Log in

View Full Version : multiAVCHD - author Blu-ray/AVCHD (Blu-ray players, camcoders, Viera TV) + (HD) DVD


Pages : 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 [125] 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162

setarip_old
31st July 2010, 20:17
As my mother used to say, "Take your eyes in your hands and look!"

Talk about hiding in plain sight ;>}

Thanks - But why CAN'T it be another "per-title" option? It only takes effect if such a track is discovered, just like the other default audio settings, doesn't it?

deank
31st July 2010, 20:20
It certainly can, but most people want per-title options and not per-compilation.

There is an option in Subtitles&Audio to convert HD audio to AC3, which should give similar results to what you need. :)

Dean

MrVideo
31st July 2010, 20:21
I don't know why you decided that you need avsinfo.exe.

Because of post 6183 from Groucho2004.

I'm still curious as to the difference between multiUPDATE.exe and multiAVCHD.exe.

deank
31st July 2010, 20:37
multiAVCHD executes multiUPDATE.exe when it needs update.

multiUPDATE downloads multiAVCHD.rar, extracts it and launches the extracted multiAVCHD.exe.

Dean

MrVideo
31st July 2010, 20:51
multiAVCHD executes multiUPDATE.exe when it needs update.

multiUPDATE downloads multiAVCHD.rar, extracts it and launches the extracted multiAVCHD.exe.

Ah. Thanks for the info.

MrVideo
31st July 2010, 21:11
I don't know if the updater logs what it does somewhere, but it had issues with the first place that it tried to go to and failed with the next 4 mirrors.

For example:

Trying next mirror...
Downloading...
URL: Mirror#2
HttpSendRequest error
HttpQueryInfo: 12019
Error! (0)

The HttpQuery error line only showed up with mirror #2.

It seems to be downloading from mirror #5, as there are no errors, and nothing else to indicate that it is, or isn't, working.

The suggestion:

Don't put out a message that it is downloading unless it is really downloading, i.e., the connection was fully established and the file it is after is really being downloaded. In other words, something like:

Trying next mirror...
URL: Mirror#2
Downloading...

Thanks.

MrVideo
31st July 2010, 21:22
Here is the complete session log:

Initialising...
Checking...
Downloading...
[14:56:38] Connecting...
[14:56:40] Connect failed!
[14:56:40] Connecting...
[14:56:47] Connect failed!
[14:56:47] Connecting...
[14:56:49] Connect failed!
Error! (500)
Trying next mirror...
Downloading...
URL: Mirror#1
HttpSendRequest error
Error! (0)
Trying next mirror...
Downloading...
URL: Mirror#2
HttpSendRequest error
HttpQueryInfo: 12019
Error! (0)
Trying next mirror...
Downloading...
URL: Mirror#3
HttpSendRequest error
Error! (0)
Trying next mirror...
Downloading...
URL: Mirror#4
HttpSendRequest error
Error! (0)
Trying next mirror...
Downloading...
URL: Mirror#5
InternetReadFile failed for http://multiavchd.deanbg.com/multiAVCHD.exe
Done!

After the done, another updater GUI poped up with the following message:

(RT) Application "C:\multiAVCHD\multiAVCHD.exe" cannot be executed - Line 148.

Sure enough, multiAVCHD is broken. I'll have to install from a previous download and try the update again.

Is there a way to make the updating process more robust so that the current installed version isn't corrupted?

UPDATE: Tried it again. This time it started the download from mirror #2, But this time got the InternetReadFile failure on the RAR file. multiAVCHD.exe gone again.

Third try resulted in the same failure at mirror #2. Going to go fetch the latest version manually.

Capsbackup
31st July 2010, 21:56
This just happened to me too. I had another copy, so just copied that into multiAVCHD folder, and when it opens and asks for an update, just say no.
I am sure deank will correct this soon!

setarip_old
31st July 2010, 22:23
Sure enough, multiAVCHD is broken. I'll have to install from a previous download and try the update again.Just change "multiAVCHD.exe_backup" back to "multiAVCHD.exe"...

MrVideo
31st July 2010, 22:24
... and when it opens and asks for an update, just say no.

Since it is the latest, the update question is not asked :D

Don't know why the updater has so much trouble, since a manual download went real smooth.

setarip_old
31st July 2010, 22:26
@deank There is an option in Subtitles&Audio to convert HD audio to AC3I was already aware of that - and that option is what led me to ask the question about MA to core ;>}

MrVideo
31st July 2010, 22:27
Just change "multiAVCHD.exe_backup" back to "multiAVCHD.exe"...

Oh, didn't know it existed. If the updater knows that it failed, maybe it should put it back. In reality, the file shouldn't be renamed until it is known that a good download was completed and that it has what it needs to do the install.

Just a thought.

setarip_old
31st July 2010, 22:32
In reality, the file shouldn't be renamed until it is known that a good download was completed and that it has what it needs to do the install.If it wasn't renamed BEFORE the downloading began, it would be OVERWRITTEN - and then you'd have no fallback.

Now that you're aware of the file's (probable) existence, why don't you simply try what I suggested?

Capsbackup
31st July 2010, 22:43
Just change "multiAVCHD.exe_backup" back to "multiAVCHD.exe"...

I was not aware of this, but after checking I do not have this file in my multiAVCHD folder!

MrVideo
31st July 2010, 22:45
If it wasn't renamed BEFORE the downloading began, it would be OVERWRITTEN - and then you'd have no fallback.

Not really. As indicated in a Dean posting, the RAR file is downloaded. It can be extracted in a temp location. Once the RAR extraction took place without error, then the executable can be renamed and the install begun.

Now that you're aware of the file's (probable) existence, why don't you simply try what I suggested?

No need, since I downloaded the latest version manually and did the install, as I posted above. But, I'll know the next time this happens.

Still doesn't mean that the automatic updating process can't be made more bulletproof.

MrVideo
1st August 2010, 03:37
Ya, I know, I've been posting for a while, but haven't had the time to actually sit down and work on a project. Well, today I am.

I have the latest build (762).

I encoded the H.264 file myself, creating a 1080p23.976 MKV wrapped program. The audio is AC3 - 2.0.

##########

I just crashed mAVCHD getting it to use a chapter list text file. It complained about not being able to read all of the delimited text. It is a text file that was created by VideoReDo and is a simple file with one timecode entry per line. There are eight entries in the format hh:mm:ss:ff. With the frame being 00 through 23.

Why does mAVCHD use 3-place decimals for the frame? That is not industry standard. Is there an option to tell mAVCHD to use the industry timecode standard?

##########

Now I get to start the project over again. This time I will save the project before dealing with the chapter marks. :D

deank
1st August 2010, 08:09
I added HH:MM:SS:FR support - you'll get it in the next update.

multiUPDATE is updated to use different servers.
multiAVCHD will automatically download and update multiUPDATE.

rendez2k
1st August 2010, 09:19
deank - is there any chance you could add an option in fit all to reduce any HD audio to AC3 (as a 1-off setting for that kind of project)? I know theres an overall option for that but for most discs I'm not bothered. However, if I'm trying to fit an MKV on a DVD-5 I do often want to reduce the audio size to get more video bitrate.

Edit: Is fit all even taking the reduce audio setting into account? With it enabled, the bitrate appears to be no different? The suggested DVD5 setting in transcode is much higher than the fit-all suggested bitrate (and the AR appears to be off too!)
http://i29.tinypic.com/xd8k5s.jpg

deank
1st August 2010, 09:41
When dealing with one title only I'd suggest that you use "Bitrate suggestions" option in [transcode] window. There you get quite accurate bitrate values, taking audio conversion into account.

About the AR: use compliant files. If not - open [transcode] and make sure all is set properly. multiAVCHD can't magically always do everything for everyone and every video file - it is not a one-click-application after all. :)

JoeH
1st August 2010, 13:18
You should use this for that purpose:
http://www.iol.ie/~schubert/avsexec.zip

This has no user interface (apart from messageboxes for error handling) and just steps through a script frame by frame.

Also, it uses the proper Avisynth API and not VFW.

Let me know if it works.

What type of speed difference does using AVSEXEC / AVSINFO make? Can it be used by normal users of say, MeGUI or other typical X264 encoding programs to speed things up?

deank
1st August 2010, 14:24
AvsInfo/AvsExec are not for video encoding - simply for executing avisynth scripts.

***

Latest update includes Boot-Logo when launching multiAVCHD:

http://multiavchd.deanbg.com/multiAVCHD_boot.jpg

When prompted to update, please do. If not, use multiUPDATE to update your multiAVCHD, which in its turn will do all necessary updates itself.

There will be a short delay when you launch multiAVCHD for a first time after the update, while .\res\splash.bmp (http://download.videohelp.com/deanbg/splash.bmp) (550KB) and .\multiBOOT.exe (http://download.videohelp.com/deanbg/multiBOOT.exe) (50KB) are downloaded (total of 600kb). If you wish, you can download these 2 files manually and save them in their proper locations. After a complete reinstall, multiAVCHD icon/link launches multiBOOT.exe. It is required, because multiAVCHD boots really-really slow and some users double/tripple click/start multiAVCHD and it causes troubles.

If you don't want to do a full-download/reinstall, please change your multiAVCHD shortcut to point to multiBOOT.exe.

Please, do not start multiAVCHD.exe directly - it will take more (~2x) time to launch.

Dean

Capsbackup
1st August 2010, 17:02
deank, I have just tried, downloaded twice from 2 different sites, the latest full version, but when I try to install, I get a message that says:
Installer integrity check has failed. Common causes includes incomplete download and damaged media. Contact the installer's author to obtain a new copy.
More information at: http://nsis.sf.net/NSIS_Error :confused:

deank
1st August 2010, 17:05
I'm uploading the installer atm and it is not complete. The upload broke couple of times and you probably don't have the full installer (38 061 834 bytes).


You don't really need it. Just launch multiUPDATE.exe, then close multiAVCHD and restart it. And change the shortcut to multiBOOT.exe which should be present in multiAVCHD's folder.

Capsbackup
1st August 2010, 17:13
OK, done. It works. It appears to launch multiAVCHD in about 7 seconds now. :)
I was not having any problems with it before, seemed to launch acceptibly fine for me!
Of course, with my age does come patience! :p

deank
1st August 2010, 17:16
Same here (launches for 7 seconds), but on my older desktop PC it takes like 120 SECONDS?! And even 10-15 seconds delay may cause users to launch it again and...

The problem is that multiAVCHD is compiled each time it is started and it takes some time on slower machines.

I think the splash screen is a nice looking way to wait for main app to appear.

Dean

Capsbackup
1st August 2010, 17:17
I think the splash screen is a nice looking way to wait for main app to appear.

Agreed!

deank
1st August 2010, 17:22
Full-download link (http://download.videohelp.com/deanbg/multiAVCHD_4.1.exe) is now available.

MrVideo
1st August 2010, 18:55
I added HH:MM:SS:FR support - you'll get it in the next update.

Thanks, very much appreciated.

Was that the reason for the failure with the chapter file that I have?

multiUPDATE is updated to use different servers.
multiAVCHD will automatically download and update multiUPDATE.

Attempting autoupdate. Will do manual if it fails.

deank
1st August 2010, 19:13
multiAVCHD supports about 10 chapter formats, it is that :FR(ame) one was not one of them.

Internally multiAVCHD uses HR:MM:SS.MLS - milliseconds and not frames. multiAVCHD uses all types of framerate conversion and the only way to keep all correct is to use TIME-stamps and not frames and that's why if you load a chapter file with :FR format it will be converted to .MLS milliseconds.

Dean :rolleyes:

MrVideo
1st August 2010, 19:15
I added HH:MM:SS:FR support - you'll get it in the next update.

Since I couldn't find a way to set the timecode format, I have to assume that you didn't mean the update that just happened, but for build 763.

deank
1st August 2010, 19:16
Build 762 accepts HR:MM:SS:FR format. Try to load your chapter file - it should work.

It will even work if your chapter file looks like

00:00:01.001
00:02:03:20
1681
00:12:06
00:18:12;00:19:31.434

The update uses the same build number, because of minor changes.

MrVideo
1st August 2010, 19:21
Internally multiAVCHD uses HR:MM:SS.MLS - milliseconds and not frames. multiAVCHD uses all types of framerate conversion and the only way to keep all correct is to use TIME-stamps and not frames and that's why if you load a chapter file with :FR format it will be converted to .MLS milliseconds.

In that case, any idea why the chapter file that was created from VideoReDo would cause mAVCHD to crash?

File attached.

deank
1st August 2010, 19:23
The attachment needs approval. Can you copy/paste it in [ code ] or post in multiFORUM threads so I can take a look?

Before you do, please use multiUPDATE.exe to update your multiAVCHD - just to be on the safe side.

A chapter file with:

00:00:00:00

each on separate line works for sure and I tested it yesterday.

I used this one for testing:

00:00:00:00
00:07:59:20
00:15:59:11
00:23:59:22
00:31:59:03
00:39:59:14
00:47:59:05
00:55:59:06
01:03:59:17
01:11:59:18
01:19:59:09

and got

00:00:00.000
00:07:59.800
00:15:59.440
00:23:59.880
00:31:59.120
00:39:59.560
00:47:59.200
00:55:59.240
01:03:59.680
01:11:59.720
01:19:59.360

for 25fps video.

MrVideo
1st August 2010, 19:25
Build 762 accepts HR:MM:SS:FR format. Try to load your chapter file - it should work.

It didn't crash this time, but it only used the chapters that ended in :00

The other 6 were ignored.

MrVideo
1st August 2010, 19:28
The attachment needs approval. Can you copy/paste it in [ code ] or post in multiFORUM threads so I can take a look?

Sorry, didn't notice that it was tagged pending. Not used to that.

00:00:29:23
00:01:08:00
00:13:16:01
00:21:10:00
00:31:59:06
00:38:33:12
00:42:56:23
00:47:51:19

MrVideo
1st August 2010, 19:30
Before you do, please use multiUPDATE.exe to update your multiAVCHD - just to be on the safe side.

Oh, and the update was accomplished. Now using the new boot startup.

deank
1st August 2010, 19:31
Pasting your chapter marks works and got me this:

00:00:00.000
00:00:29.920
00:01:08.000
00:13:16.040
00:21:10.000
00:31:59.240
00:38:33.480
00:42:56.920
00:47:51.760

(assuming 25fps video)

MrVideo
1st August 2010, 19:46
Pasting your chapter marks works and got me this:
(assuming 25fps video)

As noted in your other forum (probably should have noted it here):

1080p23.976

deank
1st August 2010, 19:57
Hmm... Just tried with 23.976 and chapter marks are missing. Let me check why... I guess it is a rounding issue.

/./

Execute multiUPDATE.exe - it is fixed now. Sorry about that.

Dean

MrVideo
1st August 2010, 20:32
Execute multiUPDATE.exe - it is fixed now. Sorry for that.

'Tis OK. Just making sure that I wasn't seeing things.

BTW, when someone imports a 29.97 frame/sec project and imports a chapter list, how do you know that they are using DFTC or NDFTC? With 720p59.94, it is even worse, since two consecutive frames have the same timecode, normally marked with a character change before the frame value, i.e., 00:00 and 00.00

Just imported the chapter text file (after adding a chapter manually at time 0), and noticed that the math is was off: (TC Mine Yours)

00:00:00:00 000 000
00:00:29:23 958 293
00:01:08:00 000 000
00:13:16:01 042 708
00:21:10:00 000 000
00:31:59:06 250 .25
00:38:33:12 500 501
00:42:56:23 958 293
00:47:51:19 792 459

The frames must be frame divided by 24, not 23.976. I just noticed that VideoReDo is using NDFTC with 23.976 video. I'm not even sure what the formula is with 23.976 video using DFTC. I'll have to figure that out.

And no, it is not a typo in that the .25 is displayed as: 00:31:59..25

In the little thumbnail display of the video, you can't get to frame 000, it always displays as 001. Plus the F>> button doesn't skip a single frame, it seems to skip 6 frames, i.e., four clicks and you are at the 1 second mark, which does display as 1.000. All three of the times that are displayed within the first second are .001 too high.

deank
1st August 2010, 20:51
The .25 is fixed.

Rounding 23.976 to 24 will not be implemented.

You can manually enter a chapter-jump point in "CHAPTERS" tab. Enter a value and click the box where you entered it - preview point will seek to the position.

Dean

MrVideo
1st August 2010, 20:53
I just noticed that VideoReDo is using NDFTC with 23.976 video. I'm not even sure what the formula is with 23.976 video using DFTC. I'll have to figure that out.

I'm not sure that there can be DFTC for 23.976 video. The first point at which 23.976 fps come out at an even number of frames is 25 minutes, or 35964 frames. At 24 fps that is 36000 frames. It is only off 36 frames, or about 1.5 seconds every 25 minutes. Yes, it is an error. But the 1.44 frames that it is off every minute doesn't work out evenly, that is to just frames forward at every minute, or two minutes (that I can easily see)

deank
1st August 2010, 21:03
23.976fps 2 hour movie with chapter set at the last second:

23.976 = frame 172627
24.000 = frame 172800 (there is no such frame and all subsequent chapter processing / thumbnail extraction / menu jumps will fail).

The difference is 7 seconds and all tools within multiAVCHD work with millisecond precision, so there is no way I use rounding up or non-dropframe format for drop-frame framerates.

Dean

MrVideo
1st August 2010, 21:58
The .25 is fixed.

What about the other math errors? 23/23.976 is not .293, etc.

UPDATE: I see that the other math error is fixed as well.

You can manually enter a chapter-jump point in "CHAPTERS" tab. Enter a value and click the box where you entered it - preview point will seek to the position.

I just now noticed that you can set a jump value for the F>> and <<F buttons. Like I said, I need new glasses :D

There is something strange going on with the chapter time entry box. While playing with a value that is already there, and building from that, here is what happens:

Chapter box value - thumbnail displayed image value
00:00:29.293 00:00:29.292
00:25:29.293 00:25:29.292
00:25:29.29 00:24:09:965 <-- way out of place
00:25:29.290 00:25:29.289
00:25:29.289 00:25:29.288
00:25:29.288 00:25:29.288
00:25:29.287 00:25:29.287
00:25:29:12 00:24:09.965 <-- way out of place

Rounding errors for the rest of them. I would expect an entered timecode to go to the nearest appropriate frame, but the thousandths values changes a lot a well.

The .29 and .290 should be handled the same.

For the last sample, it doesn't want to work with frames as the last value, even though I indicated it was a frame value with the : in front of it.

UPDATE: I just noticed that the 24:09.965 location is your starting point for this file.

You jump there if you don't know how to handle the entered value? If so, not very intuitive to the user.

MrVideo
1st August 2010, 22:11
23.976 = frame 172627
24.000 = frame 172800 (there is no such frame and all subsequent chapter processing / thumbnail extraction / menu jumps will fail).

Ah, misunderstanding. The divide by 24 only for the frame, not the whole timecode. I know that basing it on the whole timecode value will be a major problem. In any event, not a big deal. With years of editing, even with Umatic tape, I'm used to timecode with frames, not decimals for frames.

The difference is 7 seconds and all tools within multiAVCHD work with millisecond precision, so there is no way I use rounding up or non-dropframe format for drop-frame framerates.

See the above posting about strange millisecond errors.

How do you know if someone is supplying NDFTC or DFTC with a chapter list for 29.97 video? With VideoReDo, I can set it for DFTC (my preferred time display) or NDFTC and the chapter list that is created will be written based upon my setting.

deank
2nd August 2010, 10:07
Frame -> milliseconds conversion is fixed

00-23 (for 23.976 and 24fps)
00-24 (for 25.000 fps)
00-29 (for 29.970 and 30fps)

Also, CHAPTER time code box and CHAPTER-THUMBNAIL time code box allow usage of HH:MM:SS:FR format.

Please note that the delimiter is : and not . (dot), so 00:01:02.03 will not be accepted (as you posted in your example).

Dean

Karl Beem
2nd August 2010, 15:30
I'm new to Multiavchd. I can't get menus to work. The guide says that you can have up to 252 scrollable titles. When I try, they don't scroll, nor can they be selected or executed. I can't produce anything like the "Favorite clips" menu in the guide. I have no control over where the selections go. BTW, I'm only interested in hard drive playback - I have no interest in burning to optical. Which project output type should I choose?

deank
2nd August 2010, 15:32
What player will you use?

Karl Beem
2nd August 2010, 15:36
What player will you use?

Possibly TMT3.

deank
2nd August 2010, 15:38
Then you can use:

[Sony Playstation 3]
[AVCHD relaxed]
[Blu-ray]

All three will work with TMT3. To play the structure with the menu you'll have to drag the BDMV folder to TMT3.

Dean