BD Rebuilder Beta - Bug Reports Only

[rusty666]

should i be useing x264 lavf encodeing?

[jdobbs]

That really doesn't matter.

[rusty666]

will you be doing 3d full menu bd25 at some point?

[jdobbs]

Yes. I have a few things to figure out first, though.

[DoctorM]

Hello, this is my first post.

I have been using BD Rebuilder for awhile now, and I seem to be having an issue that just started with the past couple of versions that I've used (I'm not sure if these correspond to the last 2 releases, as I may not have installed each release that was made). Now, the video will always be re-encoded, even when it will fit on the target (BD25). This used to work fine for me (prior to the past 2 versions that I've used). I've done several movies where converting the LPCM or DTS HD-Master to AC3 allowed the size to be small enough for the video to not be re-encoded. Now, this is not occurring. In fact, I have a movie now that I just tried. It's "The Artist" and it's 22.9 GB with the DTS HD-Master audio track. It's just barely above the BD25 threshold, and clearly by converting the DTS HD-Master to AC3, it would be below the BD25 threshold, but BD Rebuilder is still is trying to re-encode the video. Another example, I had A Clockwork Orange, with a few audio tracks (1 LPCM, and 2 AC3, I believe). I selected just 1 AC3 track (which ended up bringing the final size to well under 20 GB, but it still tried to re-encode the video. If I change the target size to BD50, it works (no re-encode), so I don't think that any of my other settings are forcing it to re-encode or anything in the video itself (interlaced, etc.). Did something change with the size calculations? Again, this just started happening with the past 2 versions that I've used (prior versions seemed to be OK). Any thoughts? Here are my settings and log(s).

---------
SETTINGS
---------
[Options]
VERSION=0.46.0.13
MODE=0
ENCODE_QUALITY=3
ONEPASS_ENCODING=0
AUTO_QUALITY=0
TARGET_SIZE=23500
AUTO_BURN=2
AUDIO_TO_KEEP=eng;
SUBS_TO_KEEP=all
SD_CONVERT=0
OPEN_GOP=0
RESIZE_1080=0
RESIZE_1440=0
RESIZE_720=0
DEINTERLACE=1
SD_TO_1080=0
CONVERT_WIDE=0
DTS_REENCODE=1
AC3_REENCODE=0
AC3_640=1
AC3_192=0
KEEP_HD_AUDIO=0
KEEP_HD_LPCM=0
AVCHD=1
REMOVE_WORKFILES=0
MOVIE_ONLY_LOOP=1
REMOVE_OUTPUT=0
USE_FILTERS=0
BDMV_CERT_ONLY=0
USE_LAVF=0
IVTC_PULLDOWN=0
ASSUME_DVD_PAL=0
UNMASK_CHAPTER=0
COMPLETION_BEEP=0
DGDECNV=0
OUTPUT_3D=0
NEROAAC=0
SUPTITLE=0
AUDIO_TRACK_LIMIT=1
SUBTITLE_TRACK_LIMIT=1
CUSTOM_TARGET_SIZE=23500
IGNORE_3D=0
OUTPUT_SBS=0
[Paths]
SOURCE_PATH=H:\
WORKING_PATH=D:\TEMP\BD-RB\

---------------------------
LOG (with BD25 target size)
---------------------------
[03/02/14] BD Rebuilder v0.46.13 (beta)
[11:22:31] Source: THE_ARTIST
- Input BD size: 22.99 GB
- Approximate total content: [01:40:36.071]
- Target BD size: 22.95 GB
- Windows Version: 6.1 [7601]
- Quality: Highest (Very Slow), Two Pass
- Decoding/Frame serving: DirectShow
- Audio Settings: AC3=0 DTS=1 HD=0 Kbs=640
[11:22:32] PHASE ONE, Encoding
- [11:22:32] Processing: VID_00015 (1 of 1)
- [11:22:32] Extracting A/V streams [VID_00015]
- [11:35:53] Reencoding video [VID_00015]
- Source Video: MPEG-4 (AVC), 1920x1080
- Rate/Length: 23.976fps, 144,721 frames
- Bitrate: 29,643 Kbs
- [11:35:53] Reencoding: VID_00015, Pass 1 of 2
[11:37:28]PHASE ONE aborted by user request

---------------------------
LOG (with BD50 target size)
---------------------------
[03/02/14] BD Rebuilder v0.46.13 (beta)
[11:38:57] Source: THE_ARTIST
- Input BD size: 22.99 GB
- Approximate total content: [01:40:36.071]
- Target BD size: 46.26 GB
- Windows Version: 6.1 [7601]
- Quality: Highest (Very Slow), Two Pass
- Decoding/Frame serving: DirectShow
- Audio Settings: AC3=0 DTS=1 HD=0 Kbs=640
[11:38:58] PHASE ONE, Encoding
- [11:38:58] Processing: VID_00015 (1 of 1)
- [11:38:58] Extracting A/V streams [VID_00015]
- [11:52:05] Reencoding video [VID_00015]
- [11:52:05] Keeping original video (no reencode)
- [11:52:05] Processing audio tracks
- Track 4352 (eng): Reencoding audio to AC3...

Thanks for any help. Sorry for the long post.
-Peter

Check your settings, you may have something like resize, avisynth filters turned on or something else (although from your logs it doesn't look that way).

That said, I have the exact same problem.

[jdobbs]

I have updated the first post of this thread with a link to the latest version of BD-RB (v0.46.14). Changes for this release:- Corrected an issue that could jumble menu
order and/or edited titles when importing
from BD sources.
- Modified LAVF processing so it now uses the
original M2TS file for encoding rather than
the extracted video file. This should reduce
the required working folder free disc space,
may slightly improve speed, and streamline
the processing to match other modes.
- Added a new hidden option. Some series DVDs
use chapters to separate episodes (especially
those with short episodes) rather than PGCs
(playlists). Setting IMPORT_DVD_CHAPS=1 tells
BD-RB to break out chapters as playlists
during IMPORT so the individual episodes can
be accessed in the IMPORT menu.
- Fixed an issue in which (under some very odd
and exceptionally rare circumstances) it may
be possible that the Quick-Play menu will be
selected unnecessarily for encoding in a full
backup, or even be selected as the feature in
a movie-only backup.
- Updated logic so FORCE_NOENCODE will apply to
BD-5/9 sources as well as BD-25/50 as long as
no other user setting (such as resizing, ivtc,
filters, etc.) overrides it.
- Other minor corrections and cosmetic fixes.

[inge70]

THX for Great Job :)

[Sharc]

Thank you jdobbs.
Very happy to see the Quick-Menu issues fixed :)

[raul124]

Amazin, just Amazin, your the MAN jdobbs!!

[Peter_A]

Check your settings, you may have something like resize, avisynth filters turned on or something else (although from your logs it doesn't look that way).

That said, I have the exact same problem.

Thanks, DoctorM, for the response. But if my settings were the issue, then wouldn't it also try to re-encode when I select BD50 (and change nothing else)? It does not re-encode with BD50 target but does with BD25 target (even when the final size would be well under BD25 max size). I'm fairly sure that something has changed with the size calculations, as I did not have this issue in the past. I've just noticed it with the past 2 versions that I've used. I was hoping that jdobbs would have some insight into the issue.

I will try the new version (0.46.14), but I didn't see anything in the notes that would indicate any change related to the issue that I'm experiencing.

[soneca]

Thanks for the new version! ;)

[rusty666]

can anyone help if i import a mkv sbs 3d it says failed video encode?

[Peter_A]

Thanks, DoctorM, for the response. But if my settings were the issue, then wouldn't it also try to re-encode when I select BD50 (and change nothing else)? It does not re-encode with BD50 target but does with BD25 target (even when the final size would be well under BD25 max size). I'm fairly sure that something has changed with the size calculations, as I did not have this issue in the past. I've just noticed it with the past 2 versions that I've used. I was hoping that jdobbs would have some insight into the issue.

I will try the new version (0.46.14), but I didn't see anything in the notes that would indicate any change related to the issue that I'm experiencing.

So, I tried 0.46.14, and I am still seeing the same behavior. The BD's .INF file contents are shown below. In the INF file, SOURCE_VIDEO_SIZE is just barely smaller than SOURCE_SIZE, even though there is an HD Audio track. Does SOURCE_VIDEO_SIZE include the audio or just the video? SOURCE_VIDEO_SIZE is just larger than TARGET_SIZE, which is why I'm guessing that the video re-encode is being performed. Is that the problem? Is BD Rebuilder not accounting for the reduced space that the audio track will provide? If that's the case, fine. But again, this used to work OK for me, so I am inclined to believe that something changed regarding the calculated size. Increasing the target size to allow the video re-encode to skip (and DTS HD to be converted to AC3) results in a 20.9GB file (original file is 22.9GB).

BTW, I just checked some archived log/inf files, and this appears to have been working correctly for me, in version 0.45.05. I also see that SOURCE_SIZE and SOURCE_VIDEO_SIZE are almost identical in the 0.45.05 logs (so that may not be part of the issue), and with 2 discs (REDUCTION=.798882164854814 and REDUCTION=.899547732616703), a video re-encode was not performed (target = BD25).

[Status]
LABEL=THE_ARTIST
VERSION=v0.46.14 (beta)
SOURCE_SIZE=24688800458
SOURCE_VIDEO_SIZE=24688613376
TARGET_SIZE=24641536000
REDUCTION=.998085576647008
RESIZE_1080=0
RESIZE_1440=0
AUDIO_TO_KEEP=eng;
KEEP_HD_AUDIO=0
SUBS_TO_KEEP=all
BACKUP_MODE=0
MOVIEONLY_TYPE=0
USE_LAVF=0
INSTANCES=1
DGDECNV=0
SSIF_MODE=0
QUICK=0
[00015]
AUDIO=1
PGS=1

[jdobbs]

The reduction could come from video or audio. But if there is a reduction indicated -- it has to come from somewhere. If the older version didn't encode while the new one did -- I'd wager the settings were different. There hasn't been any changes in the algorithm in a long time.

You're looking at things that are meant for BD-RB's use, not the end-user, and you are drawing incorrect conclusions. There are two checks to see if the disc will fit.

1. At the overall level. If the reduction percentage indicates no need to encode, then none is done. That obviously isn't the case in your example because there is a required reduction.

2. At the individual M2TS level. After extraction, the end-size for the PGS and audio is calculated based upon your settings. Then it is determined whether the video needs to get smaller. If not, it is kept intact as long as it isn't a part of a multipart playlist. In that case, it cannot be kept intact because that could result in a mismatch of quality/codecs -- and since step 1 told us there is a required reduction we know not to keep it.

[Sharc]

In Quick-Play Reauthoring mode the secondary video gets included (and encoded), while when importing the same BD the secondary video is skipped. Is this different behaviour intended, or do I miss something?
(Quick menu threshold settings are the same for both cases, if this should matter).

[Peter_A]

The reduction could come from video or audio. But if there is a reduction indicated -- it has to come from somewhere. If the older version didn't encode while the new one did -- I'd wager the settings were different. There hasn't been any changes in the algorithm in a long time.

That's very interesting, and thanks for the response. I agree that the reduction needs to come from somewhere. In this case, less than 0.2% reduction is needed, and the audio is being changed from DTS HD Master to AC3. So, the algorithm does not seem to be accounting for an audio reduction at all and re-encodes the video. Is that the way that it's supposed to work? If I bump the target size up slightly so that no reduction is needed, it does not re-encode the video, so I don't think that any of my other settings are to blame.

[jdobbs]

Sorry, but I trust BD-RB's calculations more than yours. It very definitely accounts for the audio. I'm guessing you are failing to account for muxing overhead -- which in some codecs can be as much as 30% of the audio size (and is almost never less than 14%).

There is also the inaccuracies of encoding and muxing that has to be taken into account. No one wants to spend three hours encoding something only to find that it won't fit.

If you want to take that chance, it's up to you and you can change the target sizes. But it isn't going to happen at the default level.

[jdobbs]

In Quick-Play Reauthoring mode the secondary video gets included (and encoded), while when importing the same BD the secondary video is skipped. Is this different behaviour intended, or do I miss something?
(Quick menu threshold settings are the same for both cases, if this should matter).Yes. That is intended. You'll also notice that importing turns a multipart playlist into a single M2TS, while Quick-Play retains the original MPLS structure.

While you can certainly import a single disc, the more-focused intent of IMPORT is to combine multiple discs, generally series discs or trilogies & the like.

[Sharc]

I see. Thank you for the explanation.

[DarWun]

A little over a month a go I posted about an issue I was having with FRIM failing during the rebuild of 3D titles if I used the computer to do anything else. I was not able to resolve the issue before real world stuff took over and left me without my computer until a few days ago. Since getting up and running again, I have done a few 3D title movie-only rebuilds to BD25. I am happy to report that rebuild no longer fails when it is running in the background and I am running other CPU intensive applications.

The only thing that is different with my PC since I first reported the issue is that it now sits in Toronto instead of Calgary. In that time three new versions of BD-Rebuilder were released. So I'm going to assume that something changed with either the version of FRIM being used or the way BD-Rebuilder works when running in the background and attribute it to jdobbs great work.;)

[Peter_A]

Sorry, but I trust BD-RB's calculations more than yours. It very definitely accounts for the audio. I'm guessing you are failing to account for muxing overhead -- which in some codecs can be as much as 30% of the audio size (and is almost never less than 14%).

There is also the inaccuracies of encoding and muxing that has to be taken into account. No one wants to spend three hours encoding something only to find that it won't fit.

If you want to take that chance, it's up to you and you can change the target sizes. But it isn't going to happen at the default level.

Well, with all due respect, at this point, I'm not sure that I trust BD-RB's calculations more than mine. The original ISO file is 23,547 MB. BD-RB is set for 23,500 MB, which means a reduction of only 47 MB is needed. So, the algorithm employed cannot determine that changing from DTS HD to AC3 will save more than 47 MB? If that's the case, I posit that there is an issue with the algorithm. In fact, it saved over 2 GB. The final ISO is 21,422 MB (more than 2 GB lower than the BD-RB setting of 23,500 MB). Bumping the target size up slightly makes it perform correctly, i.e. not re-encode the video.

You say that no one wants to encode for 3 hours to end up with a file that's too big to write. I agree. I would also say that no one wants to spend 3 hours re-encoding video when that step is completely unnecessary.

Edit: By the way, the 21,422 MB final ISO that I referenced above, was after I bumped up the target size and no video re-encode was done. So, the 2 GB+ reduction was entirely from the audio (DTS HD -> AC3), yet the algorithm was unable to determine that at least 47 MB would be saved.

[DarWun]

I have BD-Rebuilder set to delete working and output folders after the rebuild is complete. On a recent 3D movie only backup with BDRB v46.13 I accidentally left the "Burn to ISO" option enabled in the output dropdown menu. The iso was created at the end of the build after which the working and output folders were deleted as expected. Then BDRebuilder attempted to create the iso a second time, which it could not do since the output folder no longer existed.

[Ch3vr0n]

because they got deleted after the iso was created, normal behavior. You told it to delete working & output folders. Which it did and it doesn't mean if you left it on accidentaly. 3D backsup HAVE to be in iso due to the way the disc layout is.

[jdobbs]

Well, with all due respect, at this point, I'm not sure that I trust BD-RB's calculations more than mine. The original ISO file is 23,547 MB. BD-RB is set for 23,500 MB, which means a reduction of only 47 MB is needed. So, the algorithm employed cannot determine that changing from DTS HD to AC3 will save more than 47 MB? If that's the case, I posit that there is an issue with the algorithm. In fact, it saved over 2 GB. The final ISO is 21,422 MB (more than 2 GB lower than the BD-RB setting of 23,500 MB). Bumping the target size up slightly makes it perform correctly, i.e. not re-encode the video.

You say that no one wants to encode for 3 hours to end up with a file that's too big to write. I agree. I would also say that no one wants to spend 3 hours re-encoding video when that step is completely unnecessary.

Edit: By the way, the 21,422 MB final ISO that I referenced above, was after I bumped up the target size and no video re-encode was done. So, the 2 GB+ reduction was entirely from the audio (DTS HD -> AC3), yet the algorithm was unable to determine that at least 47 MB would be saved.Well with all due respect in return... since I am the author of the software, the calculation to be used will be decided by me.

I've answered your questions courteously. I've looked at the calculations very carefully -- and I'm not going to argue this point, no matter how many times you repeat your posts.

What I will do, when I find the time, is run "The Artist" and follow the code and see why it decides that reencoding is necessary. But I'm confident the reasoning will follow the rules I've already stated numerous times in the past several pages.

[jdobbs]

because they got deleted after the iso was created, normal behavior. You told it to delete working & output folders. Which it did and it doesn't mean if you left it on accidentaly. 3D backsup HAVE to be in iso due to the way the disc layout is.Just a little teaser... I've studied the spec and done some testing... and actually you don't even have to create the SSIF files at all -- and that's the only reason the ISO output is required. That is just one way to do 3D within the spec. There are others...

You'll see some changes, as well as possible full disc 3D backup in the next release (depending upon how far along I get and the results of more testing).

[Ch3vr0n]

if u pull that one off, you'd make me a VERY happy camper and no matter how tight my budget it is i WILL find the cash to do another donation then. Just a question though, if there's other ways to do it besides "SSIF" and the crappy "anaglyph". Why do studios always pick ssif then. Will standalone be able to play your 'alternate' method?

[DarWun]

because they got deleted after the iso was created, normal behavior. You told it to delete working & output folders. Which it did and it doesn't mean if you left it on accidentaly. 3D backsup HAVE to be in iso due to the way the disc layout is.

I realize the iso is required for 3D rebuilds. But I would have thought BDRebuilder would do a check to see if the iso is already being created before going and creating another one. It's not a bug per se. Just something I found odd.

[jdobbs]

if u pull that one off, you'd make me a VERY happy camper and no matter how tight my budget it is i WILL find the cash to do another donation then. Just a question though, if there's other ways to do it besides "SSIF" and the crappy "anaglyph". Why do studios always pick ssif then. Will standalone be able to play your 'alternate' method?:D I've already "pulled it off" -- I've done several tests of 3D encodes without SSIF files, and it works on all three of the standalone players I've tried (one of them was a 2D player and it played back in 2D). Of course all the other rules for 3D on BD apply... you still have to have a 3D enabled player and a 3D enabled monitor.

Generally it's easier to author using SSIF, especially when you are planning to create an image anyway. But, there are some discs that use a single M2TS to hold 3D content (without a SSIF) -- example: open up AVATAR with BDEDIT and look a some of the 3D content. Honestly it's not some big dark secret (since it's documented within the BD 3D spec), it's just really useful when you consider the possibility of full backups.

But, TSMUXER doesn't support what you see on AVATAE... but it does support creating two M2TS files without SSIF -- and then only the MPLS and CLPI files have to be modified to make it work in 3D without SSIF.

With that said, you never know what might come along when applying this concept to full backups, and all my tests have been with short subject encodes... so I'll not make any guarantees yet.

[jdobbs]

I have BD-Rebuilder set to delete working and output folders after the rebuild is complete. On a recent 3D movie only backup with BDRB v46.13 I accidentally left the "Burn to ISO" option enabled in the output dropdown menu. The iso was created at the end of the build after which the working and output folders were deleted as expected. Then BDRebuilder attempted to create the iso a second time, which it could not do since the output folder no longer existed.That's definitely a bug. I'll fix it.

[DoctorM]

I remember reading back with someone reporting that during extraction the audio track was blank. I think it was a DTS issue(?)
Anyway, impossible to search this thread with any keywords that doesn't return just about every post with a log.

Can anyone tell me what the issue was? I just saw this happen today and was trying to figure out a work around.

[Peter_A]

Well with all due respect in return... since I am the author of the software, the calculation to be used will be decided by me.

I've answered your questions courteously. I've looked at the calculations very carefully -- and I'm not going to argue this point, no matter how many times you repeat your posts.

What I will do, when I find the time, is run "The Artist" and follow the code and see why it decides that reencoding is necessary. But I'm confident the reasoning will follow the rules I've already stated numerous times in the past several pages.

Thanks for your response, and I do appreciate your time. It's great to be able to have a dialogue with the person who is developing the software.

I'm just a bit frustrated, because this was working for me before, and I cannot figure out why it is not now, especially since you've confirmed that nothing has changed with respect to these calculations. The last 3 titles that I have tried are exhibiting this behavior. I only mentioned "The Artist" due to the very small amount that it needed to be shrunk, and the large amount of space saved by the audio re-encode. Of course, by modifying the target size, I have found a way around this behavior.

Maybe you'll be able to see something, and if so, I look forward to hearing about it. Thanks again.

[Ch3vr0n]

that was for the pip stream relying on dts-express, not a 3D issue. That has been fixed recently with tsmuxer being updated to handle dts-e

[mparade]

@jdobbs

- [21:49:39] Processing: VID_00066 (55 of 102)
- [21:49:39] Extracting A/V streams [VID_00066]
- Error in attempt to MKVMERGE.
[21:50:21] - Failed to retrieve audio, aborted

If I remember correctly, I could make a back up of this BD (using BD-RB version 0.45.05) with which I am currently having a problem like above. Unfortunately, I could not test it due to an expiration message of the older version. I hoped I could bridge through the "problem" I have with only one BD from my BD stock. :(

[Peter_A]

@jdobbs

- [21:49:39] Processing: VID_00066 (55 of 102)
- [21:49:39] Extracting A/V streams [VID_00066]
- Error in attempt to MKVMERGE.
[21:50:21] - Failed to retrieve audio, aborted

If I remember correctly, I could make a back up of this BD (using BD-RB version 0.45.05) with which I am currently having a problem like above. Unfortunately, I could not test it due to an expiration message of the older version. I hoped I could bridge through the "problem" I have with only one BD from my BD stock. :(


I also just tried to use 0.45.05, today, to do a sanity check and confirm that what I am trying to do now worked then. I also received the expiration message and was therefore unable to test.

[jdobbs]

I also just tried to use 0.45.05, today, to do a sanity check and confirm that what I am trying to do now worked then. I also received the expiration message and was therefore unable to test.There have been about 58 improvements and bug fixes since v0.45.05. Going back in time isn't an option, so what good would that do anyway? I also can't really help without enough information to make a logical conclusion. I don't have the name of the disc, I don't have the log, I don't have the INI file, I don't even have the circumstances surrounding the problem. With those conditions the best I can do is throw down a bunch of chicken bones and hope the pattern somehow leads me the problem. This applies to both these posts.

Things can't get fixed with posts that just say "...it doesn't work"

[Peter_A]

I can't really help without enough information to make a logical conclusion. I don't have the name of the disc, I don't have the log, I don't have the INI file, I don't even have the circumstances surrounding the problem. With those conditions the best I can do is throw down a bunch of chicken bones and hope the pattern leads me the problem. This applies to both these posts.

Things can't get fixed with posts that just say "...it doesn't work"

Sorry, jdobbs, I was referring to the video re-encode issue which we discussed earlier and I had stated appeared to be working for me in version 0.45.05. I wanted to try that version again, to see if that version did work with the discs I am experiencing issues with now. Unfortunately, due to the expiration, I was unable to test.

[mparade]

There have been about 58 improvements and bug fixes since v0.45.05. Going back in time isn't an option, so what good would that do anyway? I also can't really help without enough information to make a logical conclusion. I don't have the name of the disc, I don't have the log, I don't have the INI file, I don't even have the circumstances surrounding the problem. With those conditions the best I can do is throw down a bunch of chicken bones and hope the pattern somehow leads me the problem. This applies to both these posts.

Things can't get fixed with posts that just say "...it doesn't work"

Sorry for that. Attached now the files required.
I have tried internal LAVF for decoding also with the same results, unfortunately.

[Sharc]

I have Extras which are reported by BD-RB as 1080i,29.97fps.
I resize to 1440x1080, and the script becomes accordingly
#Created by BD Rebuilder - v0.46.14 (beta)
LoadPlugin("C:\Program Files VIDEO\DGdecNV\DGDecodeNV.dll")
DGSource("C:\TEMP\BD REBUILDER\WORKFILES\VID_01054.DGI", fieldop=0)
SeparateFields()
BicubicResize(1440,540)
Weave()
ConvertToYV12().AssumeFPS(30000,1001)
It gets encoded as interlaced (MBAFF), but with ugly blended frames with visible combing.
I tried various settings, but as soon as I resize to 1440x1080 I get the blended frames. When I don't resize I am getting a progressive 3:2 pattern at 29.97 without blended frames but with the 3:2 judder.

Now I found that the source is actually not interlaced, but telecined.
The script
TFM().tdecimate() seems to produce perfect progressive frames with no blends at 23.976 fps, which can be resized to 1440x1080.

Is there an option or a setting in DB-RB which I possibly missed?

[jdobbs]

Turn on deinterlacing.

You'll also find that most of the time you'll find that the combing effects are only visible on a computer, but aren't there when you play back using a standalone and a monitor.

[DoctorM]

@Peter_A - Turn the clock back on your PC to Feb of this year and you can test 45.05 (it's what I'm doing right now).
I do agree, I've seen that recent builds calculate: Full BD with all audio / Target Size = Compression Ratio.
No deduction are made for stripped or compressed tracks.

It my be a per PC issue of some sort, otherwise a lot of people should be reporting undersized discs.

[Sharc]

Turn on deinterlacing.

You'll also find that most of the time you'll find that the combing effects are only visible on a computer, but aren't there when you play back using a standalone and a monitor.
Yep, you are right.
Actually I made 2 mistakes :o :o :
1) When leaving it as interlaced and watching the result on PC, I had the deinterlacer of the SW player disabled, hence the blended/combed frames. The TV applies its deinterlacer on material which is flagged as interlaced, and the result looks good on TV. The TV seems to bob to 60fps.
2) When I selected "Use DEINTERLACER on interlaced sources" I had -- from a previous test -- "DEINTERLACER_TYPE=4" in the .ini which removed the deinterlacer from the script; therefore the ugly result. Leaving the "DEINTERLACER TYPE= " alone I get a "3+2blended" same rate (29.97fps) pattern which also looks ok on TV.
Eventually I tried "DEINTERLACER_TYPE=3" (smoothdeinterlacer) which produced a "3+2repeats" same rate (29.97fps) pattern which also looks decent on TV.

At the end, 1) i.e. leaving it interlaced and let the TV take care of the rest looks best for me.

(I found that the "IVTC sources with 3:2 pulldown" had no effect on the script)

I understand that restoring a 23.976fps progressive stream with "telecide().tdecimate()" or similar changes the framrate which may have implications on audio sync and pitch (?).

[jdobbs]

Yep, you are right.
Actually I made 2 mistakes :o :o :
1) When leaving it as interlaced and watching the result on PC, I had the deinterlacer of the SW player disabled, hence the blended/combed frames. The TV applies its deinterlacer on material which is flagged as interlaced, and the result looks good on TV. The TV seems to bob to 60fps.
2) When I selected "Use DEINTERLACER on interlaced sources" I had -- from a previous test -- "DEINTERLACER_TYPE=4" in the .ini which removed the deinterlacer from the script; therefore the ugly result. Leaving the "DEINTERLACER TYPE= " alone I get a "3+2blended" same rate (29.97fps) pattern which also looks ok on TV.
Eventually I tried "DEINTERLACER_TYPE=3" (smoothdeinterlacer) which produced a "3+2repeats" same rate (29.97fps) pattern which also looks decent on TV.

At the end, 1) i.e. leaving it interlaced and let the TV take care of the rest looks best for me.

(I found that the "IVTC sources with 3:2 pulldown" had no effect on the script)

I understand that restoring a 23.976fps progressive stream with "telecide().tdecimate()" or similar changes the framrate which may have implications on audio sync and pitch (?).If the source shows symptoms of telecining -- but wouldn't IVTC it is probably hard-telecined. But that would be unusual for a 1080 source. I don't think I've seen one yet. The only way to get around that would be for me to add a hidden option similar to IVTC_480i that would do the same on 1080 sources. It's risky, though, as leaving it on can make a mess of a source that isn't hard-telecined. It's probably best to just deinterlace it.

Since interlaced sources are native to the TV (HDTV uses 1080i), it's likely that it didn't need to deinterlace it. It just presents the two interlaced pictures at 60 fields per second. That is, of course, unless you have your player set to output 1080p instead.

[DoctorM]

I previously had a bonus feature from a DVD. It was mixed frame rate but largely 23.976 with soft telecine flags. I couldn't find any combination of settings that worked well in BDRB. The closest was the defaults that gave me 29,97p output. I ended up doing it manually and it needed to honor pulldown flags and be IVTC'd to get to 24p.
If I missed a setting there I'd like to know what.

[mparade]

@Peter_A - Turn the clock back on your PC to Feb of this year and you can test 45.05 (it's what I'm doing right now).
I do agree, I've seen that recent builds calculate: Full BD with all audio / Target Size = Compression Ratio.
No deduction are made for stripped or compressed tracks.

It my be a per PC issue of some sort, otherwise a lot of people should be reporting undersized discs.

This "turn the clock back" idea and using 45.05 instead of the current one seems to be working on my "issue with only one BD" also.

Thank you.

[Sharc]

If the source shows symptoms of telecining -- but wouldn't IVTC it is probably hard-telecined. But that would be unusual for a 1080 source. I don't think I've seen one yet. The only way to get around that would be for me to add a hidden option similar to IVTC_480i that would do the same on 1080 sources. It's risky, though, as leaving it on can make a mess of a source that isn't hard-telecined. It's probably best to just deinterlace it.

Since interlaced sources are native to the TV (HDTV uses 1080i), it's likely that it didn't need to deinterlace it. It just presents the two interlaced pictures at 60 fields per second. That is, of course, unless you have your player set to output 1080p instead.
The 29.97fps Extra actually has sections with progressive video (interviews) encoded interlaced as per blu-ray standard, followed by hard telecined snippets of the feature (which originally is 23.976fps). As you mention, the TV seems to present everything at 60fps, so encoding as interlaced gives good results.
For playback via projector or PC deinterlacing seems to be the better choice.

[Sharc]

I still do have a problem with Quick-Play backup and edited menus. I have 2 episodes and 1 extra. The Quick-Menu looks ok on the TV, but when I select any one of the episodes the movie does not start but the TV screen turns black with "no video signal" message after some time. When I select the Extra, the Extra playback starts as expected. Weird. I still try to figure out if it depends on a particular workflow.
As a workaround I can import the already encoded disc, check/edit the Quick menu and re-author the disc.
Wild guess: Could the menu problem be related to the inclusion of the secondary video in the Quick-Play backup mode?

[Sharc]

I still do have a problem with Quick-Play backup and edited menus. I have 2 episodes and 1 extra. The Quick-Menu looks ok on the TV, but when I select any one of the episodes the movie does not start but the TV screen turns black with "no video signal" message after some time. When I select the Extra, the Extra playback starts as expected. Weird. I still try to figure out if it depends on a particular workflow.
As a workaround I can import the already encoded disc, check/edit the Quick menu and re-author the disc.
Wild guess: Could the menu problem be related to the inclusion of the secondary video in the Quick-Play backup mode?

Another observation:
1) The edited titles.inf which is written to the source folder
[Titles]
Title=GAME OF THRONES Season 1 Disk 2
Threshold=32400000
00073=Episode 3 [00:57:14.097]
00074=Episode 4 [00:55:37.792]
00075=Episode 5 [00:54:16.503]
[Position]
00073=1
00074=2
00075=3
[Selected]
00073=-1
00074=-1
00075=-1
2) The titles.inf which is written to the target folder folder
[Titles]
Title=GAME OF THRONES Season 1 Disk 2
00001=Episode 3 [00:57:14.097]
00002=Episode 4 [00:55:37.792]
00003=Episode 5 [00:54:16.503]
i.e. the sections [Position] and [Selected] are missing on the target. I then can playback the Title 00003=Episode 5. Selection of 00001 or 00002 turns the TV screen black and the player idles.

3) When I now IMPORT this non-working disc, a complete titles.inf (same as 1) with all sections will be created, which eventually works.

Could 2) be the reason for the Quick-Play failure?

[D-Train61]

Hi JDobbs,
I just downloaded and am running the latest version of your program, 0.46.14 and I'm familiar with your software and have made backups of my Blu-Rays in the past just fine but today I'm trying to backup "The Hobbit: An Unexpected Journey", output size=BD25, and as soon as it starts I get the following error message:
PHASE ONE, Encoding
- [21:41:52] Processing: VID_00001 (1 of 3)
- [21:41:52] Extracting A/V streams [VID_00001]
- Error in attempt to extract audio/subs.
- Bitstream exception Unknown exception. It does not have to be! Please contact application support team for more information.
[21:41:56] - Failed to retrieve audio, aborted

So, since it says "It does not have to be! Please contact application support team for more information." that's what I'm doing. I went into my settings and tried changing my audio encoding options like checking the "use 448 kbs for AC3 encoding" and "Keep HD Audio for BD25" to no avail. I also selected only English as the subtitles to backup.
Any ideas on what I can try next?
Thanks!

[Sharc]

Hi JDobbs,
I just downloaded and am running the latest version of your program, 0.46.14 and I'm familiar with your software and have made backups of my Blu-Rays in the past just fine but today I'm trying to backup "The Hobbit: An Unexpected Journey", output size=BD25, and as soon as it starts I get the following error message:
PHASE ONE, Encoding
- [21:41:52] Processing: VID_00001 (1 of 3)
- [21:41:52] Extracting A/V streams [VID_00001]
- Error in attempt to extract audio/subs.
- Bitstream exception Unknown exception. It does not have to be! Please contact application support team for more information.
[21:41:56] - Failed to retrieve audio, aborted

So, since it says "It does not have to be! Please contact application support team for more information." that's what I'm doing. I went into my settings and tried changing my audio encoding options like checking the "use 448 kbs for AC3 encoding" and "Keep HD Audio for BD25" to no avail. I also selected only English as the subtitles to backup.
Any ideas on what I can try next?
Thanks!
This message comes from tsMuxer

[D-Train61]

So how do I fix it? Do I simply need to upgrade to a newer version of tsMuxer? Or is it that BD-RB can't find tsMuxer? :confused: