D2SRoBa 3.80 Released

[tylo]

3.80 Released
http://home.broadpark.no/~tylovset/d2sroba/D2SRoBa380.exe

----------------------------------------------------------------
v3.60
The greatest new feature is conditional filtering for DVD output (or when selecting a fixed number of CD's for SVCD). In this case, the filter will be triggered when Q goes over the 'Auto Q' value (now renamed to 'Threshold Q' value).

The setup can optionally configure the conditional filtering, by modifying the standard resizers (including FACAR). It then also installs the filter with Undot().Deen() and the undot.dll and deen.dll in your avisynth plugin folder if not present.

Note: Although the avisynth.ini resizers adds an 'Import(...)' line, the filtering isn't triggered when using D2S alone (e.g multipass) - the imported script is then empty!

Check the history page for what else is new.

v3.50
- new icon
- added: ini entry: minimize (0=no, 1=yes, 2=(default) when log win is minimized)
- added: ini entry: max_num_cds (default=32). set to 2 if you want max 2 CDs
- improved: tweak to skip some unneeded sample tests
- improved: changed some default values

v3.49
- added: browse for Rejig on setup (if not already set)
- improved: small optimalization/fix in Qsearch

v3.4.8
- reuploaded to work with ReJig.exe
- fixed: bugs from v3.4.7 and v3.4.6 (cce open and shutdown)
- changed: Cmd. line in Requant.exe - compat. with ReJig.exe (config ini file)
- improved: small thing in QSearch again (hmm)
- fixed: Avisynth setup compability issue
- improved: more testing - e.g. for not using frameserving
- improved: robustness with cce 2.50 text input
- improved: more log info

v3.4.4
- fixed: Q search bug intoduced in v3.4.2
- changed: creation of D2SRoBa.avs - may fix a problem with corrupt .avs files
- changed: very minor tweaks

v3.4.2
- added: new build of Requant.exe (modified from Makira's latest source - included)
- fixed: Q search in rare situations
- improved: better first guess of Q (for fixed number of CDs)
- improved: log text
- tips: Use EclCCE v1.8b or later, and check 'Hide shutdown button' checkbox

v3.4.1
- fixed: Now calculates num. of sample frames (instead of launching CCE for it)

v3.4.0
- fixed: Requant.exe modified to not create read-only output files (source included)
- fixed: Batch run: Now uses global d2sroba.ini file if project ini file is missing
- added: Support for EclCCE v1.8. Supports older versions too (shows a note)
- added: With EclCCE v1.8, CCE opens minimized, thanks to RB
- added: Greyed out GUI controls which are not used
- added: Re-introduced "educated guessing" on first Q for fixed num. CDs
- changed: Replaced 'Adjust Q' with 'Adjust %' (default 1.0%)
- changed: Batch command line is now like D2S's. Try D2SRoBa.exe -?
- changed: Num CDs can be up to 32 now
- changed: D2SRoBa_log.txt now in same directory as dvd2svcd_log.txt
- changed: Some default values
- improved: Made Q search more robust
- improved: Log text more accurate

Note the new batch line: Equal to DVD2SVCD command line, but '-d2s:...' must be first argument. Example: The following will open a project and use the local d2sroba.ini file if it is there, otherwise the global:

D2SRoBa -d2s:"path\dvd2svcd project file.d2s"

[bobwillis]

Hi,

In the meantime, using 3.4.0rc3, I have had two instances where "waiting for CCE window" did not appear in the D2SRoBa window. DVD2DVD & CCE completed the job, but of course without RoBa's intervention, the size of the job was wrong. I rebooted the machine and tried again with no joy. However, on the third attempt (after closing D2SRoBa & D2S, then re-launching D2SRoBa) it worked. I'm confused, but at the moment I've given you all the info I can about the problem. I will post more if I can establish more deterministic information.

EDIT: I've tried it several times this evening, and I can't reproduce the fault. If no-one else reports it, don't worry about it. My machine was behaving very strangely yesterday evening. Spumux wasn't working, and then I came home tonight to find it was working ok again - I've phoned Mulder & Scully, but they said they'd retired.

Regards,
Bob

[Gideon25]

Batch function now works FLAWLESSLY! I have fixed my little guide and we now have BATCH avi PAL-->Ntsc conversions!
http://forum.doom9.org/showthread.php?s=&postid=460310#post460310

No more going thru steps needlessly! Now I can setup 10-15 files to convert overnight.

[DDogg]

tylo, I notice RB has a new version of eclcce out. I think one particular item may cause you some grief if I remember correctly.

"- Fixed: create_new_vaf ECL option was reversed. Now create_new_vaf=1 really sets "Create new file" and create_new_vaf=0 sets "Use existing file".

Note to all D2s beta Testers - test site has been updated

[dimirass]

Using D2SRoBa v3.4.0 RC3, I have got a Transcoding pass result: 66.1% on target (2819503323 / 4266563190).

Checking the transcoded file I realised that transcoding stopped after 45 minutes of movie.Is there any explanation for this?

On the other hand it seems that the "read only" problem has been fixed.Log file is following:


--------------------------------------------------------
D2SRoBa v3.4.0 RC3, by Tylo
--------------------------------------------------------
- WIN_XP - AutoIt 3.0.100.0
- 2004-03-24 21:59:11
--------------------------------------------------------
- Output type : DVD
- CD size : 4470
- Number of CD's : 1
- Sample percent : 2.0
- Adjust Q : 0.6
- Cond. sizing pass : Yes (Transcode if opv sz > 0.0% above target, and Q <= 25)
(Encode if opv sz < -2.0% from target, or Q > 25)
- Clean previous : Yes
- Run mode : Normal
--------------------------------------------------------
- Waiting for CCE window ...
- Detected a CCE encode window
- Detected the CCE main movie encode window... trying to kill it.
- Finished shutdown of CCE and D2S
- Avisynth setup: FitCD
- Setting high process priority
- Using CCE version 2.67
--------------------------------------------------------
- Movie length : 01:33:02 (139567 frames, 25 fps)
- D2S audio brate : (448 + 0) kbps
- D2S video estim. : 6114 kbps, 1 CDs
- Select Ranges : every 600, select 12 frames
- Sample frames : 2796
--------------------------------------------------------
Calculations:
- 1 CD: video_br=6114 (6114) audio1_br=448 video_sz=4266563190 fill=100% cbr=0
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- Number of CDs : 1
- Target mpv BR : 6114 (max 8500) kbps, size=4266563190 bytes
--------------------------------------------------------
Search for Q:
- Estim. mpv BR : 3472 (Q=32, err=-43.2%, size=2423553201, sample sz=48551984)
- Estim. mpv BR : 6640 (Q=10, err=8.6%, size=4634193849, sample sz=92838608)
- Estim. mpv BR : 5851 (Q=14, err=-4.3%, size=4083216478, sample sz=81800664)
- Estim. mpv BR : 6042 (Q=13, err=-1.2%, size=4216420101, sample sz=84469184)
- Determined Q : 12 = round(13 - 0.6 + (-1.2)/3.1)
--------------------------------------------------------
- 2004-03-24 22:58:06
- Start movie OPV encoding (Q 12)
- OPV pass result: 101% on target (4311029064 / 4266563190) Speed factor: 1.22
--------------------------------------------------------
- 2004-03-25 00:14:23
- Start transcoding (Requant 98.97%)
- Transcoding pass result: 66.1% on target (2819503323 / 4266563190) Speed factor: 8.25
--------------------------------------------------------
- 2004-03-25 00:25:40
- Recover DVD2SVCD

[DDogg]

Checking the transcoded file I realised that transcoding stopped after 45 minutes of movie.Is there any explanation for this? I can just offer a suggestion. Before anybody can really figure it out you would first have to swear on your firstborn that you had run the prime95 torture test (http://mersenne.org/gimps/p95v238.exe) for 6-24 hours without any failure, and do so at the approximate same ambient room temperature that you had at time of failure.

Failure due to CPU heat would be the very first area I would check. I say this cause I had it happen to me the other day when the room was unusually hot. Took me forever to figure out what was going on.

[sPeziFisH]

Finding Nemo (with KISS (comment: in my opinion it is not good to use KISS with animated catoon as every pixel is where it has to be, no 'noise' (I would say))

--------------------------------------------------------
D2SRoBa v3.4.0 RC4, by Tylo
--------------------------------------------------------
- WIN_XP - AutoIt 3.0.101.0
- 2004-03-25 17:31:17
--------------------------------------------------------
- Output type : SVCD
- CD size : 800
- Number of CD's : auto
- Sample percent : 1.0
- Auto Q / Max Q : 40 / 60
- Adjust % : 1.5
- Cond. sizing pass : Yes (Transcode if opv sz > 0.0% above target, and Q <= 40)
(Encode if opv sz < -3.0% from target, or Q > 40)
- Cut last image : If opv sz > 20% oversized
- Audio after video : Yes (192 kbps)
- Cond. filtering : If enabled and est. oversize < 10%
- Adjust max br : No
- Clean previous : Yes
- Run mode : Normal
--------------------------------------------------------
- Waiting for CCE window ...
- Detected a CCE encode window
- Detected the CCE main movie encode window - closing it
- Finished shutdown of CCE and D2S
- Avisynth setup: FACAR
- Using CCE version 2.67
--------------------------------------------------------
- Movie length : 01:36:26 (144670 frames, 25 fps)
- D2S audio brate : (192 + 0) kbps
- D2S video estim. : 2054 kbps, 2 CDs
- Select Ranges : every 1200, select 12 frames
- Sample frames : 1452
--------------------------------------------------------
Calculations:
- 1 CD: video_br=923 (923) audio1_br=192 video_sz=667652050 fill=100% cbr=0
- 2 CD: video_br=2054 (2054) audio1_br=192 video_sz=1485760900 fill=100% cbr=0
- 3 CD: video_br=2480 (3185) audio1_br=192 video_sz=1793908000 fill=79.2% cbr=1
--------------------------------------------------------
Computing target number of CDs:
- Estim. mpv BR : 1573 (Q=40, err=?%, size=1138514640, sample sz=11426856)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- Number of CDs : 2
- Target mpv BR : 2054 (max 2480) kbps, size=1485760900 bytes
--------------------------------------------------------
Search for Q:
- Estim. mpv BR : 1866 (Q=25, err=-9.2%, size=1349956820, sample sz=13549024)
- Estim. mpv BR : 2072 (Q=15, err=0.9%, size=1498807902, sample sz=15042988)
- Estim. mpv BR : 2051 (Q=16, err=-0.1%, size=1483857872, sample sz=14892940)
- Determined Q : 14 = Round(16 + (-0.1 - 1.5)/1)
--------------------------------------------------------
- 2004-03-25 17:55:32
- Start movie OPV encoding (Q 14)
- OPV pass result: 1% on target (14976960 / 1485760900) Speed factor: 65.51


...here I cancelled the encode, forgot to set cond.siz.pass under to -1.0 (don't like -3.0) and I wanted to do an encode without deen.
Problem:
Q=15 gives an error of +0.9%
Q=16 gives an error of -0.1%
'Adjust %'=1.5 makes D2SRoBa use Q=14 - Q=16 is close enough, ok, with some estimating-impresiceness (sample not 100%), Q=15 is still the best (?), requant could do the rest if necessary.
Q=14 would probably cause an error bigger than 0.9%

Next encode:

Finding Nemo (I used undot but who cares (I don't really know if it is a good idea to use undot, deen or evenboth (aka KISS) on material like that - as you can see, with KISS (see above) gives you a better Q, but I didn't give me the possibility to compare the quality).
Interesting: estimated bitrates of Q=16 and Q=21 are very close (hehe, a really mystery Q-machine).
Now I set 'Adjust %' to 0.9.

--------------------------------------------------------
D2SRoBa v3.4.0 RC4, by Tylo
--------------------------------------------------------
- WIN_XP - AutoIt 3.0.101.0
- 2004-03-25 18:00:33
--------------------------------------------------------
- Output type : SVCD
- CD size : 800
- Number of CD's : auto
- Sample percent : 1.0
- Auto Q / Max Q : 40 / 60
- Adjust % : 0.9
- Cond. sizing pass : Yes (Transcode if opv sz > 0.0% above target, and Q <= 40)
(Encode if opv sz < -1.0% from target, or Q > 40)
- Cut last image : If opv sz > 20% oversized
- Audio after video : Yes (192 kbps)
- Cond. filtering : If enabled and est. oversize < 10%
- Adjust max br : No
- Clean previous : Yes
- Run mode : Crash recovery
--------------------------------------------------------
- Waiting for CCE window ...
- Detected a CCE encode window
- Detected the CCE main movie encode window - closing it
- Finished shutdown of CCE and D2S
- Avisynth setup: FACAR
- Using CCE version 2.67
--------------------------------------------------------
- Movie length : 01:36:26 (144670 frames, 25 fps)
- D2S audio brate : (192 + 0) kbps
- D2S video estim. : 2054 kbps, 2 CDs
- Select Ranges : every 1200, select 12 frames
- Sample frames : 1452
--------------------------------------------------------
Calculations:
- 1 CD: video_br=923 (923) audio1_br=192 video_sz=667652050 fill=100% cbr=0
- 2 CD: video_br=2054 (2054) audio1_br=192 video_sz=1485760900 fill=100% cbr=0
- 3 CD: video_br=2480 (3185) audio1_br=192 video_sz=1793908000 fill=79.2% cbr=1
--------------------------------------------------------
Computing target number of CDs:
- Estim. mpv BR : 1657 (Q=40, err=?%, size=1199148507, sample sz=12035416)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- Number of CDs : 2
- Target mpv BR : 2054 (max 2480) kbps, size=1485760900 bytes
--------------------------------------------------------
Search for Q:
- Estim. mpv BR : 1899 (Q=28, err=-7.5%, size=1373787117, sample sz=13788200)
- Estim. mpv BR : 2068 (Q=20, err=0.7%, size=1496473654, sample sz=15019560)
- Estim. mpv BR : 2047 (Q=21, err=-0.3%, size=1481042586, sample sz=14864684)
- Determined Q : 20 = Round(21 + (-0.3 - 0.9)/1)
--------------------------------------------------------
- 2004-03-25 18:05:43
- Start movie OPV encoding (Q 20)
- OPV pass result: 98.2% on target (1459463956 / 1485760900) Speed factor: 1.38
--------------------------------------------------------
- 2004-03-25 19:15:47
- Start VBR sizing pass (2054 kbps)
- Sizing pass result: 100% on target (1485646316 / 1485760900) Speed factor: 1.38
- No Audio 1 (D:\nemo\Encoded_audio_1_FS1.mp2)
--------------------------------------------------------
- 2004-03-25 20:25:55
- Recover DVD2SVCD
--------------------------------------------------------
- 2004-03-26 00:11:03
- D2SRoBa successfully finished
--------------------------------------------------------

With 'Adjust %'=0.9 I am really satisfied, under 1.0 seems to be my 'sweet spot' (hmmm ;-P)

Tell me if sharing the logs or only quoting the interesting parts is better instead of making such a big post.
Experimenting with 'Finding Nemo' is possible if wanted.
So far Tylo, great work, once again.

[ddogg-Edited with size tags]

[DDogg]

tylo, I have a DVD test just started using 2 audio tracks. The line from the log [rc6] seems to indicate D2SRoBa is only using 1 of the audio tracks in the calcs. This may not be so, but I thought I would bring up the subject in the oft chance you check this post before the encode is completed and I actually know whether the size target is accurate.

/add1:I think you are picking it up after re-examining the log and seeing the D2S bitrate estimation, but the line in bold is confusing.

/add2: No problem, the filesize is fine. The text of the log was confusing to me. Just a cosmetics issue, if one at all.

- Using CCE version 2.50
--------------------------------------------------------
- Movie length : 01:29:12 (128343 frames, 23.976 fps)
- D2S audio brate : (448 + 384) kbps
- D2S video estim. : 2471 kbps, 1 CDs
- Select Ranges : every 1200, select 12 frames
- Sample frames : 1284
--------------------------------------------------------
Calculations:
- 1 CD: video_br=2471 (2471) audio1_br=448 video_sz=1653401072 fill=100% cbr=0
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- Number of CDs : 1
- Target mpv BR : 2471 (max 8000) kbps, size=1653401072 bytes
--------------------------------------------------------

[tylo]

Seems like we're closing in on a stable release now :)

@DDogg:
Yes, as you say, it it picking up the second audio, but the printout may be misleading, or non-complete. Don't think I'll do anything.

Regarding your request on using many segments (i.e. num of CDs). Well, it took me some time to realize what you where aiming at, but I think the basic idea is superb. /EDIT: [At first I had doubts, and thought that the content of a 4350 disk wouldn't fit on eight 545 meg disks because of the header alignment overhead (15 kbps). But the total media size is not changed, so that should not apply].
Should work great!

About reduction of CDs when filter is applied: yes, the release v3.4.0 will reduce num CDs with more than one if that should ever occur. Up to 32 segments supported!

[r6d2]

Originally posted by tylo
Seems like we're closing in on a stable release now :)@Tylo, don't forget to check out DDogg's comment on EclCCE 1.8, it may make our life miserable (although it minimizes CCE now!). Besides, the new DVD2SVCD RC1 has sort of changed the INI files, according to the readme.

Seems that all geniuses have decided to release at once! :D:D:D

I'll wait for this to cool down before updating the FAQ. :)

/Add:
I cannot believe this thread got 24000 views in just three days!

[DDogg]

tylo, being able to generate the optimized DVD elementary streams via auto SVCD mode is interesting to me. I certainly did not do a very good job of explaining my thoughts on the method, so I very much appreciate your vote of confidence. As you know, it allows the use of SVCD auto-mode and the conditional filter to create DVD elementary streams that achieve the user preset Q/quality target while ensuring the final size is always on the edge of a user specified DVD increment (1/4,1/8,1/16,and it sounds like 1/32 in the release, of a DVD disk). This ensures absolutely no wasted space, so it it easy to author multiple movies on one DVD with matched quality using one's favorite authoring tool.

However, I think the method is rather 'one off' for most people, and in fact I doubt the technique will be used much. So, I don't think it is worth too much additional effort on your part except for the segment increase [and multi-segment reduction upon conditional filter success. 32 segments would be perfect!]. That is very good news and thanks a lot for even considering it.

After the new D2SRoBa and DVD2X releases, I'll do a small writeup on the method for those that want to experiment with it.

On that note, I wanted to say how much I'm looking forward to the two releases (finally), D2SRoBa and D2S together allow extreme high quality (the highest IMHO), the highest speed, and the ultra minimization of file-size to be all rolled up together in one solution for both SVCD and movie-only DVD. It is a winning combination that should be around for a long time and I think a ton of people are going to really appreciate the huge effort both you and D2S have put forth. Especially after they download them and see the end result of all these impossibly long winded and flatulent threads. (:-D!)

[windtrader]

I cannot believe this thread got 24000 views in just three days!Yeah! I see the gagles of vultures circling above, waiting for the release. :)


It is a winning combination that should be around for a long time and I think a ton of people are going to really appreciate the huge effort both you and D2S have put forth.Hey Dog, why don't you dust off that old "say thanks and how about a donation" thread. It is surely deserved again.

[r6d2]

Originally posted by windtrader
Yeah! I see the gagles of vultures circling above, waiting for the release.Humm... I think that as DDogg splitted the thread, it inherited the previous view number, so the vultures are actually the shadows of previous vultures. :)

[windtrader]

Good news! We have Shadow vultures around here and I let my kids out to play baseball; they are harmless :p

btw - anyone know how much time left for beta testing? I'd like to make sure they all muster up the energy to give it one last battery of acid tests. Humm maybe round of tests is better stated.

[manolito]

I did three successful encodes with RC6, no complaints at all. Of course I don't do any fancy stuff like convert to DVD, batch encodes, more than one audio stream and so on. Just plain SVCDs, all in size mode, about 90% are 1 CD encodes.
Adjust % default 1.8 seems to be too high. I think somewhere around 1.0 is good. Please share your experience, so I can set a good default for the release.
I abandoned jonny's Restrict Auto I-frame insertion trick and went back to the old method again (after encoding "The Insider" and getting a 18% undersized OPV encode), and with the old method a value of 3 for "Adjust %" seems to be the optimum for me.

Cheers
manolito

[manolito]

Version 3.40 release

Thanks a million! :)
Upcoming v3.4.0 will be the last release for a while.
I hope you can stick to this when the next version of DVD2SVCD comes out. You sure deserve some rest from all of us here...

Cheers
manolito

[bobwillis]

Hi Tylo,

Just a quick note of thanks for v3.4.1. I have done a couple of different DVD2DVD encodes here and everything seems fine. My last project was glueing together both sides of 'Schindler's list' using vobedit & ifoedit (giving a total of 12GB), stripping it with ifoedit leaving 9.5GB (one remaining AC3 6ch audio track), and reencoding with D2SRoBa & DVD2DVD with FACAR. The Q figure I first got was 55, so I decided to use undot&deen (KISS), and got it down to 41. Of course, the additional VBR sizing pass was used. I was surprised, it looks excellent considering the average bitrate is only 2.66Mbps!

--------------------------------------------------------
D2SRoBa v3.4.1, by Tylo
--------------------------------------------------------
- WIN_XP - AutoIt 3.0.101.0
- 2004-03-29 20:38:49
--------------------------------------------------------
- Output type : DVD
- CD size : 4470
- Number of CD's : 1
- Sample percent : 2.0
- Adjust % : 1.0
- Cond. sizing pass : Yes (Transcode if opv sz > 0.0% oversize, and Q <= 25)
(Encode if opv sz < 1.5% undersize, or Q > 25)
- Clean previous : Yes
- Run mode : Crash recovery
--------------------------------------------------------
- Waiting for CCE window ...
- Detected a CCE encode window
- Detected the CCE main movie encode window - closing it
--------------------------------------------------------
- 2004-03-29 20:40:14
- Finished shutdown of CCE and D2S
- Avisynth setup: FACAR
- Using CCE version 2.67
--------------------------------------------------------
- Movie length : 03:15:14 (280874 frames, 23.976 fps)
- D2S audio choice : 448 + 0 kbps
- D2S video calc. : 2663 kbps, 1 CDs
- Select Ranges : every 600, select 12 frames
- Sample frames : 5628
--------------------------------------------------------
Calculations:
- 1 CD: video_br=2663 (2663) audio_br=448+0 video_sz=3899563428 fill=100% cbr=0
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- Number of CDs : 1
- Target mpv BR : 2663 (max 9500) kbps, size=3899563428 bytes
--------------------------------------------------------
Search for Q:
- Sample encode : Q=26: 3592 kbps, err=34.9%, size=5261145517, sample sz=105419964
- Sample encode : Q=41: 2674 kbps, err=0.4%, size=3916545584, sample sz=78477604
- Determined Q : 41 = Round(41 + (0.4 - 1.0)/2.3)
--------------------------------------------------------
- 2004-03-29 20:52:17
- Start movie OPV encoding (Q 41)
- OPV pass result: 97.7% on target (3809771816 / 3899563428) Speed factor: 0.94
--------------------------------------------------------
- 2004-03-30 00:20:47
- Start VBR sizing pass (2663 kbps)
- Sizing pass result: 99.9% on target (3895550696 / 3899563428) Speed factor: 0.97
--------------------------------------------------------
- 2004-03-30 03:41:11
- Recover DVD2SVCD
--------------------------------------------------------
- 2004-03-30 06:14:29
- D2SRoBa successfully finished
--------------------------------------------------------


Anyway, enough of my ranting,
thanks once again,
Bob

[NuroPreak]

I tried 3.4.1 and the sample size was ok, but i'm not so good in reading the log file what went wrong and why ended i up with a 1 cd rip instead of 2 :rolleyes:

--------------------------------------------------------
D2SRoBa v3.4.1, by Tylo
--------------------------------------------------------
- WIN_XP - AutoIt 3.0.101.0
- 2004-03-30 18:37:15
--------------------------------------------------------
- Output type : SVCD
- CD size : 800
- Number of CD's : auto
- Sample percent : 1.0
- Auto Q / Max Q : 35 / 60
- Adjust % : 1.0
- Cond. sizing pass : Yes (Transcode if opv sz > 0.5% oversize, and Q <= 40)
(Encode if opv sz < 3.0% undersize, or Q > 40)
- Cut last image : If opv sz > 0.4% oversize, and opv sz < 0.5% oversize
- Audio after video : Yes (10% of video, min 224 kbps)
- Cond. filtering : If avisynth configured, and estim. oversize < 8% for Num CDs-1
- Adjust max br : Yes (total: 2754, max avg offset: 300 kbps)
- Clean previous : Yes
- Run mode : Normal
--------------------------------------------------------
- Waiting for CCE window ...
- Detected a CCE encode window
- Detected the CCE main movie encode window - closing it
--------------------------------------------------------
- 2004-03-30 19:09:38
- Finished shutdown of CCE and D2S
- Avisynth setup: BicubicResize
- Using CCE version 2.67
--------------------------------------------------------
- Movie length : 01:46:11 (159286 frames, 25 fps)
- D2S audio choice : 224 + 0 kbps
- D2S video calc. : 1818 kbps, 2 CDs
- Select Ranges : every 1500, select 15 frames
- Sample frames : 1605
--------------------------------------------------------
Calculations:
- 1 CD: video_br=789 (789) audio_br=224+0 video_sz=628383270 fill=100% cbr=0
- 2 CD: video_br=1818 (1818) audio_br=224+0 video_sz=1447909740 fill=100% cbr=0
- 3 CD: video_br=2530 (2847) audio_br=224+0 video_sz=2014967900 fill=89.7% cbr=1
--------------------------------------------------------
Computing target number of CDs:
- Sample max BR : 2530 (adjusted for 2 CDs)
- Sample encode : Q=35: 48 kbps, err=?%, size=38343366, sample sz=386356
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- Number of CDs : 1
- Target mpv BR : 789 (max 2530) kbps, size=628383270 bytes
--------------------------------------------------------
Search for Q:
- Sample encode : Q=6: 80 kbps, err=-89.9%, size=64373775, sample sz=648644
- Sample encode : Q=1: 98 kbps, err=-87.6%, size=78486216, sample sz=790844
- Determined Q : 1 = Round(1 + (-87.6 - 1.0)/0.5)
--------------------------------------------------------
- 2004-03-30 19:10:10
- Start movie OPV encoding (Q 1)
- OPV pass result: 289.4% on target (1818854860 / 628383270) Speed factor: 0.31
--------------------------------------------------------
- 2004-03-31 00:55:30
- Start transcoding (Requant 34.55%)
- Transcoding pass: 136.6% on target (858479876 / 628383270) Speed factor: 14.81
--------------------------------------------------------
- 2004-03-31 01:02:42
- Executing: C:\Program Files\AVI2SVCD\BeSweet\BeSweet.exe -core( -input "E:\RIPPEN\Audio\Extracted_audio_1.ac3" -output "E:\RIPPEN\Audio\Encoded_audio_1.mp2" -logfile "E:\RIPPEN\Audio\Encoded_audio_1.log" ) -azid( -L -3db -l 0db -c normal -s surround -g max ) -shibatch( --rate 44100 ) -2lame( -e -b 224 -m s )
--------------------------------------------------------
- 2004-03-31 01:41:32
- Recover DVD2SVCD

[tylo]

Quite strange. The samples are consistently only a fraction of the size they should be. Can you post D2SRoBa.avs and D2SRoBa.ecl too?

[NuroPreak]

here is my D2SRoba.ecl
; CCE SP Trial Version -- Encoder Control List

[item]
title=D2SRoBa.avs
vid_out=1
vaf_out=0
aud_out=0
aud_mode=0
vid_file0=E:\RIPPEN\Video\D2SRoBa_Q1.mpv
vid_file1=E:\RIPPEN\Video\D2SRoBa_Q1.mpv
vaf_file=E:\RIPPEN\Video\CCE_Video_Information_File.vaf
aud_file=E:\RIPPEN\Video\Encoded_Video_CCE_PAL.mpv.mp2
file_focused=0
video_type=2
width=480
height=576
timecode=0x01000000
tc_ref_frm=0
drop_frame=0
cbr_brate=6000
opv_q_factor=1
opv_brate_min=300
opv_brate_max=2530
vbr_brate_avg=6000
vbr_brate_min=2000
vbr_brate_max=9000
vbr_bias=20
vbr_pass=3
create_new_vaf=1
pulldown_detect=0
letter_box=0
panscan=0
aspect_ratio=2
credits_tweak=0
credits_start=0x00000
credits_brate=1000
h_filter=1
h_filter_idx=8
use_filter=1
filter_val=2
dither=0
dither_max=8
quality_prec=20
intra_dc_prec=-1
alternate_scan=0
non_linear=0
progressive=1
gop_m=3
gop_nm=5
gop_hdr=15
seq_hdr=1
seq_endcode=1
dvd=0
all_closed_gop=0
fix_gop_length=1
fix_vbv_delay=0
top_first=0
half_width=0
half_height=0
lum_level=1
packet_size=2048
frame_rate_idx=3
qmat_idx=0
mpeg1=0
mpeg1_cps=1
brate_idx=7
stereo=2
samples_per_sec=44100
crc=1

[file]
name=E:\RIPPEN\Video\D2SRoBa.avs
type=0
frame_first=0
frame_last=1605
encode_first=0
encode_last=1605
chapter=0:Chapter0
chapter=9978:Chapter1
chapter=15983:Chapter2
chapter=22160:Chapter3
chapter=32757:Chapter4
chapter=41192:Chapter5
chapter=46126:Chapter6
chapter=56094:Chapter7
chapter=66702:Chapter8
chapter=75313:Chapter9
chapter=84658:Chapter10
chapter=95301:Chapter11
chapter=113559:Chapter12
chapter=126638:Chapter13
chapter=139620:Chapter14
chapter=149952:Chapter15


and my d2sroba.avs is 11mb i bit big to post it here.. i would need multiple threads to post that one so i zipped it and uploaded it to my webspace

http://members.lycos.nl/tristandoornebosch/D2SRoBa.rar

[windtrader]

I am trying a new strategy to maximize the chances of doing one pass, avoiding the sizing pass. I am making the CD size smaller and using a minus and plus adjustment to allow for both a small undersize AND oversize result on the first pass to still fit on the DVD and avoid the second pass.

In this run, the first pass was just over the target and posted this message.

- OPV pass result: 100.4% on target (3974034424 / 3956850788) Speed factor: 1.52
- Oversized: Cutting last CD image off by 0.43427556207358%

I checked the length of the pre and post movie files and the running length is the same to the second. The output does not seem to be "cut off". Is there any real cutting off going on or is the text of the message a bit misleading?

thx


--------------------------------------------------------
D2SRoBa v3.4.0, by Tylo
--------------------------------------------------------
- WIN_XP - AutoIt 3.0.101.0
- 2004-03-31 10:40:42
--------------------------------------------------------
- Output type : DVD
- CD size : 4460
- Number of CD's : 1
- Sample percent : 1.5
- Adjust % : 1.0
- Cond. sizing pass : Yes (Transcode if opv sz > 1.5% oversize, and Q <= 50)
(Encode if opv sz < 1.5% undersize, or Q > 50)
- Clean previous : Yes
- Run mode : Normal
--------------------------------------------------------
- Waiting for CCE window ...
- Detected a CCE encode window
- Detected the CCE main movie encode window - closing it
--------------------------------------------------------
- 2004-03-31 10:54:21
- Finished shutdown of CCE and D2S
- Avisynth setup: FACAR
- Setting low process priority
- Using CCE version 2.67
--------------------------------------------------------
- Movie length : 02:08:06 (184302 frames, 23.976 fps)
- D2S audio brate : (448 + 192) kbps
- D2S video estim. : 4118 kbps, 1 CDs
- Select Ranges : every 800, select 12 frames
- Sample frames : 2772
--------------------------------------------------------
Calculations:
- 1 CD: video_br=4118 (4118) audio_br=640 video_sz=3956850788 fill=100% cbr=0
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- Number of CDs : 1
- Target mpv BR : 4118 (max 9000) kbps, size=3956850788 bytes
--------------------------------------------------------
Search for Q:
- Sample encode : Q=16: 4522 kbps, err=9.8%, size=4345418303, sample sz=65357400
- Sample encode : Q=19: 3968 kbps, err=-3.6%, size=3813442148, sample sz=57356196
- Sample encode : Q=18: 4142 kbps, err=0.6%, size=3980714963, sample sz=59872068
- Determined Q : 18 = Round(18 + (0.6 - 1.0)/4.2)
--------------------------------------------------------
- 2004-03-31 11:00:30
- Start movie OPV encoding (Q 18)
- OPV pass result: 100.4% on target (3974034424 / 3956850788) Speed factor: 1.52
- Oversized: Cutting last CD image off by 0.43427556207358%
--------------------------------------------------------
- 2004-03-31 12:25:01
- Recover DVD2SVCD
--------------------------------------------------------
- 2004-03-31 21:01:33
- D2SRoBa successfully finished
--------------------------------------------------------

[tylo]

It's misleading. There is a few other small matters that should be fixed (nothing urgent), so I will make another "frozen" version.

@NuroPreak: The D2SRoBa.avs file is supposed to be a small text (avisynth) file, equal to 'Avisynth_Script_file.avs', but with one extra line. Something is messed up. Clean the output directory and try again - report if it happens again.

/Add: Seems like 'Avisynth_Script_file.avs' is messed, because the line I added is at the end of the big file you posted.

[NuroPreak]

Originally posted by tylo
It's misleading. There is a few other small matters that should be fixed (nothing urgent), so I will make another "frozen" version.

@NuroPreak: The D2SRoBa.avs file is supposed to be a small text (avisynth) file, equal to 'Avisynth_Script_file.avs', but with one extra line. Something is messed up. Clean the output directory and try again - report if it happens again.

/Add: Seems like 'Avisynth_Script_file.avs' is messed, because the line I added is at the end of the big file you posted.

i already restarted and again same problem :(

what do i need to do about the avisynth problem?

//edit i'm trying to disable the change cd pic maybe that helps.. ill post the results


//edit i keep gettin the same problem :(

[tylo]

The old Requant_old.exe is bundled. Compare quality difference with the new Requant.exe, by experimenting in a cmd.exe window with an mpv file:
Requant.exe 95 movie.mpv test_new.mpv
Requant_old.exe 95 movie.mpv test_old.mpv

Can you see any differences between those three .mpv's?

@NuroPreak: try v3.4.2, and add 'debug=1' in D2SRoBa.ini. Look at the Log 'Avs file: ....', 'Avs file size: ..', and 'D2SRoBa.avs size:...'. Both the sizes should only be a few hundred bytes.

[Venom_IL]

Sometimes, especially when I use batch mode, when D2SRoba closes the CCE window opened by D2S in order to open hers, I get a "memory allocation error" or something of the sort from CCE (I apologize for forgetting to take a screenshot, will do next time) .. of course it's meaningless since CCE is shut down anyway but the script doesn't "press ok" and that halts the process / batch until the user presses ok only then D2sroba's CCE is launched and the process is continued

So my idea is to modify the script so that it'd automatically press ok for any message CCE might bring up..

Of course, this is nothing more than a thought for you to consider, tylo :cool:

[tylo]

@Venom_IL: I'll have to know the exact message to remove it. I haven't seen that message myself.

@NuroPreak: Please, try http://home.tiscali.no/tylohome/files/D2SRoBa343.exe and report back if this version fixes the problem you describe. (One more has reported this).

[Venom_IL]

well noted, i will take a screenshot if it happens again

[DDogg]

tylo, I would sure feel better if we knew if NuroPreak's machine will run the Prime95 torture test (http://mersenne.org/gimps/p95v238.exe) for 4 hours without error. In fact, may I suggest than anybody that wishes to report a one off error do this test before posting. For that matter, please throw in one full cycle of memtest (http://www.memtest86.com/)

The main reason for this is that CCE, when it encounters a missed beat of the CPU, will not crash like many programs. Rather, it just drops out <poof and vanish>. D2S or D2SRoBa interprets this to mean the encoding cycle is finished as the system will keep working with no error thrown. BeSweet, on the other hand can do the same thing which causes the same problem during the audio encoding cycle, or sometimes it will present a black screen lock-up. A black screen with BeSweet is 99% CPU overheat or an unstable overclock although bad ram can do the same thing.

Btw, many people are presently running with bad ram and don't know it. I've seen this on 5 machines this year. The mpeg2dec/avisynth/CCE combination will use an extra 100 megs of memory. That is why ram problems tend to show up during encoding. Running memtest for one complete cycle will tell you one way or the other.

[Holomatrix]

Hi guys, been away from SVCD for a while and just tried one with the latest versions last night and it messed up. Can you tell me where I can tell D2Sroba to not reset the DVD2SVCD bitrate settings? The reason being is I think that is what messed me up last night. Sorry, I don't have the log, re-did the movie putting back the bitrate setting in DVD2SVCD (92-138 min 1 CD) after loading D2Sroba and it worked, but when I was reading the old log I saw that D2Sroba was set for 1 CD but it also said that DVD2SVCD was set for two CD's, D2Sroba found a Q of 1 then realized later that 1.6 gig was over my 1 CD size, transcoded to the proper size of 813 MEG. Weird.


EDIT: It was in my Recycle bin. Cool. I post the old one first then the one it was supposed to do in the first place.

--------------------------------------------------------
D2SRoBa v3.4.2, by Tylo
--------------------------------------------------------
- WIN_XP - AutoIt 3.0.101.0
- 2004-04-04 21:06:12
--------------------------------------------------------
- Output type : SVCD
- CD size : 800
- Number of CD's : 1
- Sample percent : 2.0
- Adjust % : 1.0
- Cond. sizing pass : Yes (Transcode if opv sz > 0.1% oversize, and Q <= 60)
(Encode if opv sz < 2.5% undersize, or Q > 60)
- Audio after video : Yes (9% of video, min 128 kbps)
- Adjust max br : Yes (total: 2754, max avg offset: 300 kbps)
- Clean previous : Yes
- Run mode : Normal
--------------------------------------------------------
- Note: EclCCE is old (< v1.8). Please upgrade to latest version
- Waiting for CCE window ...
- Detected a CCE encode window
- Detected a CCE encode window
- Detected a CCE encode window
- Detected the CCE main movie encode window - closing it
--------------------------------------------------------
- 2004-04-04 21:14:36
- Finished shutdown of CCE and D2S
- Avisynth setup: FACAR
- Configured cond. filter: None
- Mpv file: C:\MovieRip\Encoded_Video_CCE_NTSC.mpv
- Avs file: C:\MovieRip\AviSynth_Script_file.avs
- Using CCE version 2.67 (2.67.0.23)
--------------------------------------------------------
- Movie length : 01:33:49 (134965 frames, 23.976 fps)
- D2S audio select : 128 + 0 kbps
- D2S video calc. : 2179 kbps, 2 CDs
- Select Ranges : every 600, select 12 frames
- Sample frames : 2700
--------------------------------------------------------
Calculations:
- 1 CD: video_br=1018 (1018) audio_br=128+0 video_sz=716311989 fill=100% cbr=0
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- Number of CDs : 1
- Target mpv BR : 1018 (max 2626) kbps, size=716311989 bytes
--------------------------------------------------------
Search for Q:
- Sample encode : Q=39: 793 kbps, err=-22.1%, size=558386995, sample sz=11170636
- Sample encode : Q=25: 1008 kbps, err=-1%, size=709554794, sample sz=14194776
- Sample encode : Q=24: 1032 kbps, err=1.4%, size=726618569, sample sz=14536140
- Switch to bin search
- Sample encode : Q=0: 2322 kbps, err=128.1%, size=1634022255, sample sz=32688920
- Determined Q : 1 = Round(0 + (128.1 - 1.0)/-1.#IND)
--------------------------------------------------------
- 2004-04-04 21:34:30
- Start movie OPV encoding (Q 1)
- OPV pass result: 230.6% on target (1652029524 / 716311989) Speed factor: 0.86
--------------------------------------------------------
- 2004-04-04 23:24:12
- Start transcoding (Requant 43.36%)
- Transcoding pass: 100.5% on target (719566173 / 716311989) Speed factor: 32.28
--------------------------------------------------------
- 2004-04-04 23:27:08
- Executing: C:\Program Files\DVD2SVCD\BeSweet\BeSweet.exe -core( -input "C:\MovieRip\Extracted_audio_1.ac3" -output "C:\MovieRip\Encoded_audio_1.mp2" -logfile "C:\MovieRip\Encoded_audio_1.log" ) -azid( -L -3db -C 0db -S 0db -c normal -s surround -g max ) -2lame( -e -b 128 -m s )
--------------------------------------------------------
- 2004-04-04 23:37:07
- Recover DVD2SVCD
--------------------------------------------------------
- 2004-04-05 03:06:53
- D2SRoBa successfully finished
--------------------------------------------------------

Proper one, putting back DVD2SVCD's bitrate setting for 1 CD;

--------------------------------------------------------
D2SRoBa v3.4.2, by Tylo
--------------------------------------------------------
- WIN_XP - AutoIt 3.0.101.0
- 2004-04-05 14:50:34
--------------------------------------------------------
- Output type : SVCD
- CD size : 800
- Number of CD's : 1
- Sample percent : 2.0
- Adjust % : 1.0
- Cond. sizing pass : Yes (Transcode if opv sz > 0.1% oversize, and Q <= 60)
(Encode if opv sz < 2.5% undersize, or Q > 60)
- Audio after video : Yes (9% of video, min 128 kbps)
- Adjust max br : Yes (total: 2754, max avg offset: 300 kbps)
- Clean previous : Yes
- Run mode : Normal
--------------------------------------------------------
- Waiting for CCE window ...
- Detected a CCE encode window
- Detected the CCE main movie encode window - closing it
--------------------------------------------------------
- 2004-04-05 15:00:05
- Finished shutdown of CCE and D2S
- Avisynth setup: FACAR
- Configured cond. filter: None
- Mpv file: C:\MovieRip\Encoded_Video_CCE_NTSC.mpv
- Avs file: C:\MovieRip\AviSynth_Script_file.avs
- Using CCE version 2.67 (2.67.0.23)
--------------------------------------------------------
- Movie length : 01:33:49 (134965 frames, 23.976 fps)
- D2S audio select : 128 + 0 kbps
- D2S video calc. : 1019 kbps, 1 CDs
- Select Ranges : every 600, select 12 frames
- Sample frames : 2700
--------------------------------------------------------
Calculations:
- 1 CD: video_br=1019 (1019) audio_br=128+0 video_sz=717015635 fill=100% cbr=0
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- Number of CDs : 1
- Target mpv BR : 1019 (max 2626) kbps, size=717015635 bytes
--------------------------------------------------------
Search for Q:
- Sample encode : Q=39: 746 kbps, err=-26.8%, size=525513920, sample sz=10513004
- Sample encode : Q=22: 1038 kbps, err=1.9%, size=731030424, sample sz=14624400
- Sample encode : Q=23: 1012 kbps, err=-0.7%, size=712132325, sample sz=14246340
- Determined Q : 22 = Round(23 + (-0.7 - 1.0)/2.6)
--------------------------------------------------------
- 2004-04-05 15:11:49
- Start movie OPV encoding (Q 22)
- OPV pass result: 100.1% on target (717523776 / 717015635) Speed factor: 0.75
- Oversized: Hoping to fit mpv on selected num CDs
--------------------------------------------------------
- 2004-04-05 17:16:15
- Executing: C:\Program Files\DVD2SVCD\BeSweet\BeSweet.exe -core( -input "C:\MovieRip\Extracted_audio_1.ac3" -output "C:\MovieRip\Encoded_audio_1.mp2" -logfile "C:\MovieRip\Encoded_audio_1.log" ) -azid( -L -3db -C 0db -S 0db -c normal -s surround -g max ) -2lame( -e -b 128 -m s )
--------------------------------------------------------
- 2004-04-05 17:26:13
- Recover DVD2SVCD
--------------------------------------------------------
- 2004-04-05 17:47:41
- D2SRoBa successfully finished
--------------------------------------------------------

[tylo]

@DDogg, thanks for the advice. I think that is the cause of a lot of errors reported on this board. The reason I pursued NuroPreaks error, is that SiliconSoul reported the exact same problem (in basic topics). Still, it may very well be CCE that did it, because I cannot explain it from the source code.

@Holomatrix - thanks. It has nothing to do with bitrate tab settings. Just leave them as it is. You will notice D2SRoBa computed bitrate 1018 for 1 CD (only 1 off from the 'proper one'). The problem is the 'Switch to bin search'. That is still bugged. This should be the very last (major) bug in the plugin.

/PS: Odd thing is that your samples (e.g. Q39) are slightly different in size for the two runs, although all your settings seems equal...? (Did you make any changes in the FACAR setup inbetween?)

[Holomatrix]

Ok, great, what's the new/fixed ETA release? and yes I changed some FACAR filter setings. I was using a basic script (Undot, Flux, Blockbuster) then moved to the QMF method, which looks great :)

P.S. Great progress on D2Sroba :)

[NuroPreak]

Originally posted by DDogg
tylo, I would sure feel better if we knew if NuroPreak's machine will run the Prime95 torture test (http://mersenne.org/gimps/p95v238.exe) for 4 hours without error. In fact, may I suggest than anybody that wishes to report a one off error do this test before posting. For that matter, please throw in one full cycle of memtest (http://www.memtest86.com/)

The main reason for this is that CCE, when it encounters a missed beat of the CPU, will not crash like many programs. Rather, it just drops out <poof and vanish>. D2S or D2SRoBa interprets this to mean the encoding cycle is finished as the system will keep working with no error thrown. BeSweet, on the other hand can do the same thing which causes the same problem during the audio encoding cycle, or sometimes it will present a black screen lock-up. A black screen with BeSweet is 99% CPU overheat or an unstable overclock although bad ram can do the same thing.

Btw, many people are presently running with bad ram and don't know it. I've seen this on 5 machines this year. The mpeg2dec/avisynth/CCE combination will use an extra 100 megs of memory. That is why ram problems tend to show up during encoding. Running memtest for one complete cycle will tell you one way or the other.

I tested D2SRoba + DVD2SVCD on 2 machines.. 1x pentium 4x 2.6gz and 1x at duron 1gz both i encounter the same problem.

but i could do both tests..

[NuroPreak]

Ok first test done.. on the fast computer *2.6gz 512 mb ram. with CCE 2.66 i get same error again.. :( [still 1 cd results + a d2sroba.avs from 12 mb

second test is still busy but its @ the slow pc with CCE 2.5.

--------------------------------------------------------
D2SRoBa v3.4.3, by Tylo
--------------------------------------------------------
- WIN_2000 - AutoIt 3.0.101.0
- 2004-04-06 07:42:41
--------------------------------------------------------
- Output type : SVCD
- CD size : 800
- Number of CD's : auto
- Sample percent : 1.0
- Auto Q / Max Q : 35 / 60
- Adjust % : 1.0
- Cond. sizing pass : Yes (Transcode if opv sz > 0.5% oversize, and Q <= 40)
(Encode if opv sz < 2.0% undersize, or Q > 40)
- Cut last image : If opv sz > 0.4% oversize, and opv sz < 0.5% oversize
- Audio after video : Yes (9% of video, min 224 kbps)
- Cond. filtering : If avisynth configured, and estim. oversize < 8% for Num CDs-1
- Adjust max br : Yes (total: 2754, max avg offset: 300 kbps)
- Clean previous : Yes
- Run mode : Normal
--------------------------------------------------------
- Waiting for CCE window ...
- Detected a CCE encode window
- Detected the CCE main movie encode window - closing it
--------------------------------------------------------
- 2004-04-06 08:14:54
- Finished shutdown of CCE and D2S
- Avisynth setup: BicubicResize
- Configured cond. filter: None
- Using CCE version 2.67 (2.66.1.7)
--------------------------------------------------------
- Movie length : 02:00:02 (180057 frames, 25 fps)
- D2S audio select : 224 + 0 kbps
- D2S video calc. : 1740 kbps, 3 CDs
- Select Ranges : every 1500, select 15 frames
- Sample frames : 1815
--------------------------------------------------------
Calculations:
- 1 CD: video_br=420 (420) audio_br=224+0 video_sz=378119700 fill=100% cbr=0
- 2 CD: video_br=1080 (1080) audio_br=224+0 video_sz=972307800 fill=100% cbr=0
- 3 CD: video_br=1740 (1740) audio_br=224+0 video_sz=1566495900 fill=100% cbr=0
- 4 CD: video_br=2400 (2400) audio_br=224+0 video_sz=2160684000 fill=100% cbr=1
--------------------------------------------------------
Computing target number of CDs:
- Sample max BR : 2530 (adjusted for 3 CDs)
- Sample encode : Q=35: 53 kbps, err=?%, size=47873535, sample sz=482572
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- Using num CDs : 1
- Target mpv BR : 420 (max 2530) kbps, size=378119700 bytes
--------------------------------------------------------
Search for Q:
- Sample encode : Q=6: 91 kbps, err=-78.3%, size=82667492, sample sz=833300
- Sample encode : Q=1: 110 kbps, err=-73.8%, size=99608524, sample sz=1004068
- Determined Q : 1 = Round(1 + (-73.8 - 1.0)/0.9)
--------------------------------------------------------
- 2004-04-06 08:15:07
- Start movie OPV encoding (Q 1)
- OPV pass result: 514.9% on target (1947024860 / 378119700) Speed factor: 0.93
--------------------------------------------------------
- 2004-04-06 10:24:43
- Start transcoding (Requant 19.42%)
- Transcoding pass: 120% on target (453634771 / 378119700) Speed factor: 35.46
--------------------------------------------------------
- 2004-04-06 10:28:07
- Executing: C:\Program Files\SVCD2AVI\BeSweet\BeSweet.exe -core( -input "C:\Rippen\SVCD\Audio\Extracted_audio_1.ac3" -output "C:\Rippen\SVCD\Audio\Encoded_audio_1.mp2" -logfile "C:\Rippen\SVCD\Audio\Encoded_audio_1.log" ) -azid( -L -3db -C 0db -S 0db -c normal -s surround -g max ) -ota( -r 960 1001 ) -shibatch( --rate 44100 ) -2lame( -e -b 224 -m s )
--------------------------------------------------------
- 2004-04-06 10:46:33
- Recover DVD2SVCD
--------------------------------------------------------
- 2004-04-06 16:02:39
- D2SRoBa successfully finished
--------------------------------------------------------

[tylo]

After CCE and D2S is closed down, the only thing it does is to copy the Avisynth_Script_file.avs to D2SRoBa.avs and add the "SelectRangeEvery(..)" line to it.

The Avisynth_Script_file.avs is generated by D2S, and should only be a few Kb text file. Is this file 12Mb too (i.e. corrupted)? In that case, the file coruption probably happens during close down of CCE/D2S. As I mentioned, you should add 'debug=1' in your D2SRoBa.ini after installation, so it prints some extra debug lines.

/add: if the failure happes with both CCE 2.67, 2.66 and 2.50, its most likely D2S that is causing the corrupted file. I presume you are using v121b3. Not likely, but it could be AutoIt3 which is doing something bad during file copy on your machine.

[Holomatrix]

Does 344 fix the 'Switch to bin search' problem?

[DDogg]

NuroPreak, what happens if you do a standard D2S multi-pass without using D2SRoBa? We would be interested to know if it completes correctly.

[maSp]

[00:00:00:000] +------- BeSweet -----
[00:00:00:000] | Input : G:\Movie\Extracted_audio_1.ac3
[00:00:00:000] | Output: G:\Movie\Encoded_audio_1.ac3
[00:00:00:000] +------- AC3ENC ------
[00:00:00:000] | Bitrate method : CBR
[00:00:00:000] | AC3 bitrate : 448
[00:00:00:000] | Channels Mode : 5.1
[00:00:00:000] | Error Protection: Yes
[00:00:00:000] +---------------------
[ 417225984] Conversion Completed !
[00:01:13:000] <-- Transcoding Duration

--------------------------------------------------------
D2SRoBa v3.4.4, by Tylo
--------------------------------------------------------
- WIN_XP - AutoIt 3.0.101.0
- 2004-04-06 19:18:55
--------------------------------------------------------
- Output type : DVD
- CD size : 4350
- Number of CD's : 1
- Sample percent : 2.5
- Adjust % : 1.0
- Cond. sizing pass : Yes (Encode if opv sz < -0.0%, or > 0.0% oversize, or Q > 25)
- Clean previous : Yes
- Run mode : Normal
--------------------------------------------------------
- Waiting for CCE window ...
- Detected a CCE encode window
- Detected the CCE main movie encode window - closing it
--------------------------------------------------------
- 2004-04-06 19:29:26
- Finished shutdown of CCE and D2S
- Avisynth setup: FACAR
- Configured cond. filter: None
- Using CCE version 2.67 (2.66.1.7)
--------------------------------------------------------
- Movie length : 02:09:41 (194539 frames, 25 fps)
- D2S audio select : 429 + 0 kbps
- D2S video calc. : 4131 kbps, 1 CDs
- Select Ranges : every 480, select 12 frames
- Sample frames : 4872
--------------------------------------------------------

Can someone enlighten me please?

[NuroPreak]

Originally posted by DDogg
NuroPreak, what happens if you do a standard D2S multi-pass without using D2SRoBa? We would be interested to know if it completes correctly.

Yes that works correctly i used to do 4pass with D2S on both machines with cce 2.66 and 2.5 no problems at all..

anywy i installed 3.44 now and i put debug on ill post the results tomorrow.. but the first lines look better then before except the d2sroba.avs is still 13 mb or something :( but atleast it does some testings now and its not using Q=1 but he's doing Q=18.


--------------------------------------------------------
D2SRoBa v3.4.4, by Tylo
--------------------------------------------------------
- WIN_XP - AutoIt 3.0.101.0
- 2004-04-06 19:17:32
--------------------------------------------------------
- Output type : SVCD
- CD size : 800
- Number of CD's : auto
- Sample percent : 1.0
- Auto Q / Max Q : 35 / 60
- Adjust % : 1.0
- Cond. sizing pass : Yes (Transcode if opv sz > 0.5% oversize, and Q <= 40)
(Encode if opv sz < 2.0% undersize, or Q > 40)
- Cut last image : If opv sz > 0.4% oversize, and opv sz < 0.5% oversize
- Audio after video : Yes (10% of video, min 224 kbps)
- Cond. filtering : If avisynth configured, and estim. oversize < 8% for Num CDs-1
- Adjust max br : Yes (total: 2754, max avg offset: 300 kbps)
- Clean previous : Yes
- Run mode : Normal
--------------------------------------------------------
- Waiting for CCE window ...
- Detected a CCE encode window
- Detected the CCE main movie encode window - closing it
--------------------------------------------------------
- 2004-04-06 19:40:52
- Finished shutdown of CCE and D2S
- Avisynth setup: BicubicResize
- Configured cond. filter: None
- DEBUG: Mpv file: E:\RIPPEN\Video\Encoded_Video_CCE_PAL.mpv
- DEBUG: Avs file: E:\RIPPEN\Video\AviSynth_Script_file_avs-vfapi.avi (13326336 bytes)
- Using CCE version 2.50 (2.50.1.0)
--------------------------------------------------------
- Movie length : 02:00:02 (180057 frames, 25 fps)
- D2S audio select : 224 + 0 kbps
- D2S video calc. : 1740 kbps, 3 CDs
- Select Ranges : every 1500, select 15 frames
- DEBUG: Sample avs: E:\RIPPEN\Video\D2SRoBa.avs (13326363 bytes)
- Sample frames : 1815
--------------------------------------------------------
Calculations:
- 1 CD: video_br=420 (420) audio_br=224+0 video_sz=378119700 fill=100% cbr=0
- 2 CD: video_br=1080 (1080) audio_br=224+0 video_sz=972307800 fill=100% cbr=0
- 3 CD: video_br=1740 (1740) audio_br=224+0 video_sz=1566495900 fill=100% cbr=0
- 4 CD: video_br=2400 (2400) audio_br=224+0 video_sz=2160684000 fill=100% cbr=1
--------------------------------------------------------
Computing target number of CDs:
- Sample max BR : 2530 (adjusted for 3 CDs)
- Sample encode : Q=35: 1161 kbps, err=?%, size=1045878197, sample sz=10542600
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- Using num CDs : 3
- Target mpv BR : 1740 (max 2530) kbps, size=1566495900 bytes
--------------------------------------------------------
Search for Q:
- Sample encode : Q=16: 1902 kbps, err=9.3%, size=1713097020, sample sz=17268260
- Sample encode : Q=20: 1641 kbps, err=-5.7%, size=1477491592, sample sz=14893324
- Sample encode : Q=18: 1775 kbps, err=2%, size=1598800209, sample sz=16116132
- Sample compute : Q=19: 1708 kbps, err=-1.8%, using (br(Q18) + br(Q20))/2
- Determined Q : 18 = Round(19 + (-1.8 - 1.0)/3.9)
--------------------------------------------------------
- 2004-04-06 19:57:37
- Start movie OPV encoding (Q 18)

[DDogg]

Yes that works correctly i used to do 4pass with D2S on both machines with cce 2.66 and 2.5 no problems at all.. Yes, but I was talking about now, not in the past.

[NuroPreak]

Originally posted by DDogg
Yes, but I was talking about now, not in the past.

I ment with that i tried it last week, and a couple a days back. So i was already busy with d2sroba aswell.

[tylo]

Does 344 fix the 'Switch to bin search' problem?Yes.

@maSp: D2SRoBa looks at the size of the G:\Movie\Encoded_audio_1.ac3 and the number of frames and thereby computes the framerate. That number is not critical for any computations (info only). Maybe I use wrong number of frames...

@NuroPreak: This is mysterious:
DEBUG: Avs file: E:\RIPPEN\Video\AviSynth_Script_file_avs-vfapi.avi (13326336 bytes)
How come "AviSynth_Script_file_avs-vfapi.avi" be the avs file? Isn't that created by the vfapi tool in connection with the DVD2AVI tool? Anyone. The odd thing is that your current run seems fine.

Please, rar together all the project files, and PM me a link to it:

dvd2svcd project file.d2s
DVD2SVCD.INI (in project dir)
SVCD.INI
CCE_Project_file.ecl
CCE_Project_file_MOVIE.ecl
Avisynth_Script_file.avs

I will look at it after easter sometimes.

[NuroPreak]

Ill zip it up and upload it as soon as it's finished :D

[bobwillis]

Hi NuroPreak,

Just a thought - Do you run CCE frameserved? See if safemode (frameserving) is unchecked on the D2S Encoder tab. Most people run without frameserving (box unchecked). I wonder if that is creating the avs-VFAPI.avi file?

Regards,
Bob

[maSp]

Maybe I use wrong number of frames...

Thanks Tylo. It was my fault. I was testing the latest BeSweet release and forgot to change back.

[DDogg]

DEBUG: Avs file: E:\RIPPEN\Video\AviSynth_Script_file_avs-vfapi.avi (13326336 bytes) Just as Bob said, when using CCE in the encoder tab, you would only have this file created IF safe-mode has been turned on. I forgot that was even there anymore.

/add:hmm, vfapi always wraps the audio to the signpost AVI. In this case we have the resampleaudio line in the avisynth script creating a small false audio. The resulting vfapi.avi would reflect the size of the audio.

God, what a comedy of odd events - Yep, I turned on safe mode and saw what was happening. tylo, I assume you copy the INI referenced file "AviSynth_Script_file_avs-vfapi.avi" as D2SRoBa.avs. D2SRoBa.avs, even though it is really a renamed 9 meg vfapi file, will still allow your edits, and open in cce via d2sroba.

[NuroPreak]

Whoei everything worked out for the first time since weeks i ended up having nice quality and 3cds.. ;)

I got 2 more questions.. as ddogg and bobwillis said.. should i turn of safe frameserving?

and the other question is i have 2 cds completelty filled but the third cd is only 126mb and i want all cds filled.. so it had to create 3 cds from 800mb or 2cds from 800mb can i fix this by setting the conditional sizing pass the limites to 0%, i can remember that i read it somewhere but i donno where anymore

//edit
Q. How can I make sure my CDs are always filled?
If you want to stretch your CDs fully at the cost of doing always 2 passes, check conditional sizing pass and set 0-0 in the thresholds.


//edit: tylo please clean your pm mailbox because i can't send you a pm :D

tnx alot!! :)

[DDogg]

NuroPreak, if it worked it was a complete accident (AFAIK). Yes, please never use safe mode with d2sroba. In fact, I hope tylo can check for it and do a stop process warning.

[NuroPreak]

Originally posted by DDogg
NuroPreak, if it worked it was a complete accident (AFAIK). Yes, please never use safe mode with d2sroba. In fact, I hope tylo can check for it and do a stop process warning.

without safe mode it works perfect!! now i enjoy the pleasure of d2sroba.. ripping my stuff in 3 hours or something instead of 10 / 12 hours :D also the d2sroba.avs looks ok now :D

i got 1 more question.. how come if i set

- Cond. sizing pass : Yes (Transcode if opv sz > 0% oversize, and Q <= 40) (Encode if opv sz < 0% undersize, or Q > 40)

both at 0% i ended up with a 2cd movie with cd2 only filled till 426mb? Why doesn't d2sroba choose a lower Q then?


--------------------------------------------------------
D2SRoBa v3.4.4, by Tylo
--------------------------------------------------------
- WIN_XP - AutoIt 3.0.101.0
- 2004-04-13 00:01:23
--------------------------------------------------------
- Output type : SVCD
- CD size : 800
- Number of CD's : auto
- Sample percent : 1.0
- Auto Q / Max Q : 35 / 60
- Adjust % : 1.0
- Cond. sizing pass : Yes (Transcode if opv sz > 0% oversize, and Q <= 40)
(Encode if opv sz < 0% undersize, or Q > 40)
- Audio after video : Yes (10% of video, min 224 kbps)
- Cond. filtering : If avisynth configured, and estim. oversize < 8% for Num CDs-1
- Adjust max br : Yes (total: 2754, max avg offset: 300 kbps)
- Clean previous : Yes
- Run mode : Normal
--------------------------------------------------------
- Waiting for CCE window ...
- Detected a CCE encode window
- Detected the CCE main movie encode window - closing it
--------------------------------------------------------
- 2004-04-13 00:12:19
- Finished shutdown of CCE and D2S
- Avisynth setup: BicubicResize
- Configured cond. filter: None
- DEBUG: Mpv file: E:\RIPPEN\Video\Encoded_Video_CCE_PAL.mpv
- DEBUG: Avs file: E:\RIPPEN\Video\AviSynth_Script_file.avs (442 bytes)
- Using CCE version 2.50 (2.50.1.0)
--------------------------------------------------------
- Movie length : 01:25:38 (128462 frames, 25 fps)
- D2S audio select : 224 + 0 kbps
- D2S video calc. : 1740 kbps, 2 CDs
- Select Ranges : every 1500, select 15 frames
- DEBUG: Sample avs: E:\RIPPEN\Video\D2SRoBa.avs (469 bytes)
- Sample frames : 1290
--------------------------------------------------------
Calculations:
- 1 CD: video_br=750 (750) audio_br=224+0 video_sz=481732500 fill=100% cbr=0
- 2 CD: video_br=1740 (1740) audio_br=224+0 video_sz=1117619400 fill=100% cbr=0
- 3 CD: video_br=2530 (2730) audio_br=224+0 video_sz=1625044300 fill=93.3% cbr=1
--------------------------------------------------------
Computing target number of CDs:
- Sample max BR : 2530 (adjusted for 2 CDs)
- Sample encode : Q=35: 1482 kbps, err=?%, size=952477416, sample sz=9564664
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- Using num CDs : 2
- Target mpv BR : 1740 (max 2530) kbps, size=1117619400 bytes
--------------------------------------------------------
Search for Q:
- Sample encode : Q=27: 1721 kbps, err=-1.1%, size=1105498562, sample sz=11101284
- Sample encode : Q=26: 1755 kbps, err=0.9%, size=1127411590, sample sz=11321332
- Determined Q : 26 = Round(26 + (0.9 - 1.0)/2)
--------------------------------------------------------
- 2004-04-13 00:20:38
- Start movie OPV encoding (Q 26)
- OPV pass result: 100.9% on target (1127133652 / 1117619400) Speed factor: 0.56
--------------------------------------------------------
- 2004-04-13 02:53:13
- Start transcoding (Requant 99.16%)
- Transcoding pass: 99% on target (1106604997 / 1117619400) Speed factor: 44.65
--------------------------------------------------------
- 2004-04-13 02:55:10
- Executing: C:\Program Files\AVI2SVCD\BeSweet\BeSweet.exe -core( -input "E:\RIPPEN\Audio\Encoded_audio_1.mp2.wav" -output "E:\RIPPEN\Audio\Encoded_audio_1.mp2" -logfile "E:\RIPPEN\Audio\Encoded_audio_1.log" ) -ota( -g max ) -shibatch( --rate 44100 ) -2lame( -e -b 224 -m s )
--------------------------------------------------------
- 2004-04-13 03:21:59
- Recover DVD2SVCD
--------------------------------------------------------
- 2004-04-13 08:56:55
- D2SRoBa successfully finished
--------------------------------------------------------

[tylo]

- Movie length : 01:25:38 (128462 frames, 25 fps)
- D2S audio select : 224 + 0 kbps
- D2S video calc. : 1740 kbps, 2 CDs

These numbers are taken directly from D2S project files. Unfortunately, sometimes it seems D2S either calculates wrong bitrate (in that case it will also go wrong without the plugin), or simply a wrong bitrate is written to the file. 1740kbps is way to low for this short movie on 2 CDs. Try again fresh and see if you get the same result.

[DDogg]

Note updated version, See first page of this post.

"v3.4.6 - hopefully the final version"