BeLight 0.22 beta 8 released

[drob]

Kurt still problems with batch processing, when you throw a couple of files, BeLight marks all the job as done but actually process only the last file.

[Kurtnoise]

I know and I'm busy with my real life now. I hope to fix this this week-end.

[drob]

real life always comes first, thanks for help.

[suspiciousBob]

Will the mp3 area of BeLite be changed for LAME 3.97 and the new preferred switches as documented here (http://www.hydrogenaudio.org/forums/index.php?showtopic=28124)?
BTW great job on this project fella.

[Kurtnoise]

Hi,

BeLight uses already lame_enc.dll from 3.97 beta version if you use the whole package you know...;) Concerning the settings, I already managed these recommendations. If you look the command line, you can see that BeLight uses the recommended settings most of the time. However, there are some addition in Quality Mode caused by BeSweet restriction.

The only point which must be checked today is the MP3 presets for BeLight. Of course, I'll look into this. :)

[Immi]

Hi,

BeLight uses already lame_enc.dll from 3.97 beta version if you use the whole package you know...;)

Hi Kurt,

together with WMP10 (only Win XP users) everybody get the Fraunhofer MPEG Layer-3 Audio Codec Professional (it is ACM codec located at %windir%\sytem32\l3codecp.acm after installing WMP10).
Maybe ( :stupid: I am not an programming expert) using of this codec also can be implemented into Besweet/Belight.
BR, Immi

Off topic PS:
Tip: Using of prof codec in any encoding programs (ie. EAC, Vdub etc.) apply the following registry modification

Windows Registry Editor Version 5.00

[HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Drivers32]
"msacm.l3acm"="C:\\WINDOWS\\system32\\l3codecp.acm"

[Kurtnoise]

mmmh, it's possible I think. There is a tool which works with acm filters. But I'm pretty sure that Lame is probably more interesting than the Fhg codec. :) Do you perform some tests with this codec ?


Edit : here is the tool (http://nyaochi.sakura.ne.jp/xoops/modules/mysoftwares/tc_3.html).

[Immi]

mmmh, it's possible I think. There is a tool which works with acm filters. But I'm pretty sure that Lame is probably more interesting than the Fhg codec. :) Do you perform some tests with this codec ?


Edit : here is the tool (http://nyaochi.sakura.ne.jp/xoops/modules/mysoftwares/tc_3.html).

Hi,

I look the net for this info and generally it is said that the FHF codec quite old and outdated.

At the weekend I try to check this.

The method:
So I am unable to hear any difference by my ears, consequently I had to compare the encodings sample by sample.
Source: ripped (EAC) typical music (with vocal+drums+hiss) PCMwav
Encodings: EAC wav encoding funciton using
- FraunhoferACM CBR 80 mono
- LameACM CBR 80 mono
- FraunhoferACM CBR 128
- LameACM CBR 128
- FraunhoferACM CBR 320
- LameACM CBR 320
Additionally Besweet encoding using
- Lame.dll CBR 80 mono
- Lame.dll CBR 128
- Lame.dll CBR 320

To get the differences between the original and the encoded files inverted mixing was applied (mix the encoded file into the original wav inverted).
This mixing was checked, because when the original wav is invert mixed into itself I get pure flatline(silence) as result.

Now the tricky part. To get proper difference file, the sample accurate timing is needed. Problem is that the encoding process insert silence in the begining of the audio file. So I "draw" a timer signal in the original then always cut out the not necessary silences to get the sample accurate match of original and encoded file.

For the attachments please download
http://www.mytempdir.com/211968

Attachment: the original timer signal and the timer after encoding
original - 080lame compare.jpg
Attachment: Timer signal section after application of inverted-mixing
original - 080lame difference.jpg

Results:
Codec - Encoding time - Encoded audio file size
difference file in the attachment

80kbs mono:
FHF ACM 0:19 2654892 bytes
difference original-080fhf.jpg
Lame ACM 0:12 2655574 bytes
difference original-080lame.jpg
Lame dll 0:10 2657043 bytes
difference original-080lamedll.jpg

128kbs:
FHF ACM 0:34 4241724 bytes
difference original-128fhf.jpg
Lame ACM 0:23 4249260 bytes
difference original-128lame.jpg
Lame dll 0:21 4251190 bytes
difference original-128lamedll.jpg

320 kbs:
FHF ACM 0:15 10618524 bytes
difference original-320fhf.jpg
Lame ACM 0:19 10615392 bytes
difference original-320lame.jpg
Lame dll 0:19 10627785 bytes
difference original-320lamedll.jpg

"Additional test" VBR-insane:
lame dll 0:19 8246932 bytes
difference original-vbr-insane-lamedll.jpg

Now you can see the the lost info-data is more in the Fraunhofer encoded files. Furthermore Fraunhofer only support CBR encodings.

So you are right: Lame is probably more interesting than the Fhg codec

BR, Imre

PS.: It was inresting to listen the difference file. You can clearly hear the lost audio data. It is mainly the high freqency and slighly the mid freqency part of the audio.

[Kurtnoise]

Kurt still problems with batch processing, when you throw a couple of files, BeLight marks all the job as done but actually process only the last file.
fixed...:)

http://corecodec.org/frs/?group_id=45&release_id=175#r175

[Kurtnoise]

At 48KHz the average bitrate of q-2 is 32kbps, at 8KHz the average bitrate is 10kbps, a big diference. See OggDropXpd for precise nominal bitrates for each sample rate.
Which version did you have ? coz I don't see any changes when I choose the Sampling Rate...

And bear in mind that nominal bitrate and average bitrate are 2 concepts completely different.

[dwrbudr]

@Kurt: You should again look at the files list processing, cause it is broken again in the last daily build.

Example:

1. File - Open - 1.mp3 - View BeSweet Command

BeSweet.exe -core( -input "C:\Temp\2\1.mp3" -output "C:\Temp\2\1.aac" -logfile "C:\Temp\2\1.log" ) -dimzon( -dllname bse_WinampAacPlusV2.dll -cbr 32 -mode 1 -v 2 ) -ota( -hybridgain )

2. Popup menu - Clear List

3. File - Open - 2.mp3 - View BeSweet Command

BeSweet.exe -core( -input "C:\Temp\2\1.mp3" -output "C:\Temp\2\1.aac" -logfile "C:\Temp\2\1.log" ) -dimzon( -dllname bse_WinampAacPlusV2.dll -cbr 32 -mode 1 -v 2 ) -ota( -hybridgain )

See - in the list is 2.mp3, but in BeSweet CMD - 1.mp3 :eek:

[Kurtnoise]

Indeed...shame on me. :o

http://corecodec.org/frs/?group_id=45&release_id=175#r175

[Caroliano]

Which version did you have ? coz I don't see any changes when I choose the Sampling Rate...
You must close the configuration window for it update the values... When you open again the scale will be changed.

[humax]

when i took 2 files in List to do .

First file gets skipped and second will be encoded.

First 1 comes message ready after 1 second and second file is encoding right. A NEW or KNOWN Bug ??

cu

[Kurtnoise]

works fine here with the last daily build...

@Caroliano : ok...thanks.

[humax]

Hi

Definitly doesnt work queue encoding .
And is the Quit Button gone ?? I cant find a button to stop encoding
Encoding stops not . It is at 1 second rest and the watch runs and runs

I made two Pictures. First where the clock runs and runs and second that after 1 second beelight says after one second first job is done . second job makes he then.

I am using Beelight 19.10 Build and Besweet 1.5b31

EDIT :

If i only make 1 job to do - abort Buton is available and encoding works fine
See on Picture 3.

cu


PLZ Approval my Attachements !! THX

[LoRd_MuldeR]

Hi.
The latest Build (BeLight-0.22beta_20051019) makes the GUI look this way:

http://img487.imageshack.us/img487/8203/belight3ng.th.png (http://img487.imageshack.us/my.php?image=belight3ng.png)


// EDIT

If I move the bitrate-slider in the AAC Tab (Winamp) the combobox below switches .style between csDropDownList and csDropDown. IMO there seems to be a problem, or is there a reason for this?

[Kurtnoise]

Did you modify the BeLight borders ?

[LoRd_MuldeR]

Did you modify the BeLight borders ?

I did *not* modify anything. It's the EXE file from the latest daily build.

[fight2win]

winamp has been updated, so will winamp aac support in belight will also be updated?

[Kurtnoise]

Do you know if it has been improved ? I don't think so...btw, you can replace the old dll by the new one.




@Lord_Murder : I found the problem... :rolleyes: Sorry for that.

[ron spencer]

3 bugs, all after you hit start and GET TO THE TRANSCODE WINDOW.

1. The time elapsed and left have 12 in from of them...representing a clock, this is wrong....also AM is listed there.

2. If you try to resize the transcode log big dialog box the red cancel button does not move to the right place...it ends up over the time left bars.

3. Again in the transocde window area if you try to move over the information components so you can see them (the box above the progress bars) they will bounce back


thanks!!!!

[LoRd_MuldeR]

np :)
But did you upload a new build which includes the fix ???

[Kurtnoise]

No coz I'm not at home during this week. So, maybe the next weekend...

[ron spencer]

ok thanks

[nexus]

Hi,

we are currently working on a better german translation at Doom9 | Gleitz DVD-Forum.

A few issues:


The translations of Tempo_Hint_part1 and Tempo_Hint_part2 aren't used, it's always english
Pitch_Hint_part1 and 2 and Rate_Hint_part1 and 2 aren't shown.


You can find the latest translation in this thread (http://forum.gleitz.info/showthread.php?t=25495) (German forum).

[Kurtnoise]

Danke schön... :D

[drob]

Kurt i can confirm, with latest nightly build, Que still isn't working.

[ron spencer]

that is an old build...he has not been able to make a new one as out of town i think

[nexus]

Danke schön... :DBitte schön. :)

Newly discovered issue: :D
In AAC Encoding Tab the descriptions aren't sourced out to the .ini, is that right? This concerns FAAC and Winamp encoder.

You (all) can find almost every day an updated version of Deutsch.ini in the thread mentioned above. ;)

[Kurtnoise]

Newly discovered issue: :D
In AAC Encoding Tab the descriptions aren't sourced out to the .ini, is that right? This concerns FAAC and Winamp encoder.
not an issue. It's because I don't translated these parts yet...I'm working on it. :)

You (all) can find almost every day an updated version of Deutsch.ini in the thread mentioned above. ;)
ok...did you start with the Megamacy translation or did you made the German translation from scratch ?



@Drob : I know...I'm working on it.

[megamacy]

hm, seems like i'm out of work now ^^

but my actual translation is for the latest daily build ;)
(http://corecodec.org/forum/forum.php?thread_id=158&forum_id=128)

[nexus]

ok...did you start with the Megamacy translation or did you made the German translation from scratch ?We started with the English.ini because I didn't knew that there's already a Deutsch.ini. But I thinks ours is better. ;)

It's because I don't translated these parts yet...I'm working on it.Can you post it if you add new lines to the language.ini?

@megamacy: Thought you won't work on the Deutsch.ini anymore?! It doesn't make sense to work on two different Deutsch.ini files. If there are any problems please contact me via Germans forum PM or post in the accordant thread there.

[Kurtnoise]

Can you post it if you add new lines to the language.ini?
yep...in the next build.

[nexus]

yep...in the next build. Great!

Everybody seems to be satisfied with our latest translation so it will be a good idea if you add our language file in the next build. As always: here (http://forum.gleitz.info/showthread.php?t=25495) . ;)

[Kurtnoise]

Check out the lastest daily build (http://corecodec.org/frs/?group_id=45&release_id=209#r209) to update your file...;)

[Elias]

So how long will BeLight be Beta? I don't think it's fair to call it beta any longer, because it's pretty stable nowadays :)

[MacAddict]

Latest daily build breaks Nero encoding with "unable to get bsn interfaces"??? Or is it just me?

[Kurtnoise]

So how long will BeLight be Beta? I don't think it's fair to call it beta any longer, because it's pretty stable nowadays :)
No it isn't...

[Kurtnoise]

Latest daily build breaks Nero encoding with "unable to get bsn interfaces"??? Or is it just me?
be sure to have bse_Nero7WA.dll...I'll add condition to switch bse_Nero7 and bsn asap.

[MacAddict]

Thanks Kurt, that did the trick :-)

[nexus]

Check out the lastest daily build (http://corecodec.org/frs/?group_id=45&release_id=209#r209) to update your file...;)Deutsch.ini updated. Maybe you should include it in the daily build.

[Marcovelo]

Hello experts !

I'm trying to encode a 6ch wav in AAC using BeLight 0.2.2.0 beta 8 and BeSweet v1.5b31.

I'm getting this log:

BeSweet v1.5b31 by DSPguru.
--------------------------
Using Shibatch.dll v0.25 by Naoki Shibata & DSPguru (shibatch.sourceforge.net).
Using bse_FAAC by dimzon, Build Sep 14 2005, 11:26:17

Logging start : 11/06/05 , 18:54:40.

BeSweet.exe -core( -input C:\Temp\util_Audio\BeLight\TestSequence_48_16b.wav -output C:\Temp\util_Audio\BeLight\TestSequence_48_16b.aac -logfile C:\Temp\util_Audio\BeLight\TestSequence_48_16b_3.log ) -dimzon( -dllname bse_FAAC.dll -b 128 --obj-type LC -6chnew )

[00:00:00:000] +------- BeSweet -----
[00:00:00:000] | Input : C:\Temp\util_Audio\BeLight\TestSequence_48_16b.wav
[00:00:00:000] | Output: C:\Temp\util_Audio\BeLight\TestSequence_48_16b.aac
[00:00:00:000] | Floating-Point Process: No
Error 84: error configuring bsn!
Quiting...
[00:00:00:000] Conversion Completed !

Logging ends : 11/06/05 , 18:54:40.

Strange, because I didn't invoke -bsn option (Nero), -dimzon( -dllname bse_FAAC.dll ...

By the way, I was able to encode my 6chwav using FAAC.exe (1.24) from commandline:
-> faac.exe TestSequence_48_16b.wav -q 125 -o TestSequence_48_16b.aac

The resulting .aac seems to be correct when played in VLC 0.8.2, with correct channel mapping (5.1).

Any suggestion about my problem with BeLight would be very appreciated...
Thanks in advance

Marco

PS: Big thanks to Kurtnoise and Dimzon for BeLight and plugins .

[Kurtnoise]

Deutsch.ini updated. Maybe you should include it in the daily build.
Thanks..I'll included it tomorrow. I need to tune different things before to release a new build.

[Kurtnoise]

Marcovelo : what kind of Sampling Rate did you have concerning your input stream ?

[Marcovelo]

Kurt:
my imput stream is Wav (non extensible): TestSequence (http://marcovelo.free.fr/Tests/TestSequence_48_16b.zip)
samplerate = 48000
channels = 6
codec = PCM
bitspersample = 16

I built this input wav using BeSweet from 6 mono wav 48Khz 16 bits:
besweet -core ( -input "list.mux" -output "TestSequence_48_16b.wav" -6chwav )


Thanks for your advices

Marco

[tebasuna51]

For me work without problems.
BeSweet v1.5b31 by DSPguru.
--------------------------
Using Shibatch.dll v0.25 by Naoki Shibata & DSPguru
(shibatch.sourceforge.net).
Using bse_FAAC by dimzon, Build Sep 14 2005, 11:26:17

Logging start : 11/07/05 , 01:24:28.

BeSweet.exe -core( -input G:\TestSequence_48_16b.wav
-output G:\TestSequence_48_16b.aac
-logfile G:\TestSequence_48_16b.log )
-dimzon( -dllname bse_FAAC.dll -b 128 --obj-type LC -6chnew )

[00:00:00:000] +------- BeSweet -----
[00:00:00:000] | Input : G:\TestSequence_48_16b.wav
[00:00:00:000] | Output: G:\TestSequence_48_16b.aac
[00:00:00:000] | Floating-Point Process: No
[00:00:00:000] | Source Sample-Rate: 48.0KHz
[00:00:00:000] +---------------------
[00:00:05:836] Conversion Completed !
[00:00:01:000] <-- Transcoding Duration

Logging ends : 11/07/05 , 01:24:29.
The wav6 is absolutely correct and the aac plays fine and well channel mapping.

[LoRd_MuldeR]

The new built (BeLight-0.22beta_20051106.zip) still looks like this:

http://img471.imageshack.us/img471/1971/belight4xd.png

[Kurtnoise]

weird...try to remove BeLight.ini file and reload BeLight coz I really don't know why. :confused:


btw, for those who want to test a new preversion I updated a new daily build : http://corecodec.org/frs/?group_id=45&release_id=212#r212

[Marcovelo]

@tebasuna51:

Thanks for your test.
I'll check again my configuration, as I still don't see why it does not work on my side.
I just tried again with the same (short) path for file names, but I still get the same error...
I think I have exactly the same version of BeSweet (v1.5b31) and bse_FAAC.dll (Build Sep 14 2005, 11:26:17)...
Very strange...
I will tell you if I found the solution...


@Kurt:

I also experience the same problem of display than reported by LoRd_MuldeR on today's BeLight build.
Deleting BeLight.ini and running again BeLight.exe is not better.
This small pb is not present with BeLight 0.21 on the same PC.