BD Rebuilder Beta - Bug Reports Only

[MrT.]

thanks jdobbs ! :D

[jdobbs]

It's got two quad-core Xeons and 12 GB of RAM. That's a total of 8 physical cores, running 16 threads. So I'd say your guess is pretty much on the nose, if BD-RB thinks threads = processors.


The encoding speed during Pass 2 actually got faster when one of the three instances quit after finishing its part. That makes me very interested to see what happens during Pass 2 with 6-way split.

A single instance maxes out my CPU(s) but only during Pass 2. Pass 1 has been the biggest bottleneck in the past; a single instance barely breaks the 10% CPU usage barrier. No wonder 10 instances worked great for Pass 1 FPS-wise. Too bad the system ran out of RAM.


10 might have been fine if the PC had had enough RAM. I'll set multiprocess=6 for the next test.
Interesting that your system sets the NUMBER_OF_PROCESSORS environment variable to the thread count rather than the actual "number of processors". That's kinda' cheating a bit.

Unfortunately, if you're not seeing any "left-over" processor time it's unlikely you'll get any speed improvement in pass 2. But using 6 instances on your system probably wouldn't slow it down either.

You know, there might actually be a way to use multiple processes in pass 1 and only one in pass 2. I think all I'd have to do is combine the stats and mbtree files before the run -- and then use the original AVS file (before the split). I'd have to research the mbtree file format, though... but the source code is available so that's not a huge challenge. I'm not sure how many systems other than yours would see much benefit, though.

[mayhem2408]

How do I set the Movie and Menu Only mode as shown in MrT post? It appears buy the new files and the Blank option is close. Can it be enabled yet?

[jdobbs]

How do I set the Movie and Menu Only mode as shown in MrT post? It appears buy the new files and the Blank option is close. Can it be enabled yet? Search this thread for "ENABLE_TEST". It is still experimental -- but there is a post explaining how to use it.

[mayhem2408]

Sweet. Found It. Enabled It. Now testing it. Thanks JD

[jdobbs]

Sweet. Found It. Enabled It. Now testing it. Thanks JD Hopefully you found this link (http://forum.doom9.org/showthread.php?p=1526277&highlight=enable_test%27#post1526277).

[bluballz]

jdobbs is there an undocumented setting for the size limit, or how bdrb calculates what are extras, when it runs quicker encode for extras?

For example, the way I have it setup, it does 1 pass crf 18 encodes on what it deems "extras" but does 2 pass high quality on what it defines the main feature. Currently it is running 2 passes on a 7gb extras video file which is 38 minutes, has 1 video 1 audio and 3 subtitle tracks.

Awesome work by the way, keep it up :)

[AmigaFuture]

I, Too, am noticing Sweet speed increase with my 8 Core system (i7 930) and slightly overclocked with 6GB of RAM and multiple HDDs. I originally had the PageFile disabled, but noticed for the Re-Encodes to operate successfully with MULTIPROCESS it has to be turned ON, haha. I was using MULTIPROCESS=1 1st and thought "Wow! Nice!" I'm testing from a Clean 0.40.1.

I'm glad to see the ENABLE_TEST as I missed that before. I'll check that, now, too. My encodes from MULTIPROCESS is looking great...and in sync.

[Yordan5]

I, Too, am noticing Sweet speed increase with my 8 Core system (i7 930) and slightly overclocked with 6GB of RAM and multiple HDDs. I originally had the PageFile disabled, but noticed for the Re-Encodes to operate successfully with MULTIPROCESS it has to be turned ON, haha. I was using MULTIPROCESS=1 1st and thought "Wow! Nice!" I'm testing from a Clean 0.40.1.

I'm glad to see the ENABLE_TEST as I missed that before. I'll check that, now, too. My encodes from MULTIPROCESS is looking great...and in sync.

Well, I guess that's the minimum spec one needs to have in order to take advantage of MULTIPROCESS. I have 4 Core Processor (i5 2500K not overclocked) and 6Gb of RAM. I ran the same file (bluray 26Gb to bluray DVD5) three times. Without MULTIPROCESS, with MULTIPROCESS=2 and with MULTIPROCESS=3. The fastest was MULTIPROCESS=2 which finished just 2min quicker. The other two were identical in time.

[bassnut]

Sweet ..... Took a bit longer than I thought to get back as my OC needed some tweeking. I Ran Real Steal last night ..... All settings exactly the same on both runs.

AMD PII 1090T @ 4.2GHZ 4 GIG DDR3 1800 CL7 RAM

Old BDRebuilder
[10:13:03] BD Rebuilder v0.39.07 (beta)
- Source: REAL_STEEL
- Input BD size: 46.12 GB
- Approximate total content: [03:58:29.961]
- Target BD size: 22.95 GB
- Windows Version: 6.1 [7601]
- Auto Quality: Good (Very Fast), ABR
- Audio Settings: AC3=0 DTS=0 HD=1 Kbs=640

[13:02:50]PHASE TWO - Rebuild Started
- [13:02:50] Rebuilding BD file Structure
[13:03:06] - Encode and Rebuild complete
- WORKFILES folder removed.
[13:03:07]JOB: REAL_STEEL finished.

2Hours 50 Min

New BDRebuilder

[22:09:01] BD Rebuilder v0.40.02 (beta)
- Source: REAL_STEEL
- Input BD size: 46.12 GB
- Approximate total content: [03:58:29.961]
- Target BD size: 22.95 GB
- Windows Version: 6.1 [7601]
- Auto Quality: Good (Very Fast), ABR
- Audio Settings: AC3=0 DTS=0 HD=1 Kbs=640

[23:35:13]PHASE ONE complete
[23:35:13]PHASE TWO - Rebuild Started
- [23:35:13] Rebuilding BD file Structure
[23:35:29] - Encode and Rebuild complete
- WORKFILES folder removed.
[23:35:29] JOB: REAL_STEEL finished.


1 Hour 29 Min

[colinhunt]

Interesting that your system sets the NUMBER_OF_PROCESSORS environment variable to the thread count rather than the actual "number of processors". That's kinda' cheating a bit.
Yeah ;)


But using 6 instances on your system probably wouldn't slow it down either.
1080p24 original, 290 138 frames, 2-pass Highest Quality, --tune film

2-way split
Pass 1 - 119 minutes
Pass 2 - 301 minutes

3-way split
Pass 1 - 102 minutes
Pass 2 - 295 minutes (last 25% with 2 instances)

5-way split
Pass 1 - 94 minutes
Pass 2 - 308 minutes

6-way split
Pass 1 - 94 minutes
Pass 2 - 308 minutes

v0.39.07 - 1 instance
Pass 1 - 168 minutes
Pass 2 - 305 minutes

Conclusions: more instances = better for pass 1, as long as RAM doesn't run out. Pass 2 benefits little from additional instances; best balance might be found somewhere between 3 and 6 instances.

You know, there might actually be a way to use multiple processes in pass 1 and only one in pass 2. I think all I'd have to do is combine the stats and mbtree files before the run -- and then use the original AVS file (before the split). I'd have to research the mbtree file format, though... but the source code is available so that's not a huge challenge. I'm not sure how many systems other than yours would see much benefit, though.
Pass 1 certainly benefits from more instances. It's another matter if it's worth the hassle involved, seeing how little encoding times differ in pass 2.

** update
Fixed processing time for 6-way pass 1. Forgot to carry the one, so to speak.
Added results from 5-way run.

[busch42]

Just installed the new version of BDrebuilder, Bought the new Bluray movie Drive yesterday and ripped to hard drive, Now I went to use BDrebuilder to shrink down to a bd25 and it came up, Failed video encode, aborted. Here's the full report.

[06:42:58] BD Rebuilder v0.40.02 (beta)
- Source: DRIVE
- Input BD size: 44.82 GB
- Approximate total content: [03:03:08.684]
- Target BD size: 22.95 GB
- Windows Version: 6.1 [7601]
- Auto Quality: Good (Very Fast), ABR
- Audio Settings: AC3=0 DTS=0 HD=0 Kbs=640
[06:42:58] PHASE ONE, Encoding
- [06:42:58] Processing: VID_00703 (1 of 13)
- [06:42:58] Extracting A/V streams [VID_00703]
- [06:43:13] Reencoding video [VID_00703]
- Source Video: MPEG-4 (AVC), 1920x1080
- Rate/Length: 23.976fps, 2,857 frames
- [06:43:13] Reencoding: VID_00703, Pass 1 of 1
- [06:44:23] Video Encode complete
- [06:44:23] Processing audio tracks
- Track 4352 (eng): Keeping original audio
- [06:44:23] Multiplexing M2TS
- [06:44:31] Processing: VID_00002 (2 of 13)
- [06:44:31] Extracting A/V streams [VID_00002]
- [06:44:47] Reencoding video [VID_00002]
- Source Video: MPEG-4 (AVC), 1920x1080
- Rate/Length: 23.976fps, 6,863 frames
- [06:44:47] Reencoding: VID_00002, Pass 1 of 1
- [06:47:03] Video Encode complete
- [06:47:03] Processing audio tracks
- [06:47:03] Multiplexing M2TS
- [06:47:18] Processing: VID_00705 (3 of 13)
- [06:47:18] Extracting A/V streams [VID_00705]
- [06:47:38] Reencoding video [VID_00705]
- Source Video: MPEG-4 (AVC), 1920x1080
- Rate/Length: 23.976fps, 3,169 frames
- [06:47:38] Reencoding: VID_00705, Pass 1 of 1
- [06:48:52] Video Encode complete
- [06:48:52] Processing audio tracks
- Track 4352 (eng): Keeping original audio
- [06:48:52] Multiplexing M2TS
- [06:49:02] Processing: VID_00702 (4 of 13)
- [06:49:02] Extracting A/V streams [VID_00702]
- [06:49:24] Reencoding video [VID_00702]
- Source Video: MPEG-4 (AVC), 1920x1080
- Rate/Length: 23.976fps, 3,241 frames
- [06:49:24] Reencoding: VID_00702, Pass 1 of 1
- [06:50:45] Video Encode complete
- [06:50:45] Processing audio tracks
- Track 4352 (eng): Keeping original audio
- [06:50:45] Multiplexing M2TS
- [06:50:56] Processing: VID_00701 (5 of 13)
- [06:50:56] Extracting A/V streams [VID_00701]
- [06:51:16] Reencoding video [VID_00701]
- Source Video: MPEG-4 (AVC), 1920x1080
- Rate/Length: 23.976fps, 3,625 frames
- [06:51:16] Reencoding: VID_00701, Pass 1 of 1
- [06:52:52] Video Encode complete
- [06:52:52] Processing audio tracks
- Track 4352 (eng): Keeping original audio
- [06:52:52] Multiplexing M2TS
- [06:53:02] Processing: VID_00704 (6 of 13)
- [06:53:02] Extracting A/V streams [VID_00704]
- [06:53:21] Reencoding video [VID_00704]
- Source Video: MPEG-4 (AVC), 1920x1080
- Rate/Length: 23.976fps, 3,673 frames
- [06:53:22] Reencoding: VID_00704, Pass 1 of 1
- [06:54:47] Video Encode complete
- [06:54:47] Processing audio tracks
- Track 4352 (eng): Keeping original audio
- [06:54:47] Multiplexing M2TS
- [06:54:57] Processing: VID_00700 (7 of 13)
- [06:54:57] Extracting A/V streams [VID_00700]
- [06:55:16] Reencoding video [VID_00700]
- Source Video: MPEG-4 (AVC), 1920x1080
- Rate/Length: 23.976fps, 3,841 frames
- [06:55:16] Reencoding: VID_00700, Pass 1 of 1
- [06:56:44] Video Encode complete
- [06:56:44] Processing audio tracks
- Track 4352 (eng): Keeping original audio
- [06:56:44] Multiplexing M2TS
- [06:56:54] Processing: VID_00454 (8 of 13)
- [06:56:54] Extracting A/V streams [VID_00454]
- [06:57:51] Reencoding video [VID_00454]
- Source Video: MPEG-4 (AVC), 1920x1080
- Rate/Length: 23.976fps, 6,596 frames
- [06:57:51] Reencoding: VID_00454, Pass 1 of 1
- [07:00:47] Video Encode complete
- [07:00:47] Processing audio tracks
- Track 4352 (eng): Keeping original audio
- [07:00:47] Multiplexing M2TS
- [07:01:00] Processing: VID_00451 (9 of 13)
- [07:01:00] Extracting A/V streams [VID_00451]
- [07:02:15] Reencoding video [VID_00451]
- Source Video: MPEG-4 (AVC), 1920x1080
- Rate/Length: 23.976fps, 7,825 frames
- [07:02:15] Reencoding: VID_00451, Pass 1 of 1
- [07:05:51] Video Encode complete
- [07:05:51] Processing audio tracks
- Track 4352 (eng): Keeping original audio
- [07:05:51] Multiplexing M2TS
- [07:06:11] Processing: VID_00453 (10 of 13)
- [07:06:11] Extracting A/V streams [VID_00453]
- [07:07:38] Reencoding video [VID_00453]
- Source Video: MPEG-4 (AVC), 1920x1080
- Rate/Length: 23.976fps, 8,970 frames
- [07:07:38] Reencoding: VID_00453, Pass 1 of 1
- Encode failed. Aborting.
- BD-Rebuilder v0.40.02 (beta)
- Windows Version: 6.1 [7601]
- AVISYNTH Version: 2.5.8.0, Ok
- HAALI Splitter: 1.9.42.1, Ok
- FFDSHOW: 3882, Ok
- WIN7 preferred AVC CODEC: Ok
- WIN7 preferred VC-1 CODEC: Ok
- WIN7 preferred MPEG2 CODEC: Ok
- FFDSHOW VC-1 set to "wmv9", Ok
- FFDSHOW MPEG2 set to "libavcodec": Ok
- FFDSHOW AVC set to "ffmpeg-mt": Ok
- X264: Ok
- AFTEN: Ok
- FAAC: Ok
- MP4BOX: Ok
- WAVI: Ok
- TSMUXER: Ok
[07:07:39] - Failed video encode, aborted

[jdobbs]

Well, I guess that's the minimum spec one needs to have in order to take advantage of MULTIPROCESS. I have 4 Core Processor (i5 2500K not overclocked) and 6Gb of RAM. I ran the same file (bluray 26Gb to bluray DVD5) three times. Without MULTIPROCESS, with MULTIPROCESS=2 and with MULTIPROCESS=3. The fastest was MULTIPROCESS=2 which finished just 2min quicker. The other two were identical in time. As I've said, the improvement with MULTIPROCESS will be approximately equal to the amount of unused processor utilization. Most quad cores max out and use 100% in pass two. So all you might get is a small increase in pass one that is balanced by the additional scan time.

But... what about your next processor and the ones after that? Ahhh...

[jdobbs]

@busch42

Can't say for sure -- but most of the time when it works its way through that many files and then fails deep into the process it is a sign of system instability (usually related to overheating). Are you overclocked?

The other possibility is a corrupt source. You might try ripping again. Also, sometimes a new high-volume release can throw a new angle at the decrypters and it might take a week or so for them to adjust. Of course that's just conjecture...

[Capsbackup]

I did a full backup to BD-25 of Drive, no problems with mine. ;)

[jdobbs]

jdobbs is there an undocumented setting for the size limit, or how bdrb calculates what are extras, when it runs quicker encode for extras?

For example, the way I have it setup, it does 1 pass crf 18 encodes on what it deems "extras" but does 2 pass high quality on what it defines the main feature. Currently it is running 2 passes on a 7gb extras video file which is 38 minutes, has 1 video 1 audio and 3 subtitle tracks.

Awesome work by the way, keep it up :) No. For the most part anything that is not a part of the main feature is considered an "extra". But there are occasional exceptions -- mainly when the M2TS is large (over ~5GB) or has ATCDelta set.

[RobertM]

As I've said, the improvement with MULTIPROCESS will be approximately equal to the amount of unused processor utilization. Most quad cores max out and use 100% in pass two. So all you might get is a small increase in pass one that is balanced by the additional scan time.


This agrees with my findings. My system is an i7-950 (4 core) with 6GB running at 3.6GHz. I see virtually no difference (<1%) between MULTIPROCESS=0 and MULTIPROCESS=1 in either Pass 1 or Pass 2.

With Hyper-Threading turned ON (so, 8 threads) I see a slight degradation in Pass 1 (1.5% slower) and a slight improvement in Pass 2 (5% faster).

So, on this system, it seems to be pretty much a wash.

[busch42]

jdobbs, Sorry to bother you but I think it had nothing to do with BDrebuilder, I tried DVDfab Bluray ripper and that didn't work either, Seems funny DVDfab let me rip to hard drive and it plays fine on computer but I can't shrink to a BD25, But I did just use VSO Bluray Converter and that worked fine, But it only does main movie and converts to a DVD5, Capsbackup, Can you tell me how you did it.

[busch42]

Oh, I'm not overclocked, And the movie Driver is cinavia protected.

[omegaman7]

Cinavia has nothing to do with it...

[busch42]

omegaman7, Have any suggestions because last week I backed up a bluray no problems. I'm using a AMD Phenom 9850 quad core processor 2.5GHz, 4GB memory, Windows7 64 bit.

[omegaman7]

I'll be running "Drive" today. I don't expect problems. Anydvd HD did update recently. If there was a new encryption, i'm sure they are aware of it ;) Cinavia is merely an audio watermark. It requires a specific hardware/firmware to detect it. At the PC level, it doesn't come into play.

I rarely use dvdfab. I've had more problems with it, than anydvd hd.

[busch42]

I might have to try Anydvd, Been using Fab for so many years with no problems but might have to switch. Looks like capsbackup had no problem.

[AmigaFuture]

Interesting...

Without MULTIPROCESS so I would know the speed. FPS is what I was used to seeing.

[16:14:51] BD Rebuilder v0.40.01 (beta)
- Source: JEFF_DUNHAM_SPARK
- Input BD size: 11.02 GB
- Approximate total content: [01:20:25.376]
- Windows Version: 6.1 [7601]
- MOVIE-ONLY/ALTERNATE OUTPUT mode enabled
- Mode: MKV Container, 1920x1080, 192Kbs AC3
- Automatic cropping of borders enabled
- Audio Settings: AC3=0 DTS=1 HD=0 Kbs=640
[16:14:54] PHASE ONE, Encoding
- [16:14:54] Processing: VID_00010 (1 of 1)
- [16:14:54] Extracting A/V streams [VID_00010]
- [16:17:46] Reencoding video [VID_00010]
- Source Video: MPEG-4 (AVC), 1920x1080
- Rate/Length: 29.970fps, 144,617 frames
- Bitrate: 7,448 Kbs
- [16:17:46] Reencoding: VID_00010, Pass 1 of 2
- [18:19:43] Reencoding: VID_00010, Pass 2 of 2
- [20:15:12] Video Encode complete
- [20:15:12] Processing audio tracks
- Track 4352 (eng): Reencoding audio to AC3...
- [20:15:49] Multiplexing M2TS
[20:15:49]PHASE ONE complete
[20:15:49]PHASE TWO - Rebuild Started
- [20:15:49] Rebuilding stream 00010 [1 of 1]
- [20:15:49] Building ALTERNATE OUTPUT Structure
[20:18:43] - Encode and Rebuild complete
[20:18:43] JOB: JEFF_DUNHAM_SPARK finished.


Same as before except with MULTIPROCESS=1 which you can really see the difference. Very noticeable higher FPS.

[13:13:47] BD Rebuilder v0.40.01 (beta)
- Source: JEFF_DUNHAM_SPARK
- Input BD size: 11.02 GB
- Approximate total content: [01:20:25.376]
- Windows Version: 6.1 [7601]
- MOVIE-ONLY/ALTERNATE OUTPUT mode enabled
- Mode: MKV Container, 1920x1080, 192Kbs AC3
- Automatic cropping of borders enabled
- Audio Settings: AC3=0 DTS=1 HD=0 Kbs=640
[13:13:54] PHASE ONE, Encoding
- [13:13:54] Processing: VID_00010 (1 of 1)
- [13:13:54] Extracting A/V streams [VID_00010]
- [13:16:50] Reencoding video [VID_00010]
- Source Video: MPEG-4 (AVC), 1920x1080
- Rate/Length: 29.970fps, 144,617 frames
- Bitrate: 7,448 Kbs
- [13:16:50] Reencoding: VID_00010, Pass 1 of 2
- [14:26:26] Reencoding: VID_00010, Pass 2 of 2
- [16:06:25] Video Encode complete
- [16:06:25] Processing audio tracks
- Track 4352 (eng): Reencoding audio to AC3...
- [16:07:08] Multiplexing M2TS
[16:07:08]PHASE ONE complete
[16:07:08]PHASE TWO - Rebuild Started
- [16:07:08] Rebuilding stream 00010 [1 of 1]
- [16:07:08] Building ALTERNATE OUTPUT Structure
[16:11:16] - Encode and Rebuild complete
[16:11:16] JOB: JEFF_DUNHAM_SPARK finished.

Around an hour difference. I used a timer on both, and the logs display what I got. Quality is set to Auto for these.

[LowDead]

I just say WOW.. Now we talking speed. I though I would do a test with multiprocessing. I set it to 1 so bdrb could decide what was best for my i7 980X with 24GB RAM. First off I thought my raided SATA6GB HDDs would totally die on me when doing 6-way.. FPS was jumping between 30 to 70 and was all over the place. Couldn't get stable usage of processor either as it jumped about to.. couldn't get 100% on first pass at all. Then I switched working directory to my revodrive (four raided SSDs on a PCIe card for those who don't know what it is), et voíla. FPS stable on 100. 100% on processor usage first pass. So it's not all about having a good processor ;-)

Oh, almost forgot.. Thank you for this great update JDOBBS, this will generate a donation from me later tonight :)

edit: ~22-27 FPS on second pass

[AmigaFuture]

So it's not all about having a good processor ;-)

Oh, almost forgot.. Thank you for this great update JDOBBS, this will generate a donation from me later tonight :)

I agree, totally. Setting the directories to different drives (not just partitions) really helps. I've known that since the Amiga days. Hehe. I'm donating tonight also. Not just for the speed increase, but more for Appreciation of your Passion we get to enjoy.

[jdobbs]

Interesting...

Without MULTIPROCESS so I would know the speed. FPS is what I was used to seeing.

...

Same as before except with MULTIPROCESS=1 which you can really see the difference. Very noticeable higher FPS.

...

Around an hour difference. I used a timer on both, and the logs display what I got. Quality is set to Auto for these. Great news. Thanks for the report.

[jdobbs]

I just say WOW.. Now we talking speed. I though I would do a test with multiprocessing. I set it to 1 so bdrb could decide what was best for my i7 980X with 24GB RAM. First off I thought my raided SATA6GB HDDs would totally die on me when doing 6-way.. FPS was jumping between 30 to 70 and was all over the place. Couldn't get stable usage of processor either as it jumped about to.. couldn't get 100% on first pass at all. Then I switched working directory to my revodrive (four raided SSDs on a PCIe card for those who don't know what it is), et voíla. FPS stable on 100. 100% on processor usage first pass. So it's not all about having a good processor ;-)

Oh, almost forgot.. Thank you for this great update JDOBBS, this will generate a donation from me later tonight :)

edit: ~22-27 FPS on second pass Cool... I was hoping to see some positive reports like these...

I can't wait until I grow up so I can get me one of those super bad-ass machines too. ;)

[steveg32]

@jdobbs

@steveg32

Is MKV_INTERMEDIATE=1 set in your INI file?

In HIDDENOPTS.TXT shows:

"MKV_INTERMEDIATE=n n = 0/1 - 0=OFF (default) - if set to "1", .MKV files are used instead of .264"

"MKV_INTERMEDIATE=" as shown in my previous attachments did not appear in the INI.

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

Btw, tried another movie with DTS Express Audio (RE_AFTERLIFE) and had the same type error.

"[03:44:26] BD Rebuilder v0.40.01 (beta)
- Source: RE_AFTERLIFE
- Input BD size: 34.02 GB
- Approximate total content: [03:16:08.996]
- Target BD size: 22.95 GB
- Windows Version: 6.1 [7601]
- Auto Quality: Good (Very Fast), ABR
- Audio Settings: AC3=0 DTS=0 HD=0 Kbs=640 ......

- [04:29:58] Processing: VID_00011 (35 of 35)
- [04:29:58] Extracting A/V streams [VID_00011]
- [04:45:34] Reencoding video [VID_00011]
- [04:45:34] Reencoding secondary video [TRK_02]
- Source Video: MPEG-4 (AVC), 1920x1080
- Rate/Length: 23.976fps, 139,368 frames
- Bitrate: 15,514 Kbs
- [04:49:37] Reencoding: VID_00011, Pass 1 of 1
- [05:32:57] Video Encode complete
- [05:32:57] Processing audio tracks
- Track 4352 (eng): Keeping original audio
- Track 4355 (eng): Keeping original audio
- Track 4357 (eng): Keeping original audio
- Track 6656 (eng): Keeping original audio
- [05:32:57] Multiplexing M2TS
- Error in attempt to multiplex: MUX_00011.meta
- Can't open file: C:\USERS\STEVE\VIDEOS\MY BD-REBUILDER ENCODES\MY BD-25_GOOD_1 PASS ABR\WORKFILES\VID_00011_2.AVS.MKV
[05:33:01] - Failed to build structure, aborted"

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

Mahalo, Steve :confused:

@jdobbs ... ooops, saw your post about the release of v0.40.02. Will download & retry again later. Aloha jdobbs - U Da Man!

[setarip_old]

@busch42

Hi! I might have to try Anydvd, Been using Fab for so many years with no problems but might have to switch. Looks like capsbackup had no problem.Just as a point of information, MakeMKV (presently free) also properly processed a Full Disc backup of "Drive" (a SONY release) with no difficulty...

[Racer]

What is actually the correct setting in ffdshow.
If I use libavcodec the avs script delivers to less frames for x264.
If I use ffmpeg-mt the avs script requires too much cpu load so that it slows down x264. On my AMD X4 libavcodec is faster but then I can not reach 100% cpu load.
How is it possible to improve this?

The speed (fps) can be measured very easy with AVSMeter.

[sonate]

JD, this is a general question. I notice that sometimes I will run an app which is memory intensive while I'm running BDRB, the frame counter will stop and remain stopped for maybe a minute or two after I close the application. Then it starts back up in normal fashion and runs to completion. Does this in any way create a problem with the encode? I haven't noticed any but I just wanted to ask.

Also, putting the computer into standby while running an encode, when awakened it resumes where it left off in the particular pass not needing to restart the pass over. Again, this should be OK too?

[jdobbs]

What is actually the correct setting in ffdshow.
If I use libavcodec the avs script delivers to less frames for x264.
If I use ffmpeg-mt the avs script requires too much cpu load so that it slows down x264. On my AMD X4 libavcodec is faster but then I can not reach 100% cpu load.
How is it possible to improve this?

The speed (fps) can be measured very easy with AVSMeter. I use libavcodec, but either is fine. Not sure how either would "delivers to less frames" -- they should deliver the same number of frames if read end-to-end sequentially (as is done in an encode). But -- since neither is frame accurate in its seeking, jumping around in a player (or AVISYNTH) will give inaccurate framecounts.

[jdobbs]

JD, this is a general question. I notice that sometimes I will run an app which is memory intensive while I'm running BDRB, the frame counter will stop and remain stopped for maybe a minute or two after I close the application. Then it starts back up in normal fashion and runs to completion. Does this in any way create a problem with the encode? I haven't noticed any but I just wanted to ask.

Also, putting the computer into standby while running an encode, when awakened it resumes where it left off in the particular pass not needing to restart the pass over. Again, this should be OK too? Running something at the time of encoding shouldn't matter.

But I wouldn't put a lot of trust in an encode that resumes after standby mode, however. There are a lot of variables at play there. You're putting your trust in a perfect resume and a belief the Windows never makes a mistake in doing so... and I'm personally not willing to make that assumption.

[Ch3vr0n]

Quick question, im confused about something. If i have both LAVF enabled and multiprocessing. What should happen (eg wich method should take priority). LAVF or multiprocessing

[busch42]

Thanks, setarip old, I will have to check that out and see how it works, I just tried BDrebuilder again and it came up x264.exe stopped working, So I uninstalled everything and reinstalled all programs and now Im giving it 1 more shot, Don't know why I'm having such a hard time.

[jdobbs]

Quick question, im confused about something. If i have both LAVF enabled and multiprocessing. What should happen (eg wich method should take priority). LAVF or multiprocessing LAVF will take priority.

[busch42]

Still having a problem, After uninstalling everything and reinstalling I got the movie to go from 16% before to go 31% now, But still comes up x264.exe has stopped working. Any suggestions anybody?

[jdobbs]

Still having a problem, After uninstalling everything and reinstalling I got the movie to go from 16% before to go 31% now, But still comes up x264.exe has stopped working. Any suggestions anybody? Does it always stop in the same place? If so -- it is almost definitely the source.

Right after it stops, copy the command from LASTCMD.TXT (in the BD-RB folder) into a command window and run it. That way you will be able to see what X264 is saying when it crashes.

[Capsbackup]

Still having a problem, After uninstalling everything and reinstalling I got the movie to go from 16% before to go 31% now, But still comes up x264.exe has stopped working. Any suggestions anybody?

If you are not over clocked, then have you checked and cleaned out all dust from your cpu fan and case? Are your fans working? X264 is very processor intense, more so than most people realize. ;)

[rippn]

I'm having trouble with it stopping too. Same point 3 attempts. Went back to v0.39.7 and no issue.

Win7 X64 no overclock.

[jdobbs]

I'm having trouble with it stopping too. Same point 3 attempts. Went back to v0.39.7 and no issue.

Win7 X64 no overclock. I need a lot more information that this. What disc? A log would help. Stopping how? What point? Where in the process?

I'm willing to work on any problem -- but first I have to have a clue as to what it is...

[Ch3vr0n]

Here's detailed comparison from a disc run with 40.02; Disc: Friends With Benefits

full backup (non-blanking) multiprocess mode, intact audio
auto-quality: Good, 1 pass
main m2ts: VID_0001; 27.619,68MB AVC
job start: 22:37:43
encode start: 22:44:04
n° X264's: 2 (0.40.01 launched 3)
cpu usage: 100%
ram usage: 4% into build 3.75GB / 80% into build 4.25-4.45GB
FPS count: ~36FPS
encode finish: 00:10:16
encode time: 1h26min
job end: 00:59:47 (not counting iso burning)
job time: 2h22min
output size: 23.338.112kb; 22.2GB

smaller video files utilised mostly 100% cpu with 1 x264 and arround 3.25GB of ram @ ~40FPS


full backup (non-blanking) LAVF mode, intact audio
auto quality: Good, 1 pass
Main m2ts: same one
Job start: 18:35:01
encode start: 18:41:46
n° x264's:: 1x x264-64
cpu usage: 100%
ram usage: 4% into build 2.91GB / 80% into build 2.89GB
FPS count: ~36FPS
encode finish: 19:55:32
encode time: 1h14min
job end: 20:42:18
job time: 2h07min
output size: 22.539.520kb (slightly smaller); 21.4GB

Total comparison: MP vs LAFV = MP = +15minutes (probably the time it took to split the main m2ts)
will try a different codec next

[busch42]

Wiped all junk off hard drive, Reinstalled all required programs still no luck, But I did manage to extract the main movie with tsMuxeR and save it as a bluray, But it is still 28GB. So Tried again with BDrebuilder to shrink but still comes up x264.exe has stopped working, After trying for 2 days it is getting frustrating, I even tried a program that was recommened called Blue Cloner, But everytime I hit the start button it would come up (failed attempt). I'm now trying again after going into setup and enabled Use X264's internal LAVF for decoding.

[jdobbs]

Wiped all junk off hard drive, Reinstalled all required programs still no luck, But I did manage to extract the main movie with tsMuxeR and save it as a bluray, But it is still 28GB. So Tried again with BDrebuilder to shrink but still comes up x264.exe has stopped working, After trying for 2 days it is getting frustrating, I even tried a program that was recommened called Blue Cloner, But everytime I hit the start button it would come up (failed attempt). I'm now trying again after going into setup and enabled Use X264's internal LAVF for decoding. There was a new version of X264 included with this release. Just for testing, try replacing it with the one from v0.39.07 and see if that makes a difference.

[busch42]

Thanks jdobbs, I will try that. I just got 37% done after I enabled Use X264's internal LAVF for decoding, but then it came up x264.exe stopped working again. Before I could only get about 14% done. But I did notice on the last version of BDrebuilder that when you clicked on settings there was A word there called Encoder that you could click on and make sure it was set to x264, But I don't see it on this version.

[jdobbs]

Thanks jdobbs, I will try that. I just got 37% done after I enabled Use X264's internal LAVF for decoding, but then it came up x264.exe stopped working again. Before I could only get about 14% done. But I did notice on the last version of BDrebuilder that when you clicked on settings there was A word there called Encoder that you could click on and make sure it was set to x264, But I don't see it on this version. It never did anything, and it's unlikely that I'll ever add an alternate encoder -- so I removed it.

[AmigaFuture]

Cool... I was hoping to see some positive reports like these...

I can't wait until I grow up so I can get me one of those super bad-ass machines too. ;)

You don't have to grow "up" to have a "super computer", Friend. The parts can be gathered from friends, family, job(s) and others' you Know and Help. Law of Attraction is a powerful Friend, the ~Best~. :-)

[Sharc]

@jdobbs
Source is Blu-ray 25fps MBAFF.
I resize it to 720 for a BD5, encode using CRF prediction, and it comes out oversized as BD9. (CRF final was 12.8 which is already an indication of oversizing).
BD-RB introduces frame duplication for 50fps for blu-ray compliance. Is this duplication possibly the reason for oversizing? Has anything been changed with respect to earlier version (I am pretty sure that the sizing came out right in earlier versions)?

[RobertM]

Just got a "PredictAndEncode" error when trying to backup "Skyline" with "MULTIPROCESS=1" set. The error happens immediately after extracting the A/V streams. I tried several times, re-extracting the data each time, with the same result. Then I set "MULTIPROCESS=0" and Pass 1 is now progressing normally.

BTW; I've done several successful backups with "MULTIPROCESS=1" set.

Here's the log:

[17:11:23] BD Rebuilder v0.40.01 (beta)
- Source: SKYLINE
- Input BD size: 25.61 GB
- Approximate total content: [01:32:46.394]
- Target BD size: 7.84 GB
- Windows Version: 6.1 [7601]
- MOVIE-ONLY mode enabled
- Quality: High Quality (Default), Two Pass
- Audio Settings: AC3=1 DTS=1 HD=0 Kbs=640
[17:11:26] PHASE ONE, Encoding
- [17:11:26] Processing: VID_00004 (1 of 1)
- [17:11:26] Extracting A/V streams [VID_00004]
- [17:14:54] Reencoding video [VID_00004]
- Source Video: MPEG-4 (AVC), 1920x1080
- Rate/Length: 23.976fps, 133,460 frames
- Bitrate: 10,517 Kbs
- [17:14:54] Reencoding: VID_00004, Pass 1 of 2
- [17:14:54] PredictAndEncode() 00006 2809
[17:15:55] - Aborted by user request