Missing .m2v when encoding

[pg55555]

Joe_smish
Sorry to hear this. I have done more than 200 successful backups with RB, since version 0.46

I would recomend try other DVD before giving-up. Maybe your problem is DVD specific.
If not, it should be your setup.

[pg55555]

eriksen76

Look at Rockas site (www.dvd-rb.com), he has all the old public versions available for download

[eriksen76]

Thanx alot, didn't know I could get the old versions there

I will try with version 0.79 whick seems to be the version before the m2v file check introduced in 0.80

Will let you know how it went

By the way a small question. I have always placed the dgdecode file in the plugin directory of avisynth. Does it make any diffrences to place it in dvdrebulder directory? Just a thought...

By the way I use the right version

/Eriksen76

[lark]

should not make any difference as long as the path is correct (i.e. the loadplugin statement in the AVS files points to the correct dll).

edit: and if its in the avisynth plugin directory, you don't even need the loadplugin statement.

regards
t :)

[SpazzHH]

Originally posted by eriksen76
I have always placed the dgdecode file in the plugin directory of avisynth. Does it make any diffrences to place it in dvdrebulder directory? Just a thought...

By the way I use the right version

/Eriksen76


jdobbs suggests a seperate copy of the DGDecode be kept in the DVD-RB directory to keep some other program from accidently upgrading the .dll to a version that would not be compatable with RB. If the .dll is in the avisynth directory they could easily do that without someone realizing it. Safety precaution with a VERY important file.

[eriksen76]

So what you're saying is only of copy of Dgdecode, which is to be placed in the DVDRB directory.

None in avisynth directory?

By the way right now I'm doing a 2 pass test on the same movie, but this time with Dvdrb 0.79

Will let you know how it went

/Eriksen76

[SpazzHH]

From what I understand by reading other posts on this board, that is his preference. That is why the installer puts it there instead of the Avisynth directory.

[jdobbs]

I'd like to make this perfectly clear. The "missing .M2V file" bug is NOT RELATED TO DVD-RB. It is caused by Version 2.70 of CCE which I believe to be too buggy to use. In every test I performed on it the files were created properly in earlier versions of CCE (both SP and Basic) using the exact same .ECL files. Others have done the same.

I implemented a workaround in DVD-RB Freeware v0.82.1 and Pro v0.84.1 in order to make it work -- but that doesn't mean I am willing to accept the blame for an error that is clearly with Cinemacraft encoder.

If you want your .M2V files to show up use one of these versions of DVD-RB, or better yet, use a stable version of CCE or one of the other encoders (ProCoder/QuEnc/HC) for encoding.

[jdobbs]

Originally posted by eriksen76
I just did a a movie from sractch!! - NO pre-processing (Latest RB)

"ERROR: The input file V05000100002001.avs is missing frames. Rebuild aborting

[file]
name=C:\\NOTTING HILL\D2VAVS\V05000100002001.avs
frame_first=0
frame_last=3175
encode_first=0
encode_last=3175

Something strange has happend with the latest versons of DVDRB. I never before have had any of the explained errors when using this program. Something has got to be fixed ;)

/Eriksen76 That is exactly the problem I have mentioned. Use a stable version of CCE. Go post this on their forum!!!

[eriksen76]

Sorry Jdobbs I have not read this anywhere

Will try with a more stable version.

Thanx for pointing me this out, will let you know how further testing went.

Which CCE is the most stable according to you?

EDITED: Just for the fun of it I tried rebuilding the above mentioned project in "3 click mode" with DVDRB 0.79 and so far it has not given me any errors during rebuild??? It has successfully run through the problematic VTS mentioned above.

/Eriksen76

[jdobbs]

Originally posted by Joe_smish
I've never had a sucessful back up with this software yet. Preprocessing or not. I think its time to give up.:( The reason I've waited so long to respond to this thread is exactly this abrasive attitude...

If you install DVD-RB properly, don't preprocess, don't play with the sources, and let it do its job -- it works.

[pg55555]

DGDecode to be used by RB can be anywhere in your PC as long as the path point to it correctly.

The problem is that DGDecode is also used by other programs, which during installation can overwrite the one in the plugins folder, generating problems in RB working.

So the recomendation is to put the righ DGDecode version in a place to be used exclusivelly by RB, such as the RB folder, and point the Path to it in RB setup. So, you are free to have in the Avisynth plugins folder any other version those other programs require.

[jdobbs]

@eriksen76,

What DVD-RB version are you using? If you continue to have problems, we can work together to fix them. The only variable I can think of that has happened recently is the introduction of CCE v2.70 into the mix and I think that's the source of your headaches. If not, I'll get it fixed. But in the instances where I could repeat the missing M2V problem consistently, it was caused by overflow errors in CCE v2.70. I then simply pointed to an older version of CCE and ran the same job without error.

[jdobbs]

Originally posted by eriksen76
Sorry Jdobbs I have not read this anywhere

Will try with a more stable version.

Thanx for pointing me this out, will let you know how further testing went.

Which CCE is the most stable according to you?

EDITED: Just for the fun of it I tried rebuilding the above mentioned project in "3 click mode" with DVDRB 0.79 and so far it has not given me any errors during rebuild??? It has successfully run through the problematic VTS mentioned above.

/Eriksen76 CCE SP v2.50 or v2.67, CCE Basic v2.69. I'll go back to v0.79 and see if anything has changed that might affect this. Did you try redoing the entire job or just the rebuild? The problem with earlier versions of DVD-RB and CCE 2.70 is that they didn't have the "catch" in -- and an .M2V that was short would just be integrated as-is into the output stream and you might have had a "blank" or "frozen" section that only had sound. That's why I added that framecount check... it's better to not create a VTS than to have one that is in error but it is unknown.

[jdobbs]

Just as an example -- in doing Sex and the City, Season 5, Disc 2 (or was it 1?) -- several thousand frames were missing following the HBO "whitenoise" screen on the last episode.

In earlier versions I made the assumption that if the .M2V was created by CCE it was good. In doing that section the encoder bombed at the beginning of the segment during the first M2V pass, only created a few frames, and didn't display any error messages or pass them back in the process termination code.

It worked fine with any other version of CCE.

[eriksen76]

Originally posted by jdobbs
CCE SP v2.50 or v2.67, CCE Basic v2.69. I'll go back to v0.79 and see if anything has changed that might affect this. Did you try redoing the entire job or just the rebuild? The problem with earlier versions of DVD-RB and CCE 2.70 is that they didn't have the "catch" in -- and an .M2V that was short would just be integrated as-is into the output stream and you might have had a "blank" or "frozen" section that only had sound. That's why I added that framecount check... it's better to not create a VTS than to have one that is in error but it is unknown.

As VIP member I always use the latest version (0.84.1)

Right now I'm doing a full encode with DVDRB 0.84.1 & CCE 2.67 to see if that has solved the problem with the missing frames.

I was just a bit confused because when I suddenly had 2 movies in a row with 2 different errors. Never happend before.

First error was Notting Hill with the missing frames during rebuild (No errors during encode).

The other one is Seven Years In Tibet which I'm also doing some testing on now with the new setup. In order not to confuse too much I will await futher explanations about this problem until this test is done. But this one has problems during encode. It might be due to v2.70.02.00 so I wait a bit with details

B back

/Eriksen76

[jdobbs]

I really want to put an end to this missing .M2V and missing frames thing. If DVD-RB somehow contributes to it or causes it -- I need to know that. Otherwise I need to find a way to somehow prevent CCE from bombing during the encode...

If someone has a DVD that consistently has a missing .M2V -- could you try it with an earlier version of CCE and see if it repeats? Please report it here in this thread -- as this is my highest priority right now.

DVD-RB really shines when it uses CCE, and I want to make it work -- buggy version or not.

[TK8472]

Originally posted by jdobbs
I'd like to make this perfectly clear. The "missing .M2V file" bug is NOT RELATED TO DVD-RB. It is caused by Version 2.70 of CCE which I believe to be too buggy to use. In every test I performed on it the files were created properly in earlier versions of CCE (both SP and Basic) using the exact same .ECL files. Others have done the same.

I implemented a workaround in DVD-RB Freeware v0.82.1 and Pro v0.84.1 in order to make it work -- but that doesn't mean I am willing to accept the blame for an error that is clearly with Cinemacraft encoder.

If you want your .M2V files to show up use one of these versions of DVD-RB, or better yet, use a stable version of CCE or one of the other encoders (ProCoder/QuEnc/HC) for encoding.

In case it was overlooked in my posts, all my .M2V errors have been with CCE SP v2.67.00.27. I've never installed 2.70.

[jdobbs]

I'm making my assumption based upon my own experience.

Give me more details. The problem is that the check for missing .M2V files didn't go in until version .79 and .80. If they were missing before then you would have gotten some other cryptic error message during rebuild.

[jdobbs]

Does anyone think custom matrices might somehow have an influence on this? That's the only thing I can think of that was added recently that might change CCE processing...

[jdobbs]

Originally posted by eriksen76
As VIP member I always use the latest version (0.84.1)

Right now I'm doing a full encode with DVDRB 0.84.1 & CCE 2.67 to see if that has solved the problem with the missing frames.

I was just a bit confused because when I suddenly had 2 movies in a row with 2 different errors. Never happend before.

First error was Notting Hill with the missing frames during rebuild (No errors during encode).

The other one is Seven Years In Tibet which I'm also doing some testing on now with the new setup. In order not to confuse too much I will await futher explanations about this problem until this test is done. But this one has problems during encode. It might be due to v2.70.02.00 so I wait a bit with details

B back

/Eriksen76 Thanks. I'm eager to here your results.

[Fishman0919]

Does anyone think custom matrices might somehow have an influence on this? That's the only thing I can think of that was added recently that might change CCE processing

and turning off CCE's adaptive quantizer matrices feature by default

CCE 2.70 might be abit more pick with that turned off...then again adaptive quantizer matrices haves been in CCE 2.67 and up and is on by default

[eriksen76]

Originally posted by jdobbs
Thanks. I'm eager to here your results.

So far no problems with Seven Years In Tibet. Actually it has gone through the encode of the problematic .m2v file.

Seems to be working :) . Final Conclusion to come later though

/Eriksen76

[jdobbs]

Originally posted by Fishman0919
and turning off CCE's adaptive quantizer matrices feature by default

CCE 2.70 might be abit more pick with that turned off...then again adaptive quantizer matrices haves been in CCE 2.67 and up and is on by default That's true too. I wish I could repeat this problem in some way other than on one single disc. That way I could be sure I'm not chasing the wrong error.

[jdobbs]

Originally posted by eriksen76
So far no problems with Seven Years In Tibet. Actually it has gone through the encode of the problematic .m2v file.

Seems to be working :) . Final Conclusion to come later though

/Eriksen76 Well, that confirms my suspicions of v2.70 so far. I'll wait for the final result.

[jdobbs]

@TK8472,

Do you use custom matrices? Also, was the offending segment using a custom matrix?

[jdobbs]

@eriksen76

Can you tell me what the vbr_brate_avg, vbr_brate_min, and vbr_brate_max are on the offending segment?

[TK8472]

Originally posted by jdobbs
@TK8472,

Do you use custom matrices? Also, was the offending segment using a custom matrix?

No on both.

[Joe_smish]

Originally posted by jdobbs
The reason I've waited so long to respond to this thread is exactly this abrasive attitude...

If you install DVD-RB properly, don't preprocess, don't play with the sources, and let it do its job -- it works.

I'm pretty sure I did install it properly, and I didnt do any preprocessing....didnt work.:(
I'm sorry for an abrasive attitude, that wasnt my intentions. I was actually trying to have an optimistic outlook on this program. I've heard such great things about it and I really want to get it to work with me. Thats why I've been trying so hard to get it to work. But I'm not sure why it isnt working, b/c I installed it properly, I use the trial version of CCE 2.67, I rip the DVD and dont do anything to it. and I got the error of an .m2v missing. I'm not sure what I am doing wrong. I'm sure its me not the program. I really wish someone who used this went to Uconn and could give me a run through. That would be very helpful.

[Joe_smish]

Originally posted by jdobbs
I'd like to make this perfectly clear. The "missing .M2V file" bug is NOT RELATED TO DVD-RB. It is caused by Version 2.70 of CCE which I believe to be too buggy to use. In every test I performed on it the files were created properly in earlier versions of CCE (both SP and Basic) using the exact same .ECL files. Others have done the same.

I implemented a workaround in DVD-RB Freeware v0.82.1 and Pro v0.84.1 in order to make it work -- but that doesn't mean I am willing to accept the blame for an error that is clearly with Cinemacraft encoder.

If you want your .M2V files to show up use one of these versions of DVD-RB, or better yet, use a stable version of CCE or one of the other encoders (ProCoder/QuEnc/HC) for encoding.

I'm sorry but I wasnt blaming DVD-RB, I was just looking for a solution to the problem. I dont know anything about RB so I figured I'd ask and see IF it was a problem with RB, I wasnt blaming it. I dont know anything about CCE either, so for all I konw it could be a problem with that. I have no idea. Whats this matrices stuff people are talking about??

[jdobbs]

I just want this error to stop. If I can manipulate the input to CCE somehow (as I did in the last two versions for the error I'd received) I can get around it...

[jdobbs]

Originally posted by TK8472
No on both. Could you post your REBUILDER.INI file?

[Joe_smish]

Do you think if I had the pro version this would help with my problem? Or do you think it is a source problem, or a setup problem? I dont even know where to start trouble shooting, I feel like I've tried every thing.

[eriksen76]

Originally posted by jdobbs
@eriksen76

Can you tell me what the vbr_brate_avg, vbr_brate_min, and vbr_brate_max are on the offending segment?

Encoding not done yet. Will have a look at it when done.

It might be tomorrow, as its kinda late here in Europe

/Eriksen76

[jdobbs]

I just did some more testing on CCE V2.70. It is EXTREMELY sensitive to high demand scenes -- and using the HBO "whitenoise" intro screen as my baseline I've found that it is crashing with a buffer overflow any time the VBR maximum bitrate gets below about 8500Kbs -- which is a very high point to be crashing... for comparison, on CCE SP V2.67 you can get pretty far down into the low 7000s (on the same scene) before it has any problems.

As I pointed out in another post somewhere here, what the encoder should do when it runs out of headroom is adjust the quantizer scale to get under its maximum. It doesn't look like V2.70 is doing that (or at least not well).

Also, when V2.70 is called as a shell process or from the command line with the -batch switch on -- it doesn't display or make any indication of an error. V2.67 will kick up an error screen when it hits a problem.

[Joe_smish]

Do you have any suggestions that I can try in an attempt to solve the problem?

[jdobbs]

How many different DVDs have you tried? It's always best to do your first encode with an easy one just to make sure your configuration is right. After it's properly set up it will pretty much run on its own from then on.

[jdobbs]

Originally posted by eriksen76
Encoding not done yet. Will have a look at it when done.

It might be tomorrow, as its kinda late here in Europe

/Eriksen76 You can open the .ECL file while it is running... and look at the values. :D Sorry. You probably want to go to bed.

[eriksen76]

Well, "Seven Years In Tibet" is completed now with absolutely no problems at all.

The error with CCE 2.70 and this movie was:

"ERROR: File VXXXXXXX shows completed, but no .M2V file exists. Aborted."

That happend about 40% during the encode before when using CCE 2.7

Now I managed to get it working with latest DVDRB 0.84.1 PRO and CCE 2.67 (Trial)

So I guess the conclusion to the above mentioned error is that it's also related to CCE 2.7

Still, I will report how things went with the "missing m2v file" issue. But I think it will work. (Still encoding..)

Still I find ot very strange that when doing the rebuild of the movie with 0.79 it managed to make it work.

So something must have been changed from 0.79 - 0.84.1 PRO that makes a difference.

To sum up. I did a "one click" on this movie with CCE 2.70 and DVDRB 0.84.1 pro. In the very beginning of the rebuild it gave me the error that m2v file was missing.

But when doing the rebuild only, with the older version of DVDRB then it worked for me.

/Eriksen76

[jdobbs]

That's because all you did was the rebuild. The missing or shortened file was just ignored... v0.79 trusted CCE a lot more than v0.84 does (that's why I added the changes I did). You would be very disappointed when you played it back and had a segment that was audio-only.

I'm liking HC and QuEnc more-and-more. They both can kick through the troublesome disc I'm testing without a hitch.

[eriksen76]

Ok, thanx for the explanation - Glad there was one :)

/Eriksen76

[jptheripper]

Im curious jdobbs, and have thought this awhile

either you can make the program hcenc only (i think i might get stoned forthat comment)

or is there the option to , if the m2v fails the errorcheck twice, to encode that single segment with hcenc or quenc? its better than failing to me. I realize the prepare stages are different, i wonder how long it would take to prepare 2x the avs's/ecls

-jp

[archaeo]

either you can make the program hcenc only (i think i might get stoned forthat comment)

no, no stoning necessary but..

I like HC a lot, but until the encode speed picks up (ie multithread) I just have to go with CCE for times sake.

[jdobbs]

Well... I'm currently testing a beta of HC that HANK315 sent me on my video computer right now -- and from what I've seen so far I may be moving into your mindset as far as recommendations go. It is faster. As for multithreading... I can make DVD-RB run more than one segment at a time, you know!!!

QuEnc is getting better too.

But why would someone want to pay for a commercial encoder that needs a freeware one to rescue it when it crashes? At least from my point of view the problem is with Cinema Craft and they need to fix it.

[archaeo]

and from what I've seen so far I may be moving into your mindset as far as recommendations go. It is faster. As for multithreading... I can make DVD-RB run more than one segment at a time, you know!!!

Hey that would be great on both counts...:D
So you can optimize RB even more for us dualies?
When could this occur??

[jdobbs]

We're getting pretty OT (I'm probably more guilty than anyone). I'll let you know when I look at it a little.

[archaeo]

sounds good ;)

[JohnG]

Originally posted by jdobbs
That's because all you did was the rebuild. The missing or shortened file was just ignored... v0.79 trusted CCE a lot more than v0.84 does (that's why I added the changes I did). You would be very disappointed when you played it back and had a segment that was audio-only.

So to clarify, if I do decide to use CCE 2.7 and the process completes, I am not in danger of finding out later during playback that there are missing frames? To echo your opinion, I would much rather have the process fail than find out as I am viewing the film.
Do you suggest that I (we) just stick with 2.5sp or HC/QuEnc until Cinema Craft decides to fix this problem?

[jptheripper]

Originally posted by jdobbs
Well... I'm currently testing a beta of HC that HANK315 sent me on my video computer right now -- and from what I've seen so far I may be moving into your mindset as far as recommendations go. It is faster. As for multithreading... I can make DVD-RB run more than one segment at a time, you know!!!

QuEnc is getting better too.

But why would someone want to pay for a commercial encoder that needs a freeware one to rescue it when it crashes? At least from my point of view the problem is with Cinema Craft and they need to fix it.


Is hc really that good? i may have to check it out, i have a super bad source dvd with low bitrate ~1800 after encoding that is driving me nuts

[TK8472]

Originally posted by jdobbs
Could you post your REBUILDER.INI file?

I'm at work, but will post it as soon as I get home.