DVD Rebuilder (CCE One-Click Beta v0.46) Comments and Suggestions

[MedicineMan]

About 3 pages before this, there's a mention about audio recoding. My test DVD is a music concert, where the audio tracks (LPCM) take more than 1,5Gb. So, even with DVD-Rebuilder, the image quality isn't that great.

We're always thinking about compressing the video, but there are sometimes where "compressing" the audio would be as useful.

Could you (maybe using BeSweet) allow us to convert LPCM tracks to AC3 tracks, and so use more space for video. I agree with you that maybe there's no point in compressing or downsampling audio, except im LPCM tracks.


Sincerely


MM

[jdobbs]

Originally posted by MedicineMan
About 3 pages before this, there's a mention about audio recoding. My test DVD is a music concert, where the audio tracks (LPCM) take more than 1,5Gb. So, even with DVD-Rebuilder, the image quality isn't that great.

We're always thinking about compressing the video, but there are sometimes where "compressing" the audio would be as useful.

Could you (maybe using BeSweet) allow us to convert LPCM tracks to AC3 tracks, and so use more space for video. I agree with you that maybe there's no point in compressing or downsampling audio, except im LPCM tracks.


Sincerely


MM I can see the utility in that. But LPCM seems to be so rare (except on music videos, I guess)... and this might be a considerable amount of work... I've added it to my list, but it is at the bottom.

Thanks.

[jdobbs]

NEW VERSION OF DVD-REBUILDER (v0.44)

Attached please find a new version of DVD-RB. This version includes a couple significant fixes -- including what may be the cause of the elusive chapter stutter, and the "Error code '5'" The changes/improvements included in this version are listed below.

Again a big thanks to the beta testers. They are, as a group, really keeping me on my toes.

- Found and corrected what may have been the cause of the "stutter" at chapter points on some players. The error corrected was related to an incorrect GOP timestamp being inserted in the first GOP of every cell. This may have also contributed to the reported display time showing incorrectly when rewinding through chapter points on some players. My thanks to RB (again) for his knowledge and insight.

- Modified the Disable "Interlaced" option. It is now set individually for each modifiable VTS. When this is set the source for the selected VTS will be treated as progressive -- meaning "interlaced=true" will not be added to the ConvertToYUY() line, Zig-Zag encoding will be used, and the progressive flag will be set.

- Fixed an error associated with Half-D1 that caused the .ECL files to to not correctly reflect the horizontal resolution as selected in the Half-D1 menu.

- Fixed Runtime Error '5' error that happened in REBUILD phase.

- Added a QuEnc Options item (why didn't I have one before?) and included two options "Trellis Quantization" and "KVCD Notch Matrix"

- Inserted code to prevent individual failed jobs from aborting an entire set of batch jobs (when in Batch Mode)

- The DVD compliant flag is now set for all CCE encodes except when Half-D1 is selected.

Updated version to v0.41

The following changes have been incorporated into v0.41:

- Corrected an bug that occurred when VTSs started off with SCRs other than zero that would have resulted 0003 and 0004 errors.

- Inserted code to remove the incorrectly coined "Layer Breaks" -- the annoying pause the happens when the original DVD would reach the point at which it switches layers. It, of course, isn't needed on a DVD-5.

- Added the "Average Bitrate" to the final summary line in each PREPARE and to the batch summary.

Updated version to v0.42

- Important update: Corrected a bug in which TFF/RFF flags could be set incorrectly on in some frames. This error would have resulted in a single frame "stutter" that might randomly occur throughout the video.

- Found and corrected a bug in which streams of pictures that were marked as interlaced in the original VTSs could be mistakenly marked as progressive upon REBUILD. A special thanks to Alex Z for helping identify these two bugs.

- Based upon some good advice from DDogg and others I have changes the default values for bias to 25 quality_prec to 16.

Updated version to v0.43

- Added a new feature under the "Mode" menu called "Half-D1 and Half Space for Extras". This feature, when checked, will resize any extras (that are outside the main [largest] VTS) to Half-D1 and will also half the allocated bitrate for them. The "recovered" space will be allocated to the main VTS. For movies with a lot of extras this can significantly improve the bitrate for the main movie. The new option will NOT work with ReJig for obvious reasons (ReJig is a transcoder and cannot resize the source).

- Corrected an error in which LPCM audio that was selected for removal was not properly flagged and removed.

- Change the default value for "Dynamically Assign Cell Bitrates" (DACB) to true (checked). Note that this only affects the default, if you have selected or deselected this option and it is stored in the INI file -- the value willstill be set to your last selection. Reasoning: DACB just does a better job of properly allocating space and bitrate. Also disabled this option when ReJig is selected (it has no affect in ReJig mode, bitrate is proportional anyway).

- (0.43a) Fixed the status printout of percentage of improvement associated with reduction.

- (0.43b) Fixed a goofy error in which sometimes the main movie would also be set to Half-D1

Updated version to v0.44

- Fixed the "Error '75'" that was associated with using mounted volumes

NOTE: Some browsers require a "refresh" or "Ctrl-R" to see the new version.

A NEWER VERSION IS AVAILABLE -- SEE THE FIRST POST OF THIS THREAD

[nwg]

Thanks.

I am rebuilding a project right now. Can I stop it and rebuild with 040, to take advantage of any stutter fixes? Or does the project need to be done again.


- Added a QuEnc Options item (why didn't I have one before?) and included two options "Trellis Quantization" and "KVCD Notch Matrix"

What are these?

Thanks.

[jdobbs]

Yes. If you have already done the first two phases, the rebuild is where the stutter fix is applied.

Trellis is some kinda' super-duper quantizer. It takes about twice as long but is reported to improve quality. I really can't tell you much more than that about either Trellis or the KVCD Notch Matrix. They are available on QuEnc and I was asked by one of the folks in the forum to add them. :)

[nwg]

Excellent.

Thanks for the reply.

I have already had impressive results with QuEnc and it is good to see the runtime error 5 bug finally sorted out.

Take a break, you have done a lot of work recently.

[digidragon]

Originally posted by jdobbs
NEW VERSION OF DVD-REBUILDER (v0.40)

Again a big thanks to the beta testers. They are, as a group, really keeping me on my toes.
Thanks to you for creating the software and being so quick to squash bugs. I've wanted to try re-encoding, rather than transcoding, but the normal way seemed to be a bit too complicated for me. Also, using your software, I get to keep the DVD intact without any reauthoring.

I've only used the software twice, so I'm hardly a beta tester, but apart from the motion blur problem (which you fixed in a test build) my first attempt at using rebuilder went smoothly and I ended up with a working DVD.

I'm upgrading my OS to XP tomorrow, so when it's up and running I'll try the DVD that had the motion blur problem using 0.4, and post back.

[wmansir]

Originally posted by jdobbs
Yes. If you have already done the first two phases, the rebuild is where the stutter fix is applied.

Trellis is some kinda' super-duper quantizer. It takes about twice as long but is reported to improve quality. I really can't tell you much more than that about either Trellis or the KVCD Notch Matrix. They are available on QuEnc and I was asked by one of the folks in the forum to add them. :)

Trellis coding is a nice and clever way of encoding a sequence of quantization indices (coming from a vector quantizer, in general). The division of scalar quantization intervals into four different subsets (Ungerboeck, and Marcellin & Fischer) and then the formation of a particular trellis provides some data compaction. The optimal path selected through the trellis ensures that encoding of the selected sequence of quantization indices will give the minimum distortion. Thus the scheme takes care, up to some extent, of both rate and distortion simultaneously.

PS. I have no idea what that means, i just copied it from goggle :)

I think I'll just use jdobb's description for the settings guide.

[jdobbs]

@wmansir

Man, I was impressed as hell until I saw the "google" comment.:D

[Morbo]

On the newest version of CCE,I converted my KILLBILL in about 3 hours on a 2.8 P4....

All I left off(In case anyone else has it) was the French audio and all went well....

Nice program,I could't find anything wrong with .38!!

Cheers!

[wmansir]

@ Jdobbs

I just added this to the Settings guide:Disable "interlaced"
Source for the selected VTS will be treated as progressive -- meaning "interlaced=true" will not be added to the ConvertToYUY() line, no Deinterlacing will be applied, Zig-Zag encoding will be used, and the progressive flag will be set. See this post (http://forum.doom9.org/showthread.php?s=&threadid=58800#post354127) for more information on ConvertToYUY(). Also, note that this is only an override for incorrectly flagged progressive material that would erroneously be treated as interlaced. Progressive material, that is flagged correctly, will be treated properly automatically.

I found that no deinterlacing is applied to the selected VTSs, so I added that information. I may take out the link on ConvertToYUY as this option encompasses more than just that setting now. Please let me know if anything is incorrect.

This also got me thinking, should Deinterlaced material not use "Interlaced=True", use Zig-Zag and/or be flagged progressive? I don't deinterlace myself, so I don't know.

[haggis663]

@jdobbs, thanks for an excellent program

My DVD player automatically stretches 4:3 aspect DVDs horizontally to fill my widescreen TV. I have used this DVD Rebuilder's 4:3 to 16:9 conversion and it works well, but chops of the bits that would have been black if it was letterboxed. It would be nice to have the option of pillerboxing (left + right) the image in an optional colour. Any chance of adding this to your list?

[sup191]

My first post!! :)

This really isn't a bug report, but more of a major THANK YOU to jdobbs! I just got around updating to .40 and the horizontal panning stutter of my Tombstone NTSC DVD backup is finally gone!! I was last using .32 and the stutter was very apparent. It used to stutter on my Xbox, Sony standalone, and WinDVD. Now it's flawless on all three! Thanks so much for an awesome tool! I don't have much, but I just donated to this excellent program. When the final 1.0 comes out, I'll gladly donate again. Now if I could only find another computer to use while mine is busy tackling my DVD collection... :D

[DDogg]

DVD-RB seems to be using enough CPU cycles that it is actually slowing down CCE 2.5 from 2.20 to about 1.50 real-time on my machine. It is easy to see, just use task manager to set DVD-RB's idle priority to low and watch the speed on CCE go up. Didn't Rb try to speak to this? This may be because dvd-rb is running at normal priority and cce runs at low priority on my machine?

[Sir Didymus]

Originally posted by wmansir
Trellis coding is a nice and clever way of encoding a sequence of quantization indices (coming from a vector quantizer, in general). The division of scalar quantization intervals into four different subsets (Ungerboeck, and Marcellin & Fischer) and then the formation of a particular trellis provides some data compaction. The optimal path selected through the trellis ensures that encoding of the selected sequence of quantization indices will give the minimum distortion. Thus the scheme takes care, up to some extent, of both rate and distortion simultaneously.

PS. I have no idea what that means, i just copied it from goggle :)

I think I'll just use jdobb's description for the settings guide.

My dear Moderator, you are really a crazy man !!!
Your joke have me in stitches...

People like you are fantastic...

[Fr4nz]

Originally posted by DDogg
DVD-RB seems to be using enough CPU cycles that it is actually slowing down CCE 2.5 from 2.20 to about 1.50 real-time on my machine. It is easy to see, just use task manager to set DVD-RB's idle priority to low and watch the speed on CCE go up. Didn't Rb try to speak to this? This may be because dvd-rb is running at normal priority and cce runs at low priority on my machine?

Here CCE 2.5 goes at 2.2x, so I don't think DVD-RB takes too many cycles.

[Xitrum]

The stutter at chapter changes is fixed in this version.

Thanks very much for your awesome software.:D :D :D

[jdobbs]

Originally posted by DDogg
DVD-RB seems to be using enough CPU cycles that it is actually slowing down CCE 2.5 from 2.20 to about 1.50 real-time on my machine. It is easy to see, just use task manager to set DVD-RB's idle priority to low and watch the speed on CCE go up. Didn't Rb try to speak to this? This may be because dvd-rb is running at normal priority and cce runs at low priority on my machine? That's probably it. I'll do some testing and see what happens. I've tried a couple of alternatives to the way I do it no, and have found the "sleeping" too long makes it difficult to cancel a job (you have to click on the button a couple/few times before it responds). I'll check it out some more.

[GizmoDerMokwai]

hi, for some reason (nice win32 features on a dvd) it wouldt be nice to have a option, to set custom output file size. maybe you think about hhat...

[jdobbs]

Originally posted by GizmoDerMokwai
hi, for some reason (nice win32 features on a dvd) it wouldt be nice to have a option, to set custom output file size. maybe you think about hhat... Are you saying you'd like to cut down on the output size in order to make room for the PC extras on the original disc? If so you can set the target output size in the INI file, it is described in the REBUILDER.TXT file. All you need to do is find the size in MB of the extras, convert it to 2K sectors, and then subtract that number from, and set, the target size.

[GizmoDerMokwai]

hehe, nice, sorry, i didn't know, that there was already a feature like that :D

[jdobbs]

Originally posted by GizmoDerMokwai
hehe, nice, sorry, i didn't know, that there was already a feature like that :D That's okay -- at some point I guess I'll have to sit down and write (gulp) a manual.

[Hirogen]

@Jdobbs

First of all thank you very much for a great program, have tried different versions and up to now I still have to run into my first bug, (I hope not ;) )

Have done LOTR 1 & 2 SEE, various Farscape episode discs, Babylon 5 episode discs and some other movies all without a glitch tonight I'm going to do Moby Dick with Patrick Stewart which is about 176 minutes, I wonder how it will come out but I have great confidence.

Using:
CCE SP 2.50
eclCCE 1.7b
DVD-RB Beta 40 for the first time
Avisynth 2.54
Running an Athlon XP 2400

As soon as my paycheck comes in I will make a donation.

Again, thank you very much, you are making my life easy.

[jdobbs]

Originally posted by Hirogen
@Jdobbs

First of all thank you very much for a great program, have tried different versions and up to now I still have to run into my first bug, (I hope not ;) )

Have done LOTR 1 & 2 SEE, various Farscape episode discs, Babylon 5 episode discs and some other movies all without a glitch tonight I'm going to do Moby Dick with Patrick Stewart which is about 176 minutes, I wonder how it will come out but I have great confidence.

Using:
CCE SP 2.50
eclCCE 1.7b
DVD-RB Beta 40 for the first time
Avisynth 2.54
Running an Athlon XP 2400

As soon as my paycheck comes in I will make a donation.

Again, thank you very much, you are making my life easy. That's great to hear. Good news is still news. I'm glad I created this comments and suggestions thread, the bug ones get pretty depressing.

[nwg]

I have been using DVD-RB since the first version. I have created DVD's from almost every version and they all worked fine (using DVD+RW's in case a DVD is faulty).

It does help that I have a very nice decent Sony Player NS900 that seems to play anything I put in it.

[jdobbs]

Cool.:cool:

[cdburrner]

hey jdobbs

using rb .40 and cce. it doesnt matter if i try using cce 2.50, 2.66, 2.67... i have eclcce installed for each version and i have the RB paths pointed at the eclcce.exe's. but when i go to encode RB tells me that "No CCE path has been set. Use menu Options/Setup to configure."

i've even tried setting RB's path to the original (non eclcce) CCE program exe and i get the same thing... i checked the rebuilder.ini file and the paths are entered correctly.

i have no idea what the hell is going on here! is anyone else having this problem?

[lab-one]

Just a thought, did you first run eclcce and point it to the ccexxx.exe?

[jdobbs]

Originally posted by cdburrner
hey jdobbs

using rb .40 and cce. it doesnt matter if i try using cce 2.50, 2.66, 2.67... i have eclcce installed for each version and i have the RB paths pointed at the eclcce.exe's. but when i go to encode RB tells me that "No CCE path has been set. Use menu Options/Setup to configure."

i've even tried setting RB's path to the original (non eclcce) CCE program exe and i get the same thing... i checked the rebuilder.ini file and the paths are entered correctly.

i have no idea what the hell is going on here! is anyone else having this problem? The only way you can get that message is if the CCEPath is set to an empty string. It is set to the path of the one you have selected in the CCEOptions menu. Each of the paths in the INI file are read at startup... are you sure they are correct? Have you modified them by hand? The settings should be for CCE250=, CCEBasic=, and CCENew= (all followed by a path). The INI file that is read will be in the same directory of the dvd-rb executable you are running.

[slafe]

Originally posted by cdburrner
hey jdobbs

using rb .40 and cce. it doesnt matter if i try using cce 2.50, 2.66, 2.67... i have eclcce installed for each version and i have the RB paths pointed at the eclcce.exe's. but when i go to encode RB tells me that "No CCE path has been set. Use menu Options/Setup to configure."

i've even tried setting RB's path to the original (non eclcce) CCE program exe and i get the same thing... i checked the rebuilder.ini file and the paths are entered correctly.

i have no idea what the hell is going on here! is anyone else having this problem?

check out :http://forum.doom9.org/showthread.php?s=&threadid=73051&perpage=20&pagenumber=30

slafe...

[shh]

I may be wrong but...

In CCE mode + some pieces resized to half-D1:
Shouldn't the half-D1 pieces also get half the bitrate?
As I see it in the .ecl, bitrates seem to be adjusted in the same ratio as full-D1 pieces.

Also, I think a ConvertToRGB-option would be good, because passing YUY2 to CCE makes the encodings bit too dark (there must be some colorspace-issure in conjuction with YUY2).

[jdobbs]

Originally posted by shh
I may be wrong but...

In CCE mode + some pieces resized to half-D1:
Shouldn't the half-D1 pieces also get half the bitrate?
As I see it in the .ecl, bitrates seem to be adjusted in the same ratio as full-D1 pieces.

Also, I think a ConvertToRGB-option would be good, because passing YUY2 to CCE makes the encodings bit too dark (there must be some colorspace-issure in conjuction with YUY2). I haven't done that yet (halving the bitrate).

[gunga]

Hello Jdobbs,

I still seem to have the same problem concerning the stutter/pause problem with the movie Haute Tension(NTSC-Interlaced-Top Field) on my JVC XV-N44 DVD player. I'm running DVDRB 0.40 with CCE 2.67.00.23/ECLCCE 1.7b, dynamically assign cell bitrate selected, and decomb selected. The stuttering/pause in the video/audio seems to happen at the same places with this version as it did with the previous version 0.39. Everytime the stutter/pause happens, (and as you mentioned before) the total bitrate for me goes above 10.08Mbps, which looks like the cause of my problem. Once it goes below 10.08 I have no stutter or pause in the movie. Why is the bitrate jumping so high, even with the DVD complianance flag enabled?

Thanks....

[Joergen]

Also, I think a ConvertToRGB-option would be good, because passing YUY2 to CCE makes the encodings bit too dark (there must be some colorspace-issure in conjuction with YUY2).

I didnt notice any difference in brightness when comparing the original VOB and the M2V generated by DVD-RB in RGB mode in two DVD2AVI windows, they seem exactly the same.

Of course if you open something in YUV mode in dvd2avi your video adapter overlay is used and it will look alot darker.

EDIT: I have now confirmed it. I took two screen grabs using DVD2AVI in RGB mode of the same frame in a backup I had made with RB34 and decrypting a vob from the original disc again, pasting both screenies in photoshop and using the eyedropper tool and ctrl-z (undo/redo) compare individual pixels of the two screenshots (when undo removes the newly pasted second shot and brings it back again over the first one) and the values are max 3 units off into either direction in all shades. And in the black areas and the black bars on the picture the values are exactly 16/16/16 on both shots. You cant do any better than that unless you dont encode at all :)

[Joergen]

Originally posted by gunga
the total bitrate for me goes above 10.08Mbps, which looks like the cause of my problem. Once it goes below 10.08 I have no stutter or pause in the movie.


What ever the reason for it spiking so high, a bitrate like that SHOULD stutter since the 1X reading speed of the disc simply cant bare it. Or perhaps some players can handle it without visible fault for a few millisenconds but its still a problem.

[shh]

Joergen> Of course if you open something in YUV mode in dvd2avi your video adapter overlay is used and it will look alot darker.

I didn't mean that.
This isn't visible on the PC, but you'll notice it at the TV.
The picture is about 4% darker if you pass YUY2 directly to the CCE (via avisynth & mpeg2dec).
You can also check that with photoshop's histogramm (or with a similar program).
I think this is a known problem, but I currently don't know, *why* this happens.

jdobbs> halfing the bitrate not yet done.

Ah! Ok.
I guess you're adding that soon. :)
Anyway... Many thanks for the great program! :)

EDIT:
Oh, Joergen, you just edited your post...
So forget the obove, you've checked that yourself already. :)

EDIT2:
> ...the values are max 3 units off into either direction in all shades.

Something goes wrong with the colorspace. This shouldn't happen.

> ... black areas and the black bars on the picture the values are exactly 16/16/16 on both shots. You cant do any better than that unless you dont encode at all

I'm sorry, but 16/16/16 doesn't tell anything.
CCE can also stretch the YUY2 colorspace to 0-255 and then cut off at 16/16/16.
But, as I said, I actually don't know what the CCE is really doing, so I also don't know how to fix it. But converting to RGB24 in avisynth does help.

[Xitrum]

Hi jdobbs
I was using DVD-RB v0.40 to do the "The Haunted Mansion" movie PAL region 4.
In the "Rebuild" phase, it got to 96.6% and then this error comes up:
"DVD Rebuilder experienced a buffer overflow. Error #0004. Process must abort" with an abort button. When you click on the abort button, DVD rebuider just shutdown.

Any ideas why this error occurs?

[Joergen]

Originally posted by shh

Something goes wrong with the colorspace. This shouldn't happen.

I'm sorry, but 16/16/16 doesn't tell anything.


I understand you are an expert with these but still:

With all lossy encoding there is a transormation of the image and in this case when pixels inherit values from their neighboring pixels the values naturally change a little. In flat areas of shades the values dont change at all (the pure black area). There are other areas in the shots like brown or gray where a single pixel doesnt shift the values at all or one unit in either of the R/G/B.

If you store a JPG with say 9 on photoshop scale you get similar shifting on a pixel level.

Comparing the histograms for R/G/B the Mean varies by -0.01 and the Std. Dev by +0.03 and doesnt affect the Median.

[DDogg]

shh, may I respectfully suggest you please bring your colorspace issues up in the avisynth developers forum and let that group verified your conclusions, specifically Sh0dan or Wilbert. I'm not particularly doubting what you say, but at the moment, it seems out of the scope of a beta dvd-rb dealing with many other issues. Or, if this is already done, could you provide specific links where one of them has verified this to save jdobbs/us the time of having to research it?

[shh]

> shh, may I respectfully suggest you please bring your colorspace issues [..]

Oh, I'm sorry. I did't want to tear this thread apart with other stuff.
I'm going to analyse some pictures (deeper, again) and go to the avisynth people.

[DDogg]

shh, thanks for reading my mind :) btw, I love your stuff and hope you never quit working on it.

[unplugged]

from DVD-RB changelog:
- The DVD compliant flag is now set for all CCE encodes except when Half-D1 is selected.
As I know that CCE flag doesn't do more than set arbitrarily max bitrate to 9800.
I remember to have red this by old Doom9 guide.
I ask confirm from someone!

[GizmoDerMokwai]

hiho, just wana to say, that i've tested dvdrb also for miniDVD! :D just blanked out most titles, removed some audiostreams, annd using cce2.5 i stored my 40min inline skting dvd on one 700mb cdr with good quality! :D

[Oldeman]

Is it possible to set the status of stripped (unchecked) audio streams to zero in the VTS_PGCITI/VTS_PGC(s) when rebuilding?

Love DVD-RB by the way...:)

[VILLA21]

Originally posted by unplugged
from DVD-RB changelog:
As I know that CCE flag doesn't do more than set arbitrarily max bitrate to 9800.
I remember to have red this by old Doom9 guide.
I ask confirm from someone!
Yeah, you are right. It"s better to leave this flag unchecked.

[philos31]

Originally posted by unplugged
from DVD-RB changelog:
As I know that CCE flag doesn't do more than set arbitrarily max bitrate to 9800.
I remember to have red this by old Doom9 guide.
I ask confirm from someone!

From the CCE Manual:

2.5.4 Creating stream for DVD
If DVD compliant is selected, encoder creates a stream as DVD
compliant.
This option involves the following matters.

Changing frame size
If the frame size of the original picture does
not conform to DVD standards, the frame size should be changed.
When the frame rate is 29.97 fps, the frame size is changed to 720 × 480, and when the frame rate is 25 fps, the frame size is changed to
720 × 576. Note that changing size means not enlarging the original
image but filling up a gap with mat.

Bitrate limitation
In the DVD standard, the maximum bitrate of
Video ES is limited to 9.8 Mbps. In the MPEG-2 VIDEO international
standard (ISO/IEC 13818-2), the size of an individual picture
is limited by the concept of “VBV (Video Buffering Verifier)”. In the
concept of VBV, a stream with a 9.8 Mbps bitrate can create GOP
whose size is to the extent of a maximum of 11 Mbps.
This perfectly conforms to the MPEG-2 VIDEO international standard
(ISO/IEC 13818-2), but whether it conforms to the 9.8 Mbps
restriction of DVD depends on interpretation of the standard. If
DVD compliant is selected, instantaneous bitrate in GOP units is
controlled to be a maximum of 9.8 Mbps.
During VBR operation, 9.8 Mbps is always written to the sequence
header in spite of the specified maximum bitrate.
9.8 Mbps is the maximum bitrate allowed under the DVD standard.
9.8 Mbps is used here because in the case of the VBV model in VBR,
bit allocation planning by the encoder becomes more flexible as the
maximum bitrate becomes higher, therefore higher image quality can
be achieved.

[jdobbs]

Originally posted by unplugged
from DVD-RB changelog:
As I know that CCE flag doesn't do more than set arbitrarily max bitrate to 9800.
I remember to have red this by old Doom9 guide.
I ask confirm from someone! That's a common misonception. But it does more than that. It also doesn't change the maximum bitrate at all. It only flags the maximum as 9800 in the header. The maximum stays what you set it. Here's a quote from the CCE manual and I've highlighted the important part:

"Bitrate limitation In the DVD standard, the maximum bitrate of Video ES is limited to 9.8 Mbps. In the MPEG-2 VIDEO international standard (ISO/IEC 13818-2), the size of an individual picture is limited by the concept of "(VBV (Video Buffering Verifier)". In the concept of VBV, a stream with a 9.8 Mbps bitrate can create GOP whose size is to the extent of a maximum of 11 Mbps.

This perfectly conforms to the MPEG-2 VIDEO international standard (ISO/IEC 13818-2), but whether it conforms to the 9.8 Mbps restriction of DVD depends on interpretation of the standard. If DVD compliant is selected, instantaneous bitrate in GOP units is controlled to be a maximum of 9.8 Mbps.

During VBR operation, 9.8 Mbps is always written to the sequence header in spite of the specified maximum bitrate. 9.8 Mbps is the maximum bitrate allowed under the DVD standard. 9.8 Mbps is used here because in the case of the VBV model in VBR, bit allocation planning by the encoder becomes more flexible as the maximum bitrate becomes higher, therefore higher image quality can be achieved."

[jdobbs]

Wow. I must be goofy. I just posted the same answer as the post just before me by philos31 ... did it really take me that long to compose it???

[DMagic1]

I've gotten this error both times I tried a moie with v.040 encoding on different segments.
http://bellsouthpwp.net/d/m/dmagic1/stuff/error.JPG

[quantum]

I just finished with a multi job batch. I see each job is summarized at the end. Good. I see the total time in minutes and high/low bitrate. But I don't see the average bitrate. An oversight? This is the more important number. High and low could be spikes.