0.55b oversize 1 step CCE OPV

[Harm]

1. DVD-RB Version: 0.55b
2. Encoder being used: CCE SP 2.67.00.23
3. Using eclCCE? Yes
4. DVD: Mummy returns
5. Bug encountered: Oversize on OPV

Total created size: 4,44 GB (4.772.220.928 bytes)

Can I supply additional information? (I saved the status messages)

[ShadowKnight]

do you have anything set for "ccetargetsectors="? I have "CCETargetSectors=2260007". I've only run one OPV so far with .55b and it came out at 4.30. I havent tried it with default "ccetargetsector" settings. The same disc, re-authored with DVDReMake beforehand, came out to 4.18 and the difference is pretty much exactly what I removed. So obviously Re-Authoring can affect sizing output, but seems to make it come out smaller.
I'd go ahead and post the settings your using like, steal space, filters, half-d1, ect, ect.. and it wouldnt hurt to post the log from the Prepare phase at least.

[Harm]

Haven't fiddled with special parms....

Prepare status messages are:
[07:57:50] Phase I, PREPARATION started.
- "Steal Space from Extras" mode is enabled.
- "One Pass VBR (w/analysis)" mode is enabled.
- VTS_01: 3.526.305 sectors.
-- Scanning and writing .D2V file
-- Processed 186.316 frames.
-- Building .AVS and .ECL files
- Reduction Level for DVD-5: 65,7%
- Overall Bitrate : 3.873Kbs
- Space for Video : 3.523.762KB
- Analyzing VTS_01 for optimal Q factor.
- TargetSize (sectors): 1783024
-- Predicted size (sectors) at Q=27: 1438239
-- Predicted size (sectors) at Q=20: 1751471
- Q Value selected: 20
- Movie improvement from extra reduction = ,0%
- HIGH/LOW/AVERAGE Cell Bitrates: 4.668/410/3.873 Kbs
[08:53:25] Phase I, PREPARATION completed in 56 minutes.

Hope this helps.....

Also took a look with DVD shrink at the output of rebuilder and there was some unreferenced material reported. Could this have something to do with it?

[ShadowKnight]

unreferenced materiual prolly would make that difference... I've noticed that DVD-RB gets the information from the ifos and whatnot, but not the actuall vobs themselves (I do believe)

To correct my earlier statement, my dvd came out to 4.30 without re-authoring it, and 4.18 AFTER re-authoring it. Which to me means that neither DVD-RB nor ReMake properly figure the ACTUAL DVD, only the information provided, which is obviously not accurate in the case of unreferenced material, and with re-authored disks. At least this is what I think the problem is here.
DVD-ReMake needs to fix the references better after a re-author, so that re-builder can make up for the extra space availible. as far as my problem goes.

As for the oversizing problem, it seems that DVD-RB needs to not be so trusting of the "referenced material" and maybe have an expert option to scan the "real files themselves". This is prolly a royal pain in the butt, but it would be well woth it when you consider that it could also fix the ifo files for rebuilding, I think it should anyways, in order to remove the references to the removed audio and subtitle streams. I may have this all wrong, and am talking out of my ass, and if so I apologize, but its an explanation that explains why I always get undersized by about as much as I remove when I re-author first.

[Harm]

Unreferenced material is 98 Mb in size......

could this be the reason jdobbs?????

[jdobbs]

Originally posted by Harm
Unreferenced material is 98 Mb in size......

could this be the reason jdobbs????? No. DVD-RB gets its sizing information from the VOB files as it scans them, not the IFOs.

Unfortunately this is the risk and nature of using OPV mode. You sacrifice output sizing accuracy in exchange for encoding time. The output size is based upon a scanning/prediction mechanism and usually is pretty good -- but not always. Certain sources can be harder to predict.

You can get more accurate by changing the prediction sampling percentage (which can be set with a variable in the INI file, see the Rebuilder.txt file) -- but you pay for that increase in increased encoding time.

[jdobbs]

Also, looking at the information that was posted it looks like you've enabled "steal space from extras" on a disc that has no extras (at least none that have a VTS large enought for DVD-RB to consider them significant). As a result you've gotten a .0% improvement through extra reduction.

[Harm]

Originally posted by jdobbs
You can get more accurate by changing the prediction sampling percentage (which can be set with a variable in the INI file, see the Rebuilder.txt file) -- but you pay for that increase in increased encoding time.
Is there something to say about how much increase I can expect?

Further I conclude that you say that this is just the risk of using that option? Obviously that is no problem as long as I know how to get this project beneath the size mark....... :)

[jdobbs]

Originally posted by Harm
Is there something to say about how much increase I can expect?

Further I conclude that you say that this is just the risk of using that option? Obviously that is no problem as long as I know how to get this project beneath the size mark....... :) Yes it is the risk associated with it. Maybe in the future I can get it more accurate. But as of today I find that about 1 out of 10 movies (usually one with a lot of extras) may be slightly oversized.

My intention is to build an option at some point that will allow you to do one more pass when your final output is over -- and bring it in under the wire.

[Harm]

Originally posted by jdobbs
Yes it is the risk associated with it. Maybe in the future I can get it more accurate. But as of today I find that about 1 out of 10 movies (usually one with a lot of extras) may be slightly oversized.

My intention is to build an option at some point that will allow you to do one more pass when your final output is over -- and bring it in under the wire.
Alright, fair enough.

All these plans for enhancements and so little time.......... ;)

Getting better and better....... :)

[jdobbs]

Originally posted by Harm
Alright, fair enough.

All these plans for enhancements and so little time.......... ;)

Getting better and better....... :) Exactly. Sometimes I wonder if I'll even get out of beta before the cost of dual-layered discs falls under $1...

[Sir Didymus]

When the cost of DVD +/-R dual layer disks will be less than 1$ I will use DVD-RB for making beautiful backups of my high definition blue ray DVDs (~27 GB) into the 9GB of the cheapo DL ...

:)

I really see no reason to feel pressed by this technology step...

By the way are all of the numbers related to the sizes already included in the form of parameters into the source code of DVD-RB, right ?

:scared:

Cheers,
SD

[jdobbs]

Originally posted by Sir Didymus
When the cost of DVD +/-R dual layer disks will be less than 1$ I will use DVD-RB for making beautiful backups of my high definition blue ray DVDs (~27 GB) into the 9GB of the cheapo DL ...

:)

I really see no reason to feel pressed by this technology step...

By the way are all of the numbers related to the sizes already included in the form of parameters into the source code of DVD-RB, right ?

:scared:

Cheers,
SD Pretty much... I was thinking about HD as a possibility also. ;)