BD Rebuilder Beta - Bug Reports Only

[jdobbs]

Sorry, but physics denys it. HD Audio is too big to fit on a DVD-9. There would be little or no space left for the video.

Of course, since the human ear can't detect HD audio compared to DD5.1@640Kbs* it really doesn't matter anyway. JMHO.

Any custom size over 8500 would allow it -- but be careful, you still might end up with a terrible picture or "failure to fit" messages.


*According to Dolby Labs AC-3 audio is transparent (indistinguishable from source material in double-blind tests) at 640Kbs.

[laserfan]

Originally Posted by jdobbs

*According to Dolby Labs AC-3 audio is transparent (indistinguishable from source material in double-blind tests) at 640Kbs.

Or if you're >60yrs old, or otherwise spent years (say, in college) using headphones and wearing-out Led Zeppelin cassette tapes, then you don't bother even w/640k as 448 is plenty good enough.

Not that I know of anyone to whom this applies of course!

[omegaman7]

This is a shot from "I am number four". (They're PNG J.D.)

Jdobbs, is there a way to tweak the dark scenery? What I mean is, can it be more gentle with them? I'm still seeing significant artifacting in the darker scenes. I'm gonna burn my result and see how it looks on the plasma. The plasma has better rendering than my monitor I've noticed that lots of encoders, seem to pull bitrate from the dark scenes. ConvertX is really bad. Sure would be wonderful if we could allocate more bitrate to them. I'm probably just really critical

Original frame #17661, no compression.

Slowest encode setting frame #17671(only 10 frames off, where the last encode was 29).


Ok, the video plays fine on the BD player in the living room. Though it does downscale to 720P. It's not a full HD TV My guess is, that TMT 3 is simply not rendering it properly. THough the original renders better.

[Sharc]

Originally Posted by jdobbs

- Fixed an issue in subtitle end-times related to
  previously resized (via TSMUXER) subtitles.

I just want to report back that subtitles are fine now.
I tried from a 24 fps 720p source. PAL framerate conversion to 25 fps (speedup), audio speedup without pitch shift, and subtitles speedup with correct playback times

[Capsbackup]

Originally Posted by laserfan

Or if you're >60yrs old, or otherwise spent years (say, in college) using headphones and wearing-out Led Zeppelin cassette tapes, then you don't bother even w/640k as 448 is plenty good enough.

Not that I know of anyone to whom this applies of course!

Well if you include Jeff Beck too, I confess

[JJB]

Originally Posted by Capsbackup

Well if you include Jeff Beck too, I confess

Don't leave out Robin Trower or Tommy Bolin.....Then I will testify!

[steveg32]

Originally Posted by jdobbs

@steveg32 and others who have reported sync issues on DVD-5/9 output.


I think I've found a connection. On my test computer I had upgraded to AVISYNTH v2.58 some time ago so I can certify it as the recommended version. When I removed it and reinstalled v2.57 (the currrent recommendation) I was able to repeat the sync issue on certain discs. When I reinstall v2.58, the problem is solved.

So, though it is still not completely clear why, it appears that the DVD-5/9 audio synchronization problem is linked to using the combination of HC Encoder and AVISYNTH 2.57.

Since I have been running v2.58 for some time now, I am confident it is safe to use and I have now updated the first link of this thread so that it now links to that version.

@steveg32

Can you please download and install this version of AVISYNTH and report if it corrects your issue?

Thanks.

@jdobbs

Will do. Mahalo jdobbs!

[jdobbs]

Originally Posted by laserfan

Or if you're >60yrs old, or otherwise spent years (say, in college) using headphones and wearing-out Led Zeppelin cassette tapes, then you don't bother even w/640k as 448 is plenty good enough.

Not that I know of anyone to whom this applies of course!

Cassettes? So you decided to go "HD audio" and give up the 8-tracks, eh? Let's add Hendrix, Jethro Tull, Deep Purple, Steppenwolf, Cream, The Byrds, CSN, Jefferson Airplane, and Creedance Clearwater Revival. We can even cover Page, Beck, and Clapton together just by saying "The Yardbirds".

"Get your motor runnin'..."

All of the above are on my iPod...

[laserfan]

Originally Posted by jdobbs

*According to Dolby Labs AC-3 audio is transparent (indistinguishable from source material in double-blind tests) at 640Kbs.

BTW, am I correct in assuming that bits are applied equally to 5.1 tracks i.e. each track in this case gets 106.67Kbs? And thus my preferred (when I convert from HD audio, I use this, if the original is 640 I just leave it alone) 448 yields 75kbps?

I ask cuz in olden times when I'd rip my CDs, I did so to 160Kbps (80 per stereo track) and it sounded fine to me. I guess now that I make this comparison 448 does seem slightly weak, though I read somewhere else that 448 is as much as 9x% of the population can discern...

[Sharc]

Originally Posted by laserfan

BTW, am I correct in assuming that bits are applied equally to 5.1 tracks i.e. each track in this case gets 106.67Kbs? And thus my preferred (when I convert from HD audio, I use this, if the original is 640 I just leave it alone) 448 yields 75kbps?

I ask cuz in olden times when I'd rip my CDs, I did so to 160Kbps (80 per stereo track) and it sounded fine to me. I guess now that I make this comparison 448 does seem slightly weak, though I read somewhere else that 448 is as much as 9x% of the population can discern...

Good question. I would assume that the LFE channel with an upper cutoff of 120 Hz requires a fraction (say less than 1%) of the bitrate of any of the "regular" 5 channels. Means the LFE channel is "almost free", and we may split the 448kbps to 5 tracks rather than 6, means 448/5=90 kbps for each of the 5 "regular" channels which compares favourably with your former 80 kbps per stereo track. My speculation rather than science ...
And Dolby Digital in Cinemas is 320 kbps. So if we are aiming at Cinema experience we should be happy with even less.

And just want to add Fleetwood Mac and The Who to your list as causes for changing our ears to lowpass filters ....

[jdobbs]

Originally Posted by laserfan

BTW, am I correct in assuming that bits are applied equally to 5.1 tracks i.e. each track in this case gets 106.67Kbs? And thus my preferred (when I convert from HD audio, I use this, if the original is 640 I just leave it alone) 448 yields 75kbps?

I ask cuz in olden times when I'd rip my CDs, I did so to 160Kbps (80 per stereo track) and it sounded fine to me. I guess now that I make this comparison 448 does seem slightly weak, though I read somewhere else that 448 is as much as 9x% of the population can discern...

If I remember correctly (not a sure thing) all but the LFE use equal bitrate and the LFE uses 1/3rd as much. In 448Kbs (DVD) I think that was 96Kbs per channel (80Kbs maybe?) and 32Kbs for LFE. I'd assume 640Kbs is allocated the same way.

[HWK]

Originally Posted by jdobbs

Please note that the recommended version of AVISYNTH has changed. All users are advised to update to v2.5.8.0. A link to that version is included in the first post of this thread.

Jdobbs do you think changing inspect.exe would be good idea so it would say avisynth 2.5.7.0 is not recommended version.

[laserfan]

Originally Posted by Sharc

I would assume that the LFE channel with an upper cutoff of 120 Hz requires a fraction (say less than 1%) of the bitrate of any of the "regular" 5 channels. Means the LFE channel is "almost free"...

Originally Posted by jdobbs

In 448Kbs (DVD) I think that was 96Kbs per channel (80Kbs maybe?) and 32Kbs for LFE.

Well 5x80=400 leaving 48kbps which is over 50% of the size of the other channels. If you believe that the table below has all the legal bitrates in it (taken from the Standards doc I have, which is almost completely indecipherable to me) this is the only way the math comes-out right.

Sorry to take this thread OT but I'm likin' 448-640kbps for 5.1 better every minute.

‘000000’ (0) 32 kbps 60 42 40
‘000001’ (0) 32 kbps 60 43 40
‘000010’ (1) 40 kbps 75 53 50
‘000011’ (1) 40 kbps 75 55 50
‘000100’ (2) 48 kbps 90 65 60
‘000101’ (2) 48 kbps 90 65 60
‘000110’ (3) 56 kbps 105 75 70
‘000111’ (3) 56 kbps 105 76 70
‘001000’ (4) 64 kbps 120 86 80
‘001001’ (4) 64 kbps 120 87 80
‘001010’ (5) 80 kbps 150 108 100
‘001011’ (5) 80 kbps 150 108 100
‘001100’ (6) 96 kbps 180 130 120
‘001101’ (6) 96 kbps 180 130 120
‘001110’ (7) 112 kbps 210 151 140
‘001111’ (7) 112 kbps 210 152 140
‘010000’ (8) 128 kbps 240 173 160
‘010001’ (8) 128 kbps 240 173 160
‘010010’ (9) 160 kbps 300 217 200
‘010011’ (9) 160 kbps 300 217 200
‘010100’ (10) 192 kbps 360 260 240
‘010101’ (10) 192 kbps 360 261 240
‘010110’ (11) 224 kbps 420 303 280
‘010111’ (11) 224 kbps 420 305 280
‘011000’ (12) 256 kbps 480 347 320
‘011001’ (12) 256 kbps 480 348 320
‘011010’ (13) 320 kbps 600 435 400
‘011011’ (13) 320 kbps 600 435 400
‘011100’ (14) 384 kbps 720 521 480
‘011101’ (14) 384 kbps 720 522 480
‘011110’ (15) 448 kbps 840 608 560
‘011111’ (15) 448 kbps 840 610 560
‘100000’ (16) 512 kbps 960 696 640
‘100001’ (16) 512 kbps 960 696 640
‘100010’ (17) 576 kbps 1080 782 720
‘100011’ (17) 576 kbps 1080 783 720
‘100100’ (18) 640 kbps 1200 870 800
‘100101’ (18) 640 kbps 1200 871 800

[jdobbs]

Originally Posted by HWK

Jdobbs do you think changing inspect.exe would be good idea so it would say avisynth 2.5.7.0 is not recommended version.

It will already say v2.5.8.0 is "Ok" -- but will also say "Ok" for v2.5.7.0. The fact is that v2.5.7.0 has been working fine for a long time... I'll probably wait a couple revisions and then start enforcing v2.5.8.0 as the only "Ok". The only issue I've found is when creating DVD output -- and that seem to be selective.

[Ch3vr0n]

Originally Posted by jdobbs

It will already say v2.5.8.0 is "Ok" -- but will also say "Ok" for v2.5.7.0. The fact is that v2.5.7.0 has been working fine for a long time... I'll probably wait a couple revisions and then start enforcing v2.5.8.0 as the only "Ok". The only issue I've found is when creating DVD output -- and that seem to be selective.

You must have missed my post then. http://forum.doom9.org/showpost.php?...ostcount=12097 HP5 backup failed with 2.57 but worked with 2.58.

did get some weird behavior though. If i play the back up on my computer, both movie and extras have audio. If i play the backup on my standalone, only the movie has audio. Extra's play fine but no audio :s

[jdobbs]

Originally Posted by Ch3vr0n

You must have missed my post then. http://forum.doom9.org/showpost.php?...ostcount=12097 HP5 backup failed with 2.57 but worked with 2.58.

did get some weird behavior though. If i play the back up on my computer, both movie and extras have audio. If i play the backup on my standalone, only the movie has audio. Extra's play fine but no audio :s

I read it -- but I have doubts it is definitely v2.5.7 that is the cause (without some other contributing factor), mainly because if it were that alone it would have been reported by many, many users. v2.5.7.0 has been in use with BD-RB for over two years now. I also ran Harry Potter 5 with v2.5.7.0 on my computer and had no issues

[Ch3vr0n]

well then i don't know why it failed with 2.57 but worked with 2.58. However now i have a weird audio problem on playback lo, think i'll get a batch of double layer blanks even though they're expensive. That way i'll be able to back it up without much hassle and have printable double layers for my future 3D discs

[steveg32]

Originally Posted by jdobbs

@steveg32 and others who have reported sync issues on DVD-5/9 output.


I think I've found a connection. On my test computer I had upgraded to AVISYNTH v2.58 some time ago so I can certify it as the recommended version. When I removed it and reinstalled v2.57 (the currrent recommendation) I was able to repeat the sync issue on certain discs. When I reinstall v2.58, the problem is solved.

So, though it is still not completely clear why, it appears that the DVD-5/9 audio synchronization problem is linked to using the combination of HC Encoder and AVISYNTH 2.57.

Since I have been running v2.58 for some time now, I am confident it is safe to use and I have now updated the first link of this thread so that it now links to that version.

@steveg32

Can you please download and install this version of AVISYNTH and report if it corrects your issue?

Thanks.

@jdobbs

Big mahalos! AviSynth v2.58 fixed ALL DVD output sync issues. Also, no more oversizing too! You're amazing brah. But what I most appreciate about you is your tenacity. You never give up. True warrior spirit. Aloha, jdobbs u da man!

[jdobbs]

Originally Posted by steveg32

@jdobbs

Big mahalos! AviSynth v2.58 fixed ALL DVD output sync issues. Also, no more oversizing too! You're amazing brah. But what I most appreciate about you is your tenacity. You never give up. True warrior spirit. Aloha, jdobbs u da man!

Now that you mention it -- the output from my test run was larger with v2.57 also -- mahalo back at you for the follow-up.

Go Swords.

[Jonaldinho]

I have a 1080p video and because I live in the UK it's 25fps, it's not a movie it's a video from a videographer who gave me the HD file and I would like it to be playable on a blu-ray device. I know for blu-ray compliancy at 25fps it needs to be interlaced, I typed in the hidden option DEINTERLACER_TYPE=4 for fake interlaced, it's currently encoding the 2nd pass but the LASTCMD doesn't have --fake-interlaced in the script so I was wondering will it still be fake interlaced? Do I need to enable "use DEINTERLACER on interlaced sources" for fake interlaced to work? Thanks