!

[johnman]

thats nice but if thats all, i dont think further development would be usefull :rolleyes:

[ursamtl]

thats nice but if thats all, i dont think further development would be usefull :rolleyes:

Oh I think a lot of people are using ww. I've seen several comments to that effect. As for needfulthings, you should ask daphy about that. Last I heard, he was quite busy with work but I'm sure he hasn't completely given up. As for me, I intend to get back into surround once the autumn sets it. Lately I've been exploring, of all things, analog! I heard some transfers of vinyl LPs done from quality turntables and I must say they sounded better than anything I've ever heard on commercial CDs. So right now I'm exploring the possibility of purchasing a new turntable! Given the improved soundstage, etc., I'm actually excited about what kind of source these could be for 5.1!

[tebasuna51]

I wonder, is anyone still using the program ?.
I make a search
Key Word(s): WaveWizard
Posts Made By: tebasuna51
and I found 30 threads where I recommend the use.

[Skelsgard]

I make a search
Key Word(s): WaveWizard
Posts Made By: tebasuna51
and I found 30 threads where I recommend the use.

Same here, there´s pretty much no audio-related posts of mine where I don´t recommend (or actually order) to use WW.
Ww is my butt-saver mandatory wonder audio tool for anything surround related.
Keep the great work.. (please!!! :D )

[johnman]

It is somewhat of a strange and unique program and it is capable of doing somethings most other programs can't. But its also not very user friendly, but when you get it to do what you want its kind of cool :).

I've got a couple of free days now so i'll have a look if i can jumpstart my sourcecode and get it to work again.

now let me get a mirror up, and daphy.. if you are back again send me a msg ;).


!tebunasa, i know you have noticed a couple of bugs long long time ago, ill come to you soon about it and i hope you remember some of the bugs you found. Im afraid i dont :(. I even have a bit of a problem to use ww :P.

edit 2. my computer broke so i only have a onboard soundcard on a very old computer that produces a lot of hum out of my speakers but it will have to do for now i guess :P.


New Download link (http://www.rarewares.org/wavewiz/wavewizardv0.54b.zip)

[Skelsgard]

But its also not very user friendly
Really? The first time I used it I actually thought is was incredibly simple and right-to-the-point (no unnecesary blowed-up BS), and I´m like every other user.

[johnman]

Really? The first time I used it I actually thought is was incredibly simple and right-to-the-point (no unnecesary blowed-up BS), and I´m like every other user.

To REALLY understand ALL functionality it takes a while. i.e. if you wanna merge 3 stereo file to a single 6 ch file and encode the output in 1 step to dts... its possible but i bet most ppl need a few tries to get it to work :P.

But i agree that it is to the point :). But i would like it to be less functional and more usable. More functionallity in't always better imo.

[tebasuna51]

tebasuna, i know you have noticed a couple of bugs long long time ago, ill come to you soon about it and i hope you remember some of the bugs you found.
I only remember a problem in ursamtl upmix method (http://forum.doom9.org/showthread.php?p=792462#post792462):
"Seems that Stream manipulation and Channelmapping don't work together, the merged file in pass 5 is always in correct wav order with Channel mapping enabled or disabled."

But is solved in two steps for ac3enc.dll, and in one step without mapping thanks to the new encoder Aften with remapping included.

Maybe for a new release you need include the new ac3/aac encoders Aften, Nero, CT (winamp).
And don't forget a simple remapping option Dolby Prologic II downmix (http://forum.doom9.org/showthread.php?p=864319#post864319):
6->2#mix#inactive#Dolby Prologic II downmix
0.325 0.000 0.000 0.325 0.230 0.230 0.000 0.000 -0.282 0.163 -0.163 0.282

[johnman]

I have taken a look at the project and i really should have made more documentation. I have a little problem figuring it all out again and since i was busy with several things at once some functionallity doesn't work anymore so maybe i have to go back to the latest working release which is a real bummer since i came a long way with recording sound from the soundcard and vst's functionality and some other stuff :(.

[daphy]

Hiho,

Johnman, nice to see you back on this 'stage' :D
:)

[johnman]

Thank you daphy, nice to be back :)
I went deep into the code and i am almost back on track from where i left of some time ago.

[Robot]

I’m hoping someone can help me with this problem.

I have a 6-channel PCM wav that I’m trying to turn into an ac3 file. The problem I’m having is with the channel mapping. I cannot figure out how to map them correctly for ac3. No matter what I do, I seem to either get the center channel switched with a rear channel or the left switched with the center, etc. It’s driving me nuts!

I’m using Wavewizard to re-map the channels from the wav file. The wav file was originally encoded from a 6-channel AAC file if that helps.

Thanks

[Skelsgard]

Your output WAV PCM must be in this order for correct channel mapping: L-R-C-LFE-Ls-Rs.
What are you using for AAC --> WAV decoding?
If the tool doesn´t have a correct maping on decoding, it is gonna get you a while till you figure out what order is being used, as AAC is C-L-R-Ls-Rs-LFE (the LFE order might be wrong though, I don´t remember).
The AC3 encoder (i.e Aften) assumes that the order is right, so you get this:
C-L-R-Ls-Rs-LFE (a wrongly decoded WAV from AAC)
L-R-C-LFE-Ls-Rs (a correct WAV mapping)

So C goes to L, L to R, R to LFE, and so on and so forth.

Check your AAC decoding.
As last resource, strip into mono streams the 6-ch WAV (with Ww :D ), determine wich mono channel is wich, and remux with WAV order.

Cheers

[Robot]

This is the process I’ve been using.

1. I take the 6-channel AAC file (ripped from a QT file)

2. I put the AAC file into the Nero encoding tool and output it to a PCM WAV file.

3. I take the WAV and put it into Wavewizard (where I get stuck with the channel mapping).

4. I then take the newly created WAV and encode it to AC3 via Aften.

Thanks.

[Skelsgard]

Does the first WAV decoded thru Nero plays with correct mapping?

[Robot]

Does the first WAV decoded thru Nero plays with correct mapping?

Unfortunately, I don’t know. I don’t have a 5.1 system on my computer. I can only tell if the mapping is right if I burn to DVD. :(

[raquete]

if you have one wave(interleaved-6 channels),open in nero wave editor only to check the "map".

[Ebobtron]

does this help?
AAC center
front left channel
front right channel
rear left channel
rear right channel
LFE
Wav
front left
front right
front center
LFE
rear left
rear right

oops never mind

[Skelsgard]

A fast way I use to see the mapping is to play the WAV thru AC3filter, wich has an equalizer that shows you the levels of each audio channel. This way avoids having to strip the channels and check each channel individually.
Usually the channel with little to no level througought the track is the LFE, the one that is almost always high specially on dialogs is the Center. Left and Right have normally the same levels at the same time, and peaks in the surround channels are not as frequent as the Frontal Left and Right, but there are still more than in the LFE. Also, Left comes always before Right. If u have channel 0 and channel 1, it´s bound to be that #0 is L and #1 is Right (the same with surround channels).
This might sound complicated but you only need to see it once and you´ll understand inmediately.

Cheers.

[raquete]

This might sound complicated ..is complicated and the levels of channels don't "obey" rules...less LFE of course,all depend of the source and his separations between channels..opening one short interleaved 6-channels in nero wave editor is easy to "see" the waveforms and to hear the channels(all together of separated)...will be faster and secure but...taste is taste. :-)

[Skelsgard]

Opening with AC3filter implies no need to trim, export, wait till the file is loaded into the audio app and peak-searched.
You just open it. It doesn´t get any faster, easier or less energy consuming than that.

Cheers.

[raquete]

right Skelsgard,seems easy to recognize when you know the source but for unknow...well,i like and use AC3filter but it only show me levels and not waveforms then again,for unknow sources is hard only whatching levels...i lose something in your hints? trust,i really want to know how you use in this case.thanks! :)

[Skelsgard]

Remember that there´s nothing absolute here. There are always opinions, wich can be discussed and shared, and hopefully never silenced.
Is not "you´re wrong and I´m right", just "I like more this way because... than this other way".
And about the "rules", they´re just meant to be general guidelines, nothing more.

About what i mean.
It´s something like this, I´ll give some examples:

The first image is a standard AC3 being decoded with AC3filter.
The channels in the levels panel seem to correspond to a portion of the movie with dialog (high level C), maybe some background music? (medium levels L and R, with almost equal loudness), birds ramdomly singing on the trees? (low level Ls and Rs), and nothing blowing up (null LFE).

http://img477.imageshack.us/img477/7707/ac3stdec5.jpg

Now the next pics belong to an apparently wrongly-mapped-when-decoded WAV.

http://img388.imageshack.us/img388/5664/wavwrong01af4.jpg
http://img214.imageshack.us/img214/9360/wavwrong02gp9.jpg

Now we have a L channel with high level maintained in samples from different time (the upper and bottom images are 10 minutes apart from each other). SL and LFE have the same levels across the samples, might correspond to L and R. And so on.

Knowing this is a wrongly mapped decoded AAC, i can assume that the order is
C-L-R-Ls-Rs-LFE, instead of
L-R-C-LFE-Ls-Rs.
So I find the correlations: a high level L in dialogs wich would correspond to C, R and C sharing levels with similar overall levels wich would relate to L and R respectively. The same for LFE and Ls, relating to Ls and Rs respectively. And an almost null, low level for moments, Rs, relating to LFE.

This entire mental construction, as long as it seems, occurs in half a second.
You just click-open the WAV, check ac3filter´s properties and there you see it.

I feel this is parting away from the main issue of the thread wich is Wavewizard, maybe it can be continued in a new thread.

Cheers.

[raquete]

all clear,very clear.
this is the "bingo" part: The channels in the levels panel seem to correspond to a portion of the movie with dialog (high level C),selecting one part with dialogue will be very easy...and fast of course.
thanks Skelsgard,your explanations are very complete and don't need any audio editor to check,AC3filter can do it easily following your hints.
:)

[Robot]

OK, I'm lost now. :)

[raquete]

compilation...(resuming)
I’m hoping someone can help me with this problem.

I have a 6-channel PCM wav that I’m trying to turn into an ac3 file. The problem I’m having is with the channel mapping. I cannot figure out how to map them correctly for ac3.
Your output WAV PCM must be in this order for correct channel mapping: L-R-C-LFE-Ls-Rs.
A fast way I use to see the mapping is to play the WAV thru AC3filter, wich has an equalizer that shows you the levels of each audio channel. This way avoids having to strip the channels and check each channel individually.
About what i mean.
It´s something like this, I´ll give some examples:

The first image is a standard AC3 being decoded with AC3filter.
The channels in the levels panel seem to correspond to a portion of the movie with dialog (high level C), maybe some background music? (medium levels L and R, with almost equal loudness), birds ramdomly singing on the trees? (low level Ls and Rs), and nothing blowing up (null LFE).

http://img477.imageshack.us/img477/7707/ac3stdec5.jpg

Now the next pics belong to an apparently wrongly-mapped-when-decoded WAV.

http://img388.imageshack.us/img388/5664/wavwrong01af4.jpg
http://img214.imageshack.us/img214/9360/wavwrong02gp9.jpg

Now we have a L channel with high level maintained in samples from different time (the upper and bottom images are 10 minutes apart from each other). SL and LFE have the same levels across the samples, might correspond to L and R. And so on.

Knowing this is a wrongly mapped decoded AAC, i can assume that the order is
C-L-R-Ls-Rs-LFE, instead of
L-R-C-LFE-Ls-Rs.
So I find the correlations: a high level L in dialogs wich would correspond to C, R and C sharing levels with similar overall levels wich would relate to L and R respectively. The same for LFE and Ls, relating to Ls and Rs respectively. And an almost null, low level for moments, Rs, relating to LFE.

This entire mental construction, as long as it seems, occurs in half a second.
You just click-open the WAV, check ac3filter´s properties and there you see it.

I feel this is parting away from the main issue of the thread wich is Wavewizard, maybe it can be continued in a new thread.

Cheers.

now is very easy...
Skelsgard's posts are like a guide with pictures.
regards
:)

[DAKnn]

where download ssrcdll.dll ?? for starting wavewizard.exe
..
Wavewizard uses ssrc_hp to resample. Look at the forum for the latest release ..
Im not found ssrcdll.dll and ssrc_hp for download..

[tebasuna51]

where download ssrcdll.dll ?? for starting wavewizard.exe
..
Wavewizard uses ssrc_hp to resample. Look at the forum for the latest release ..
Im not found ssrcdll.dll and ssrc_hp for download..
ssrcdll.dll is with Wavewizardv0.54b (http://www.rarewares.org/wavewiz/wavewizardv0.54b.zip) package.

I don't need any file ssrc_hp to resample with wavewizardv0.54.

[Robot]

How do you use AC3filter to check the mapping? I see no way to open a file with it.

What am I missing? :confused:

[Skelsgard]

Open the AC3 in a Directshow player and check for the filters used to render the file. There it should be AC3Filter.
Or if you use MPC, be sure to have AC3filter doing the decoding instead of MPC´s built-in filters.
Don´t know if VLC accepts AC3filter as decoding filter.

Cheers.

[Robot]

OK, I used MPC and AC3filter to play the 6-Channel wav. This is what I got:

http://bsrfilms.com/media/ac3.jpg

I'm not sure I understand what I'm looking at. How do I re-map this using WW?

Thanks again for all your help!

[Skelsgard]

From the pic you posted, SR and LFE seem to be switched.
If you decode that stream to 6 mono WAV, you get a SR probably containing LFE and viceversa.
Load the 6 mono streams in this order to get a proper WAV:
L-R-C-SR-SL-LFE
and play it again.
Post two pics of the results if possible, a few minutes apart from each other.

Cheers.

[Robot]

Sorry, but how do I get the 6 mono waves? And how will I know which is which?

Thanks for your patience.

[Skelsgard]

With Wavewizard, of course :)
As you already have the 6ch file mixed, you can remap straight from WW.
Load the 6ch WAV, go to Edit --> Preferences. Enable Enable Channelmapping, Ignore File size in header --> Always (I don´t know how big your file is) and Output --> Wave PCM. Disable anything else.
Hit OK.
Go to Edit --> Channel mapping.
Click on New to create a custom profile. Choose From 6 to 6, Move channels and configure the mapping.
Your stream seems to be: L(ch0)-R(ch1)-C(ch2)-SR(ch3)-SL(ch4)-LFE(ch5).
To remap it to proper WAV order:
L-R-C-LFE-SL-SR --> ch0(0)-ch1(1)-ch2(2)-ch3(5)-ch4(4)-ch5(3)

It´s also posible that the order is:
L(ch0)-R(ch1)-C(ch2)-SL(ch3)-SR(ch4)-LFE(ch5).
If so remap to:
ch0(0)-ch1(1)-ch2(2)-ch3(5)-ch4(3)-ch5(4)

Once you´ve created the mapping profile, enable it and hit OK.
Now Convert.

Try to find similar sounds that should come from near-to-each-other speakers. Let´s say, bullet sounds coming out from the FL speaker are more likely to come out from the SL speaker too. If you hear these sounds coming from the SR, it´s almost certain that SL and SR are mixed.

Cheers.

[Robot]

If you hear these sounds coming from the SR, it´s almost certain that SL and SR are mixed.


I think I've almost got the mapping right, but I still don't have two surround channels (only left or right depending on how I map). Does this mean that the SR and SL are mixed? If so, how do I fix that?

Thanks.

[Skelsgard]

If the mappings are correctly applied, that shouldn´t happen.
Maybe you missed something in the mapping.

For ch0(0)-ch1(1)-ch2(2)-ch3(5)-ch4(4)-ch5(3) [L-R-C-SR-SL-LFE], remap like this:
http://img242.imageshack.us/img242/4999/capture2av9.jpg


For ch0(0)-ch1(1)-ch2(2)-ch3(5)-ch4(3)-ch5(4) [L-R-C-SL-SR-LFE], like this:
http://img212.imageshack.us/img212/1324/capture1mb6.jpg

Don´t forget to enable the mapping
http://img292.imageshack.us/img292/9591/capture3dx5.jpg

Cheers

[erdie]

anyone got a mirror to the .zip file?

seems like the rarewares one is dead...

[daphy]

look here (http://forum.doom9.org/showthread.php?p=662691#post662691) ;)

[erdie]

thx a bunch!

[homerpez]

I’m hoping someone can help me with this problem.

I have a 6-channel PCM wav that I’m trying to turn into an ac3 file. The problem I’m having is with the channel mapping. I cannot figure out how to map them correctly for ac3. No matter what I do, I seem to either get the center channel switched with a rear channel or the left switched with the center, etc. It’s driving me nuts!

I’m using Wavewizard to re-map the channels from the wav file. The wav file was originally encoded from a 6-channel AAC file if that helps.

Thanks

Similar thing here... though I'm not even making it that far.

I'm going from DTS-HD --> Tranzcode --> Wavewizard (6 CH WAV), hopefully to AC3 using Behappy.

I take the individual channel WAV's, and order them in Wavewizard: FL, FR, C, LFE, SL, SR (which I think is the proper way). I click "Merge" under Preferences, and process it. The end result is, I can't open the 6-channel wave in anything... not in Behappy, Audacity, or Nero Wave editor. It crashes each of these programs every time I load the merged WAV.

The individual WAV's do play, as they are generated by Tranzcode, it seems I must be missing something somewhere, or Wavewizard is. I'm guessing from people singing the praises of Wavewizard, it's something I'm overlooking.

Incidentally... I also tried to import the WAV's into audacity, and then merge into a flac... same result. Everything crashes instantly on load of the merged file. ???

Can someone tell me exactly how to properly merge the 6 WAV's into one, so I can then make a proper AC3 using BeHappy? Hopefully without the insta-crash?

(Of course, I have no way of testing the mapping myself, since the resulting 6-channel wave is crashing every encoder I have.......)

I'm desperately confused here... I really need to have "that one way that works"... thanks :)

[tebasuna51]

@homerpez
I don't know if Tranzcode can decode DTS-HD, but if you say the mono wavs are playable, and recognized by WaveWizard, the multichannel generated can be usable for Aften (or AftenGUI) to encode to ac3.

Maybe you have a multichannel wav greater than 4 GB and some soft refuse to play them, but no problem for Aften if you include the -readtoeof 1 parameter.

If this not work, please specify sizes and formats of wav files.

[homerpez]

I definitely think something else is amiss here... since as I posted above, it wasn't only the 6.99GB Wav that crashed all encoders/players instantly, the FLAC did it as well. The FLAC was only about 300 MB or so, so it's not a matter of size I don't think.

Each of the tracks I get from tranzcode have this attribute when I ask for 'Properties' in MPC:

Audio: IEEE Float 48000Hz mono 1536Kbps [output]

All of them do play correctly by themselves, but will not play of encode when merged into anything. I've tried WAV (Wavwizard) and FLAC (using Audacity).

In BeHappy (which is frontending Aften I guess), the WAV simply crashes the program instantly. With the FLAC, it just sits there, the processor is doing "something" but no audio is encoded from it.

Is there a program that will let me load all the mono WAV files into it directly, and encode straight to AC3?

Perhaps someone who IS doing DTS-HD encoding can lead me through their process, I may be doing something wrong when I'm merging them, or my computer itself may have a demon possession - you never know. :p

Whatever ways I'm trying (Wavwizard and Audacity) don't seem to merge properly.....

[tebasuna51]

If your mono wav's are over 1h 48min, and over 1.16 GB each, and your 6.99 GB multichannel wav are created with WaveWizard Preferences:
- Stream manipulation: Merge
- Output Format: Wave PCM or WaveFormatEx
- Others options unchecked

your wav 6.99GB is ready to be encoded with Aften with -readtoeof 1

Use command line or AftenGUI (http://kurtnoise.free.fr/index.php?dir=Aften/)

[homerpez]

OK... this is seriously making me want to kill myself.

If your mono wav's are over 1h 48min, and over 1.16 GB each, and your 6.99 GB multichannel wav are created with WaveWizard Preferences:
- Stream manipulation: Merge
- Output Format: Wave PCM or WaveFormatEx
- Others options unchecked

your wav 6.99GB is ready to be encoded with Aften with -readtoeof 1

Use command line or AftenGUI

Re-cap:

1. I demux the DTS-HD track from the m2ts stream.

2. I used Tranzcode 0.40 (TransGUI) to generate 6 mono WAV files. These mono WAV's all play correctly in MPC, WMP, etc.

3. I load the 6 WAV files into Wavwizard. I order them, top to bottom:

FL
FR
C
LFE
SL
SR

4. I applied the checkboxes as indicated. Stream manipulation, Merge, PCM WAV out, no other options. I Click 'Convert', and at the end they all say 'Sucessful'. It makes the 6.99GB wav.

5. I then try to use the 6.99GB "audio-FL_Merge.wav" file to make the AC3.

Here's what happens:

BeHappy wavsource -> Aften AC3 = Program instantly crashed, with this error:

Problem signature:
Problem Event Name: BEX
Application Name: BeHappy.exe
Application Version: 0.1.9.35190
Application Timestamp: 4601dccd
Fault Module Name: AvisynthWrapper.dll
Fault Module Version: 0.0.0.0
Fault Module Timestamp: 45f4c9c9
Exception Offset: 00001d62
Exception Code: c000000d
Exception Data: 00000000
OS Version: 6.0.6000.2.0.0.256.1
Locale ID: 1033
Additional Information 1: f5a8
Additional Information 2: 26e2a974e105888ffa55cf6ab26d94e6
Additional Information 3: e052
Additional Information 4: c349466e7c52437c92b09f29bff1e833


Aften GUI -> AC3 = Program comes up immediately with the window 'Encoding Finished', even though no AC3 was produced at all.

("Read to end" flag was checked on both GUI's, and is by default.)

Trying to play audio-FL_Merge.wav in WMP crashes. Winamp crashes. Audacity crashes. Media Player Classic crashes.

This is the error raised by MPC, maybe it can explain why this is going on???

C:\temp\TEST\audio-FL_Merge.wav::Output

Media Type 0:
--------------------------
Unknown

AM_MEDIA_TYPE:
majortype: MEDIATYPE_Stream {E436EB83-524F-11CE-9F53-0020AF0BA770}
subtype: MEDIASUBTYPE_WAVE {E436EB8B-524F-11CE-9F53-0020AF0BA770}
formattype: TIME_FORMAT_NONE {00000000-0000-0000-0000-000000000000}
bFixedSizeSamples: 1
bTemporalCompression: 0
lSampleSize: 1
cbFormat: 0


Media Type 1:
--------------------------
Unknown

AM_MEDIA_TYPE:
majortype: MEDIATYPE_Stream {E436EB83-524F-11CE-9F53-0020AF0BA770}
subtype: TIME_FORMAT_NONE {00000000-0000-0000-0000-000000000000}
formattype: TIME_FORMAT_NONE {00000000-0000-0000-0000-000000000000}
bFixedSizeSamples: 1
bTemporalCompression: 0
lSampleSize: 1
cbFormat: 0


Thanks for your collective patience, this is driving me batty. :confused:

[tebasuna51]

- Maybe you use a old Aften version without support for -readtoeof?
Last rev521 must work.

- Or send me the wav headers (mono and 6chan). Extract the first bytes with a Hex editor or VirtualDub/Mod:
VirtualDub -> Tools -> Hex Editor -> File -> Open... your wav file, Edit -> Extract Segment... -> (Address: 0, Length: 1000) Ok and save like segment.wav.
Use http://www.mytempdir.com/ to host this file.

[homerpez]

- Maybe you use a old Aften version without support for -readtoeof?
Last rev521 must work.

- Or send me the wav headers (mono and 6chan). Extract the first bytes with a Hex editor or VirtualDub/Mod:
VirtualDub -> Tools -> Hex Editor -> File -> Open... your wav file, Edit -> Extract Segment... -> (Address: 0, Length: 1000) Ok and save like segment.wav.
Use http://www.mytempdir.com/ to host this file.

I will try and do that, and I'll edit this post once I've figured out the hex thing.

Just wanted to clarify again... it's not just Aften crashing, it's every program I use, system-wide. Any time my system goes to play or encode from this multichannel wav, whatever program I'm in crashes. So it's having trouble even before Aften even has a chance to work with it.

EDIT: oops... I messed up the first time, this is the first I ever used Mytempdir.com..

This should be the segment file : http://www.mytempdir.com/1349567

The segment also crashes everything I try to load it in....

[tebasuna51]

The segment also crashes everything I try to load it in....

1) I can't reproduce this, using -readtoeof 1 I can encode the segment to a 32 ms (1 frame) ac3.

2) The header is correct (32 bit float, 48 KHz, 6 channel) only the DataLength and RiffLength are set to 0 (the correct value can't fit in the 4 bytes reserved for these fields).

Maybe some soft don't crash, only open/play a wav with 0 sec.

If you want try this:

- Extract the files in WavHead.zip (http://www.mytempdir.com/1349701) at same folder than your audio-FL_Merge.wav and double-click in runme.bat

- Now the wav file can be open like 20.8 sec with some soft and we avoid the problem with value 0.

3) To open this wav (32 float) with WavSource (AviSynth-BeHappy) you need AviSynth 2.5.7

You can open the six monowavs with BeHappy using this merge.avs file:
fl = WavSource("g:\yourpath\audio-FL")
fr = WavSource("g:\yourpath\audio-FR")
fc = WavSource("g:\yourpath\audio-FC")
lf = WavSource("g:\yourpath\audio-LF")
sl = WavSource("g:\yourpath\audio-SL")
sr = WavSource("g:\yourpath\audio-SR")
mergechannels(fl,fr,fc,lf,sl,sr)

[homerpez]

Thanks for the reply... unfortunately I think there's something seriously wrong with my machine, or the codecs on it. I hope this is a common problem so someone can tell me what to do..

First of all, I really appreciate your efforts to help me, tebasuna51... it's above and beyond. :thanks:

However, I tried the .avs script on my 6 WAV files... instant crash.

I opened up the example WAV you sent me... instant crash.

I tried to load it into BeHappy using Wavsource... instant crash.

Apparently, at the system/driver/codec level, my machine seems to want to refuse to load multichannel WAV files at all. Where do I even begin to try and fix this kind of problem? I always assumed that multichannel WAV was a standard for eons...??? I've never heard of such a thing happening, and it's something that's worked for me in the past.

I'm going to play around a bit with ffdshow, see if maybe I have something checked off somewhere I shouldn't (I don't think I have anything special going on that would cause this)... all I know at this point is, my machine won't decode multichannel WAV, or FLAC formats, out of the blue, at all. :confused::confused::confused:

[homerpez]

I FIGURED IT OUT. :D

It turns out, there WAS a checkbox I should have used in Wavewizard after all: Conversion from 32-bit (float) to 16 bit.

On much closer inspection, I discovered that the mono WAV's were all 32-bit (float) format. When I played back a couple AC3 files I had around, I noticed they were all 16-bit. Hmm.

Sure enough, I downsample the 6 wav files to 16-bit, and the multichannel WAV opens and plays!

I'm encoding to AC3 with BeHappy as I type. :D

Hopefully BeHappy knows what to do with the channels mapped that way... if not, I'll be back to ask for help again probably. :p

Thanks very much for the help thus far

[tebasuna51]

When I played back a couple AC3 files I had around, I noticed they were all 16-bit. Hmm.
Not exactly, the decoder convert the ac3 to the desired bitdepth, the ac3 don't have bitdepth because the samples are in frequency (not time) domain. The ac3 dynamic range is considered equivalent to 24 bits.
Sure enough, I downsample the 6 wav files to 16-bit, and the multichannel WAV opens and plays!
Now the wav is half size 3.5 GB and work without problems.
Hopefully BeHappy knows what to do with the channels mapped that way... if not, I'll be back to ask for help again probably.
Don't worry the channel mapping is correct.

[ACrowley]

@homerpez

Why dont you use a prof Dolby encoder like Sony/Steinberg/Sonic or Softencode

So you can encode the 6 mono waves to AC3

1. Demux DTS HD
2. Tranzcode DTS HD (core) to 6 mono waves
3. Softencode wave to ac3 (i use Steinberg Nuendo)

Or you can decode the DTS HD core with Tranzcode to a 6.1 wave with /mch switch and feed aften
So you need no wavewizard in this cases

greets

[Wilbert]

@johnman,

Could you fix the problem in this post: http://forum.doom9.org/showthread.php?p=942314#post942314?

"Wavewizard don't put a valid channel mask, then with WinHex at Offset 0x28 put 0x3F (ChannelMask= FL,FR,C,LFE,SL,SR) and now work OK"

Looking at the readme, i think this is open source. Can i get the source somewhere?

[johnman]

Hello @all who still are using wavewizard...
I got some private messages which i will address in a minute so if you send me some... u'll get an answer soon...

A long time ago, i started developing because i wanted to get multichannel sound from my music. Since the day i bought my sound card which can generate multichannel sound from a stereo source this need was pretty much fulfilled. Since then i continued only for all you guys working with audio conversion :).

I have developed more free software totally not related to audio conversion and this also has a small but very enthusiastic user base. I do this in my free time and get nothing in return (money wise). Although i like to develop nice tools as freeware, i can't maintain all the software i develop (and have developed in the past). I just continue until i find a reason to use my time on an other project. Although i haven't pulled the plug out of ww, i have no plans to continue it either in the near future.

Before i stopped, i had vst's working and audio recording also. But i wanted to put more and more and more thing in it so i didn't actually come to a release (stupid mistake, i know). I really would like to make a new release with a couple of new toys because i have put a lot of hard work into it since the last official release and got a lot of help with this from a couple of forum members right here.

So what will happen next?
Only time will tell.....

PS
On my other software projects i have searched for other pretty good skilled developers who want to help just for the fun of it, but so far i haven't found anyone. So whenever i want to get something done i do it myself which dramatically slows down anything if i have no spare time at all. Although ww is not officially opensource, i have no fundamental objection to give/share the source with someone. If you think you are up to it to (co-)develop wavewizard give me a msg. But i'm not in it for the money, so if your intentions are other then give the community a nice tool you must have a damn good reason (or a zillion dollars on your bank account :cool: ).

[johnman]

@johnman,

Could you fix the problem in this post: http://forum.doom9.org/showthread.php?p=942314#post942314?

"Wavewizard don't put a valid channel mask, then with WinHex at Offset 0x28 put 0x3F (ChannelMask= FL,FR,C,LFE,SL,SR) and now work OK"

Looking at the readme, i think this is open source. Can i get the source somewhere?

wavewizard leaves the channelmask of waveEX format blank since it has no reliable method of knowing which masks to set.

The sourcecode has never been published but i created everything myself. There is a little issue with the vst part of ww since the interface is owned by a company so im not entirely free to put the sourcecode on a site.

[Zodraz]

WaveWizard is a really great and useful program. It is only one for which I know that could merge two stereo files into one multichannel (4.0) file in one pass. Too pity you stopped its development.

Anyway, I wish you good luck on your another projects.

[sk11vengeance]

Im trying to convert a 6ch wav file to 2ch wav in Wavewizard. When I get a 2ch file from ww, it sounds different compared to a file I get when I use Tranzcode with /mch and dbpoweramp 12 to convert it to 2ch wav. The latter has more bass whereas the one from ww sounds less bassy and a little louder. I can get the same sounding file as ww when I use Tranzcode to extract all 6 files and merge the ones labeled FL and FR from Tranzcode in ww. Is this the proper way to get the 2 channel file from the 6ch DTSWAV? Thanks.

[edo1080]

I Would like to know if it's possible to make the following the resulting multichannel .wav output file sequence : L,C,R,LFE,SL,SR . I think wavewizard flags the .wav output file so that every channel is lebeled as L,R,C, etcc... I made a 5.1 wav file from 6 mono waves and loaded it into scenarist, scenarist expects the following sequence L,C,R,SL,SR,LFE but the file made but wavewizard is L,R,C, etc.. so the sequences don't mach. I can load files in the right sequence accepted by sonic, but sonc tells the wav file has not the right channel sequence. Is it possible to change the output sequence for 5.1 .wav files by channel mapping? And if so how to make this?
Thanks

[tebasuna51]

I Would like to know if it's possible to make the following the resulting multichannel .wav output file sequence : L,C,R,LFE,SL,SR .
Yes, WaveWizard output channel order is controlled by the user.

I think wavewizard flags the .wav output file so that every channel is lebeled as L,R,C, etcc...
Nope, in wav header don't exist this info.
Only if WAVE_FORMAT_EXTENSIBLE is selected there are a ChannelMask field, but WaveWizard put always 0 in this field.
Never use WAVE_FORMAT_EXTENSIBLE header with old soft not compliant with standard MS wav order.

I made a 5.1 wav file from 6 mono waves and loaded it into scenarist, scenarist expects the following sequence L,C,R,SL,SR,LFE but the file made but wavewizard is L,R,C, etc.. so the sequences don't mach. I can load files in the right sequence accepted by sonic, but sonc tells the wav file has not the right channel sequence.
Is not possible know the channel sequence in a wav file, only the creator know the order and the wav header can't say if the C channel is the second or the third.
Is it possible to change the output sequence for 5.1 .wav files by channel mapping? And if so how to make this?
Thanks
Of course, menu Edit -> Channel Mapping -> New and you can define the new order.

[edo1080]

thank you. You told the output order is set by user, but How can i know which channel is CH(0) , which one is ch(1) etc..
I can assing the file I named front_left.wav to source channel 0 and use channel mapping to assing CH(0) of soure to ch(0) of output; but if I do so, can I be sure that it will be used as front _left by the player? If I go on and assing CH(1) to CH(1) of output and load center in ch(1) I can have to following output L,C,R,SL,SR,LFE (assuming I used this assignment) but how can the player know which channel is Center , which one SL etc.. Thanks

[tebasuna51]

thank you. You told the output order is set by user, but How can i know which channel is CH(0) , which one is ch(1) etc..
I can assing the file I named front_left.wav to source channel 0 and use channel mapping to assing CH(0) of soure to ch(0) of output; but if I do so, can I be sure that it will be used as front _left by the player? If I go on and assing CH(1) to CH(1) of output and load center in ch(1) I can have to following output L,C,R,SL,SR,LFE (assuming I used this assignment) but how can the player know which channel is Center , which one SL etc.. Thanks
ch(0) = FL
ch(1) = FR
ch(2) = C
ch(3) = LFE
ch(4) = SL
ch(5) = SR
This is the standard wav order used by any wav player, wav editor and all modern audio software.
The players know the Center channel is the third and so on.

Any other wav order is wrong and only needed like input for old soft.

I don't understand your "assing CH(1) to CH(1)", all is more simple, if you need a un-standard multichannel wav for Sonic (FL-C-FR-SL-SR-LFE):

- If you have six monowavs load in wavewizard screen with the order:
FL.wav
C.wav
FR.wav
SL.wav
SR.wav
LFE.wav

- If you have a correct multichannel wav you need a mappings.ini like:
6->6#move#inactive#ac3 order for SoftEncode or ac3enc
0 2 1 4 5 3
6->6#move#inactive#multichannel wav to aac
2 0 1 4 5 3
6->6#move#inactive#DVD-A to multichannel wav
0 1 4 5 2 3
2->2#move#inactive#Swap stereo channels
1 0
2->1#mix#inactive#Mix stereo to mono
0.500 0.500
6->2#mix#active#Dolby Prologic II downmix
0.325 0.000 0.000 0.325 0.230 0.230 0.000 0.000 0.282 -0.163 0.163 -0.282
2->2#move#inactive#Duplicate Left Cannel
0 0

and use 'ac3 order for SoftEncode or ac3enc' (included also in mappingsNEW.ini (http://forum.doom9.org/showthread.php?p=690486#post690486)in last WaveWizard v0.54b)

[edo1080]

Thank you for your hints; I followed them but Sonic gives me the following error message:

specific channel assignment is not corresponding to data

What could this mean? If I well unterstood the meaning of the channels in every .wav 5.1 file is:

ch(0) = FL
ch(1) = FR
ch(2) = C
ch(3) = LFE
ch(4) = SL
ch(5) = SR

As you wrote above; now you can reassing the channels as you like depending on the encoder you are going to give the file to, but you have to know that every player or program which will accept such a file will unterstand the ch(0) as FL, the ch(1) as FR etc..

Maybe Sonic is expecting a different layout of .wav file, not the standard one, is there any other multichannel .wav profile?

Maybe the solution could be a tool allowing WAVE_FORMAT_EXTENSIBLE for 5.1 .wav files, is there any? but i've read on a MSDN document that Channel order is fixed for WAVE_FORMAT_EXTENSIBLE too, you cannot reorder them, only mask out channels you're not using

[edo1080]

Update: I've discovered the .wav files accepted by Sonic are 0xFFFE 6ch wav, but
I don't know what this could mean and how I could make one file like this

[tebasuna51]

Audioformat 0xFFFE is WAVE_FORMAT_EXTENSIBLE header, if Sonic need this header you can use WaveWizard menu Edit -> Preferences -> Output format -> WaveFormatex.

But maybe now the channel order must be the standard.

[edo1080]

also using extensible wav format is the same I get the same errors, I sincerely can't unterstand which kind of error this is

[vicfred]

is this program still supported?

[tebasuna51]

is this program still supported?

Read this post (http://forum.doom9.org/showthread.php?p=1058600#post1058600)

[Heathcliff]

i might be stupid, but i can't make a 5.1 wav from 6 mono wavs with wavewizard, however i red a lost of posts saying it's possible.
can somebody explain me how to do that ?

many thanks !

[tebasuna51]

- ADD the six monowavs in order FL, FR, FC, LF, BL, BR
- EDIT -> Preferences -> Stream manipulation -> Merge files
- Output format -> Wave PCM or WaveFormatEx (the channel mask is set to 0, must be 63)
- CONVERT

[keysersose]

Does your program support packed PCM files? specifically I want to convert 5.1 24/96 MLP files and get the 6 mono waves..

Thanks

[tebasuna51]

Does your program support packed PCM files? specifically I want to convert 5.1 24/96 MLP files and get the 6 mono waves..

WaveWizard only accept wav file input.

MLP lossless encoded format can be converted to mono wav's with eac3to soft.

[Heathcliff]

- ADD the six monowavs in order FL, FR, FC, LF, BL, BR
- EDIT -> Preferences -> Stream manipulation -> Merge files
- Output format -> Wave PCM or WaveFormatEx (the channel mask is set to 0, must be 63)
- CONVERT

many thanks :)

[napalm-187]

hi

i have demuxxed an true hd track from hd dvd

converted with eac3to to a multichannel wav

now i will make an BD compatible lpcm stream will i ue for sonic bd

when i laod the pcm the mui generator says the channlorder is wrong

wich settings i must use in wavewizzard to make an right wave file with the right channelorders?

thx

[tebasuna51]

Use:

eac3to x.wav x.pcm

not only channel mapping need to be changed, also little->big-endian sample format (not possible with WaveWizard).

[napalm-187]

hi

ok i will test this

thx
edit

ok now i have an pcm

run mui generator works

drag ves file into scenarist, dont work
Error : FilePath : D:\Downloads\eac3\eac3to\batman.pcm.ves,
It failed in reading the file.


mui generator tells this is required

6 ch L C R rs ls lfe

where is the problem?

[tebasuna51]

Sorry, I don't know mui generator or scenarist software.

[napalm-187]

Thank you for your hints; I followed them but Sonic gives me the following error message:

specific channel assignment is not corresponding to data

What could this mean? If I well unterstood the meaning of the channels in every .wav 5.1 file is:

ch(0) = FL
ch(1) = FR
ch(2) = C
ch(3) = LFE
ch(4) = SL
ch(5) = SR

As you wrote above; now you can reassing the channels as you like depending on the encoder you are going to give the file to, but you have to know that every player or program which will accept such a file will unterstand the ch(0) as FL, the ch(1) as FR etc..

Maybe Sonic is expecting a different layout of .wav file, not the standard one, is there any other multichannel .wav profile?

Maybe the solution could be a tool allowing WAVE_FORMAT_EXTENSIBLE for 5.1 .wav files, is there any? but i've read on a MSDN document that Channel order is fixed for WAVE_FORMAT_EXTENSIBLE too, you cannot reorder them, only mask out channels you're not using

hii

@
tebasuna51

i get the same errormessage in sonic scenarist bd author

i testetd all channelmappings and get always the same error

i put in wav
then wave ex
and raw
alsways the same


where is here the prob

[tebasuna51]

The channelmask flag in WAVE_FORMAT_EXTENSIBLE header only say what channels are present from:
FRONT_LEFT 0x1
FRONT_RIGHT 0x2
FRONT_CENTER 0x4
LOW_FREQUENCY 0x8
BACK_LEFT 0x10
BACK_RIGHT 0x20
FRONT_LEFT_OF_CENTER 0x40
FRONT_RIGHT_OF_CENTER 0x80
BACK_CENTER 0x100
SIDE_LEFT 0x200
SIDE_RIGHT 0x400
TOP_CENTER 0x800
TOP_FRONT_LEFT 0x1000
TOP_FRONT_CENTER 0x2000
TOP_FRONT_RIGHT 0x4000
TOP_BACK_LEFT 0x8000
TOP_BACK_CENTER 0x10000
TOP_BACK_RIGHT 0x20000

but the order can't be altered.

BTW, how can detect scenarist the channel mapping in a raw file? (only LFE, without high frequencies, can be recognized)

And scenarist, need wav or lpcm files?

[napalm-187]

hi


scenarist accept .wav and .lpcm and raw in the mui generator



but the raw file is acceptedin mui genarator and crates the files and when you drag intio scenarist data tree comes an error

[plugh]

Dwnloaded and tried out this tool, and I'm wondering if I I'm doing something wrong or if there is a glitch.

Dowloaded and extracted the zip. Went into cb configure, disabled the two Surcode and the Sonic foundary programs. Hit 'Download list' and it successfully downloaded / unzipped / installed everything else. Oh yeah, I also copied the ac3enc dll into the directory and did 'find' on it.

Grabbed a six channel wav, and did conversions to aac, vorbis, and ac3. When I do the vorbis convert, the cb window makes it appear as though I requested a Sonic foundary conversion. :confused:

The jobs log shows (for example)* Message received from wavewizard.exe. Msg ID: 0130.
Msg description: Add job
[Parameters]
DataRate=0
SampleRate=0
CodeMode=0
UseLFE=Off
DlgNorm=0
CompressionType=0
RFCmpInUse=Off
CMix=0
SurMix=0
DolbySurround=0
Deemphasis=Off
DCFilter=Off
BandWidthFilter=Off
LFEFilter=Off
Phase90=Off
Cut3dB=Off
TimeStamp=Off
Did I screw up the install? How do I fix it?
Thanks!

[tebasuna51]

...
Grabbed a six channel wav, and did conversions to aac, vorbis, and ac3. When I do the vorbis convert, the cb window makes it appear as though I requested a Sonic foundary conversion. :confused:
...
Did I screw up the install? How do I fix it?

Sorry, I can't reproduce the problem.

BTW, is not recommended use WaveWizard to encode to aac or ac3, there are better new encoders (NeroAacEnc, CT, Aften) than can be used with Foobar, BeLight or AviSynth methods (BeHappy, SoundOut, Wavi)

[plugh]

This is a one-shot using a 10 second 'channel ID clip' in order to verify channel mapping of an output device, and I needed the channel mapping and wave format controls; the built-in conversions were a nice plus.

I've poked at this some more and it seems to be purely cosmetic - the job log, the command line, and the output files are all correct...

[Wile_E_Coyote]

Sorry I'm a bit late to this party, I got here as soon as I could.
Are you sitting comfortably? Then I’ll begin...

...there is a trend, these days, for people to convert (films in particular) to .mp4 files containing video (H264) with 5.1 AAC audio.
All well and good you say, and it is, if all I wanted to do was play them on my PC with VLC or burn them to DVD. But what if I wanted (and I do) to watch them on my TV through my surround sound setup using my WD media player? I can watch the video but only listen to the soundtrack as 2 channel stereo.:angry:
So started my spiral into madness; a seemingly endless round of:
Internet.
Download software.
Internet for settings/how tos/tutorials.
Trial.
Back to the internet.
All this lasting (and still continuing) the past 2 weeks; including almost 3 solid days of my annual leave. The circles getting smaller and smaller until I feared I would disappear up my own orifice.

Then, success (I thought) the solution was “simple”:
Use Yamb to split the streams.
Use aacDECdrop to convert the AAC file to a WAV file.
Use TranzGUI (I’m a simple soul) to split the WAV into 6.
Rename the files - C=FR, FL=C, FR=FL, LFE=SL, SL=SR and, last, but not least SR=LFE to correct the AAC to AC3 channel order.
Create a .mux file with the above list in.
Use BeSweet (GUI – as I said, I’m simple) to run the .mux file to create my AC3 file.
Use MKVmerge to replace the audio file in the .mp4 with the .ac3 and mux to an .mkv file my media player is happy with.

But no,:eek: my first run was just beginners luck, TranzGUI couldn’t handle some other files (something about the header I’ve read) and BeSweet created distortion (crackles/pops – no snaps) in loud segments (again something about 3db here and 31db there) so it’s back to the net…
…then I find my Nirvana – WaveWizard and WAV to AC3 Encoder! Turns out I’d already downloaded and unpacked them during my week-long frenzy, but never got back to them. But then my channel order nightmare started again. Do I run wavewizard as-is then rename the files 0=FL, 1=FR, 2=C, 3=LFE, 4=SL, 5=SR to get WAV order, then re-rename them C=FR, FL=C, FR=FL, LFE=SL, SL=SR, SR=LFE to go from WAV to AC3 or, if my maths teacher did his job, 0=C, 1=FL, 2=FR, 3=SL, 4=SR, 5=LFE to do it in one go, then follow as before from the .mux file?
No, WAV to AC3 Encoder can do it for me with its wizard!
For Tranz files - click L and navigate to the FR file, R to C, C to FL, LFE to SR, SL to LFE & SR to the SL file.
For wavewizard – click L and navigate to the 1 file, R to 2, C to 0, LFE to 5, SL to 3 & SR to the 4 file.
Then it’s off to MKVmerge with the .ac3 file as before.
I think the above order is correct, but I’m prepared to be corrected, and if anyone knows of (I haven’t found it) or can produce, a comprehensive remapping file-to WAV-to file map (like the above AAC to WAV to AC3 version) I think there would be a large and appreciative audience for it.
Here is your “starter for ten”:

5.1 file
type
WAV FL FR FC LFE SL SR
AC3 FL FC FR SL SR LFE
DTS FC FL FR SL SR LFE
AAC FC FL FR SL SR LFE
AIFF FL SL FC FR SR LFE
FLAC FL FR FC LFE SL SR
WMA FL FR FC LFE SL SR
• FL : Front Left
• FC : Front Centre (Center)
• FR : Front Right
• SL : Surround (Rear) Left
• SR : Surround (Rear) Right
• LFE : Low Frequency Effects (Sub)

[tebasuna51]

Please forget old and discontinued programs like BeSweet, AacDECdrop or TranzGUI.
Also forget the internal channel order in compressed formats.

An AAC decoder, like FAAD2v2.7, output uncompressed WAV files with the correct mapping (FL FR FC LFE SL SR)
An AC3 encoder like Aften (used by WAV to AC3 Encode) need a correct (FL FR FC LFE SL SR) uncompressed WAV file like input.
You don't need know the internal AAC or AC3 channel order.

To recode AAC to AC3 you can use several methods:
1) Command line.
You need a decoder like FAAD2v2.7 and a encoder like Aften with this command:

Faad -w input.aac | Aften -readtoeof 1 - output.ac3

2) Foobar2000.
You need define Aften like external encoder and after load the aac and right clik -> Convert

3) AviSynth methods (SoundOut, Wavi, MeGUI, BeHappy).
You need the Bass libraries to decode and Aften to encode

[starmanj]

After all these years I still find WW to be the most efficient means of quickly converting multichannel sources to DTS. Using foobar then WW it takes just minutes. However I am moving on to DVDA format, using Surcode MLP.

HAS ANYONE figured out how to add Surcode MLP to WW's batch output options? I assume it's hidden within "settingsWW.ini" which is in hex or some compiled code.

Let's breath new life into WW!

Starmanj

[johnman]

After all these years I still find WW to be the most efficient means of quickly converting multichannel sources to DTS. Using foobar then WW it takes just minutes. However I am moving on to DVDA format, using Surcode MLP.

HAS ANYONE figured out how to add Surcode MLP to WW's batch output options? I assume it's hidden within "settingsWW.ini" which is in hex or some compiled code.

Let's breath new life into WW!

Starmanj

Hello starmanj, nice to see some ppl are still using ww :). The commandline arguments for batchconverter are actually hardcoded into the program itself. To use mlp i have to add it to ww and to the converter. Can surcode mlp be used through cli? Or is the only way of doing it to capture the output from the window? The more info you give on how to use mlp from within ww the easier for me it is to add support ;)

[djmasturbeat]

can someone upload/mirror this program somewhere, please?
all the links i tried were dead.

thanks

[poisondeathray]

can someone upload/mirror this program somewhere, please?
all the links i tried were dead.

thanks

wavewizardv0.54b
http://www.mediafire.com/?6i161jp52fyj92m

[ReinerSchweinlin]

http://www.hardware-test.de/doom9/wavewizardv0.54b.rar

[djmasturbeat]

thanks guys

[SeeMoreDigital]

I use this application all the time. Even more-so since I discovered you can use it to 'stitch' DTS.wav files together...

Thank-you JohnMan :)

[johnman]

Just a little message after all those years. Luckily i was still able to login.

Back in the day when ww was developed there was no such thing as github to host and share the sourcecode, but now there are many options. I am considering putting the whole bunch on github. I have been completely out of the audio-conversion scene for ages so im doubtful if its even worth the effort and some feedback from others would help. As far as i can see now, probably not.

Having increased my dev skill since then, i might also add some desperately needed functionality if other software is lacking in anything in particular.

[tebasuna51]

Nice to read you after long time.

I have used your tool long time ago but now there are new tools to do more easy their job because:

- Decoders can output multichannels wavs with the correct wav order signaled with maskchannel, also > 4GB, with wav, w64 or RF64 headers WAVE_FORMAT_EXTENSIBLE (also with AviSynth+ decoders)
- Encoders can read that source wav files and convert the wav order to internal order automatically. No remap is needed.
- ffmpeg can decode many audio codecs and encode automatically to many audio codecs.
- ffmpeg can do remap, mix, merge, resample, amplify, dither ... (many audio filters) and output in any format.

Thanks for your offer but I can't imagine adding new features.

[johnman]

Nice to read you after long time.

I have used your tool long time ago but now there are new tools to do more easy their job because:

- Decoders can output multichannels wavs with the correct wav order signaled with maskchannel, also > 4GB, with wav, w64 or RF64 headers WAVE_FORMAT_EXTENSIBLE (also with AviSynth+ decoders)
- Encoders can read that source wav files and convert the wav order to internal order automatically. No remap is needed.
- ffmpeg can decode many audio codecs and encode automatically to many audio codecs.
- ffmpeg can do remap, mix, merge, resample, amplify, dither ... (many audio filters) and output in any format.

Thanks for your offer but I can't imagine adding new features.

Thanks for the response. I'm glad to read there are good working alternative tools. This would make ww pretty superfluous, so ill refrain from putting it on github.

Running ww again and looking at it with my current eyes it also looks a bit outdated, so putting time into it feels a bit wasteful to say the least. You fairly mentioned the old 4gb issue, i forgot that even was a thing back then after al those years post fat32 :D

Its also nice to see the audio enthousiast are going as strong as ever. I very much still enjoy my music etc, but thats basically all the time i have to spare on this.

[SeeMoreDigital]

I'm happy to report that I still use WaveWizard most days ;)

[johnman]

I'm happy to report that I still use WaveWizard most days ;)

That's honestly nice to hear :cool:

IIRC, after the last release i put in a lot of effort to make it function like a VST hub and make it work with all the VST audio plugins that are around. Unfortunately the world never got to see that.