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

[Joergen]

Originally posted by jdobbs
We're getting way too complicated. Make a program that you need a PhD to understand, and even PhDs will avoid it.

Kinda like a modern car, even the engineers who design them dont know how to fix them.

edit: jdobbs, write CTRL-R since I think it's different than Refresh in IE (and 99.9% use IE).

And HOT D*MN another super feature!

[jdobbs]

NEW VERSION (v0.43) UPLOADED

The newest version of DVD Rebuilder is now available through the first post of this thread. The changes are noted below:

- 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).

NOTE: Some browsers require a "refresh" to see the new version.

[onesoul]

Originally posted by jdobbs
We're getting way too complicated. Make a program that you need a PhD to understand, and even PhDs will avoid it. That was only an example, could be just 0.5 and 0.75.
Do you know what would be complicated? You trying to explain me how DVD-RB works and thus explaining the dvd structure.. :p ;)

[onesoul]

Don't you think 0.75 of allocated bitrate for extras at full frame is acceptable? (considerating progressive frames)

[DDogg]

- Added a new feature under the "Mode" menu called "Half-D1 and Half Space for Extras". :D I'll just bet you know I'm a very happy guy. Thanks for taking the time to add this. Hopefully it will not come back to bite at ya. Will test tonight.

[DDogg]

onesoul, with respect, give it a rest :) This is supposed to be a quick down and dirty, possibly temporary, solution until he gets more time to think it out, down the road. Let's just be grateful for this very nice surprise.

[jdobbs]

Originally posted by onesoul
Don't you think 0.75 of allocated bitrate for extras at full frame is acceptable? (considerating progressive frames) I understand. This latest version supports the Half-D1/Half Space feature. The reason I like this so well is because you can save a lot of space, but keep the Q relatively high on the extras (people with a standard non-HDTV television may not even notice). I'm looking at options for full D1 extra reductions as well -- we'll see how that goes.

[jdobbs]

Originally posted by DDogg
onesoul, with respect, give it a rest :) This is supposed to be a quick down and dirty, possibly temporary, solution until he gets more time to think it out, down the road. Let's just be grateful for this very nice surprise. Exactly. I still like the idea of sampling to find the minimum acceptable Q on the main movie and stealing from extras to meet it. That takes time though.

[DDogg]

I wonder if the first dvd I chose was a problem one. Italian job.

-----------------
Phase I, PREPARATION started at 21:30:26
- Note: "Half-D1/Half Space for Extras" mode is enabled.
- VTS_02: 189,319 sectors.
-- Scanning and writing .D2V file
-- Processed 12,140 frames.
-- Building .AVS and .ECL files
- VTS_03: 371,554 sectors.
-- Scanning and writing .D2V file
-- Processed 31,248 frames.
-- Building .AVS and .ECL files
- VTS_04: 2,630,468 sectors.
-- Scanning and writing .D2V file
-- Processed 159,007 frames.
-- Building .AVS and .ECL files
- VTS_05: 691,869 sectors.
-- Scanning and writing .D2V file
-- Processed 67,644 frames.
-- Building .AVS and .ECL files
- VTS_06: 66,502 sectors.
-- Scanning and writing .D2V file
-- Processed 4,150 frames.
-- Building .AVS and .ECL files
- Reduction Level for DVD-5: 53.6%
- Overall Bitrate : 2,626Kbs
HALF-D1 Extras increased Main-Movie bitrate by .0%
- HIGH/LOW/AVERAGE Cell Bitrates: 7,071/300/2,626 Kbs
Phase I, PREPARATION completed in 5 minutes.

[jdobbs]

FYI. For those of you who get it I'm going to try very hard to, in my next posted version, fix the errors #0003 and #0004.

[jdobbs]

Originally posted by DDogg
I wonder if the first dvd I chose was a problem one. Italian job.

-----------------
Phase I, PREPARATION started at 21:30:26
- Note: "Half-D1/Half Space for Extras" mode is enabled.
- VTS_02: 189,319 sectors.
-- Scanning and writing .D2V file
-- Processed 12,140 frames.
-- Building .AVS and .ECL files
- VTS_03: 371,554 sectors.
-- Scanning and writing .D2V file
-- Processed 31,248 frames.
-- Building .AVS and .ECL files
- VTS_04: 2,630,468 sectors.
-- Scanning and writing .D2V file
-- Processed 159,007 frames.
-- Building .AVS and .ECL files
- VTS_05: 691,869 sectors.
-- Scanning and writing .D2V file
-- Processed 67,644 frames.
-- Building .AVS and .ECL files
- VTS_06: 66,502 sectors.
-- Scanning and writing .D2V file
-- Processed 4,150 frames.
-- Building .AVS and .ECL files
- Reduction Level for DVD-5: 53.6%
- Overall Bitrate : 2,626Kbs
HALF-D1 Extras increased Main-Movie bitrate by .0%
- HIGH/LOW/AVERAGE Cell Bitrates: 7,071/300/2,626 Kbs
Phase I, PREPARATION completed in 5 minutes.
No. I stuck the percentage line in about 10 minutes before posting without testing it. I'd say it works but the printout is wrong. Let me look at it and I'll post an "a" version.

[jdobbs]

Originally posted by jdobbs
No. I stuck the percentage line in about 10 minutes before posting without testing it. I'd say it works but the printout is wrong. Let me look at it and I'll post an "a" version. Yep. I divided the decimal value by 100 to get the percentage... duh... I was supposed to multiply by 100. Do you think that might do it? I'll post in about 5 minutes.

[Joergen]

Originally posted by jdobbs
FYI. For those of you who get it I'm going to try very hard to, in my next posted version, fix the errors #0003 and #0004.

It will be most appreciated (though I've only gotten it once and managed around it). Some people though seem to get it from not setting up avisynth and mpeg2dec :rolleyes:

Could you make a sanity check for video decoding. Perhaps a short clip that is encoded and if the output is not within a range then say theres a problem.

[jdobbs]

@DDogg,

If you have a few minutes, could you try that one again with the new version and post the percentag you get?

Thx.

[gopalkk]

Thanks for planning to fix #0003 and #0004.

Menu compression is needed bad as my sources have 15% space in it.

It would be useful if you allow eclPasses=1 in the GUI
although it works when i set it in rebuilder.ini which
saves 50% time on my 2.4 HT PC.

Another suggestion is LPCM -> DD2.0 transcoding which also
can save LOTS of space but may not apply to 90% of disks.

Rebuilder seems to take more CPU time (30%)

[gopalkk]

FYI,
i set TargetSectors=2386000
and got 4.32GB output from 7GB source.

[Joergen]

That CPU percentage problem is clearly a P4 HT problem. On my XP2100+ and AMD64 3200+ CCE gets 96-98% of the CPU constantly.

[jdobbs]

Originally posted by gopalkk
Thanks for planning to fix #0003 and #0004.

Menu compression is needed bad as my sources have 15% space in it.

It would be useful if you allow eclPasses=1 in the GUI
although it works when i set it in rebuilder.ini which
saves 50% time on my 2.4 HT PC.

Another suggestion is LPCM -> DD2.0 transcoding which also
can save LOTS of space but may not apply to 90% of disks.

Rebuilder seems to take more CPU time (30%) I guess I can increase the sleep time between checks... but when you do that the "Abort" button sometimes doesn't respond.

[DDogg]

Here is two, the first is Good Fences, some gain. But, wow, look at the second which is the one I posted above! :D (if it is right)

-----------------
Phase I, PREPARATION started at 22:08:14
- Note: "Half-D1/Half Space for Extras" mode is enabled.
- VTS_01: 3,244,391 sectors.
-- Scanning and writing .D2V file
-- Processed 213,468 frames.
-- Building .AVS and .ECL files
- VTS_02: 237,077 sectors.
-- Scanning and writing .D2V file
-- Processed 16,747 frames.
-- Building .AVS and .ECL files
- VTS_03: 199,497 sectors.
-- Scanning and writing .D2V file
-- Processed 15,186 frames.
-- Building .AVS and .ECL files
- VTS_04: 239,946 sectors.
-- Scanning and writing .D2V file
-- Processed 18,271 frames.
-- Building .AVS and .ECL files
- VTS_05: 38,852 sectors.
-- Scanning and writing .D2V file
-- Processed 2,980 frames.
-- Building .AVS and .ECL files
- VTS_08: 43,985 sectors.
-- Scanning and writing .D2V file
-- Processed 3,372 frames.
-- Building .AVS and .ECL files
- Reduction Level for DVD-5: 50.0%
- Overall Bitrate : 2,365Kbs
HALF-D1 Extras increased Main-Movie bitrate by 9.4%
- HIGH/LOW/AVERAGE Cell Bitrates: 2,660/300/2,365 Kbs
Phase I, PREPARATION completed in 5 minutes.
-----------------
Phase I, PREPARATION started at 22:15:50
- Note: "Half-D1/Half Space for Extras" mode is enabled.
- VTS_02: 189,319 sectors.
-- Scanning and writing .D2V file
-- Processed 12,140 frames.
-- Building .AVS and .ECL files
- VTS_03: 371,554 sectors.
-- Scanning and writing .D2V file
-- Processed 31,248 frames.
-- Building .AVS and .ECL files
- VTS_04: 2,630,468 sectors.
-- Scanning and writing .D2V file
-- Processed 159,007 frames.
-- Building .AVS and .ECL files
- VTS_05: 691,869 sectors.
-- Scanning and writing .D2V file
-- Processed 67,644 frames.
-- Building .AVS and .ECL files
- VTS_06: 66,502 sectors.
-- Scanning and writing .D2V file
-- Processed 4,150 frames.
-- Building .AVS and .ECL files
- Reduction Level for DVD-5: 53.6%
- Overall Bitrate : 2,626Kbs
HALF-D1 Extras increased Main-Movie bitrate by 47.3%
- HIGH/LOW/AVERAGE Cell Bitrates: 3,561/300/2,626 Kbs
Phase I, PREPARATION completed in 5 minutes.


This can't be right. I'll try to check some more.

[jdobbs]

All -- Don't use 0.43a... I just found a bug. If your main vts is anything but the first one it sets it to Half-D1. I'll post a fix (0.43b) as soon as I test it.

[jdobbs]

Alright, 0.43b is posted...

[godhead]

Good thing I was reading the forum, was just about to kick off an encode with .43a :D

Thanks for the quick catch!

[DDogg]

Italian Job .43b
-----------------
Phase I, PREPARATION started at 22:41:09
- Note: "Half-D1/Half Space for Extras" mode is enabled.
- VTS_02: 189,319 sectors.
-- Scanning and writing .D2V file
-- Processed 12,140 frames.
-- Building .AVS and .ECL files
- VTS_03: 371,554 sectors.
-- Scanning and writing .D2V file
-- Processed 31,248 frames.
-- Building .AVS and .ECL files
- VTS_04: 2,630,468 sectors.
-- Scanning and writing .D2V file
-- Processed 159,007 frames.
-- Building .AVS and .ECL files
- VTS_05: 691,869 sectors.
-- Scanning and writing .D2V file
-- Processed 67,644 frames.
-- Building .AVS and .ECL files
- VTS_06: 66,502 sectors.
-- Scanning and writing .D2V file
-- Processed 4,150 frames.
-- Building .AVS and .ECL files
- Reduction Level for DVD-5: 53.6%
- Overall Bitrate : 2,626Kbs
HALF-D1 Extras increased Main-Movie bitrate by 21.8%
- HIGH/LOW/AVERAGE Cell Bitrates: 4,624/300/2,626 Kbs
Phase I, PREPARATION completed in 4 minutes.

[Joergen]

10-30% is significant. Either means I dont have to strip some piece of extras I dont want or all the better quality for the movie.

Good job jdobbs!

[philos31]

Originally posted by Joergen
That CPU percentage problem is clearly a P4 HT problem. On my XP2100+ and AMD64 3200+ CCE gets 96-98% of the CPU constantly.

I can confirm that.

On my AMD 2600+ DVDRB uses 1-2% proc. Time.
On my P4 3Ghz HT it uses around 10-15%

[RB]

Originally posted by jdobbs
@RB,

From looking at the D2V it appears there are a lot of places that were interpreted as stills because an I-FRAME is followed immediately by another I-FRAME (the lines with a single item on it). I guess the question is why the AVS files don't reflect that...
OK, thanks for looking into this. But actually, these I-Frames are not stills, this is just a trailer on this DVD which plays sequencially and seamlessly. It just has a lot of quick scene changes at the beginning (you can really see this), so that's probably why the original encoder inserted a lot of I-Frames.

[RB]

Originally posted by jdobbs
FYI. For those of you who get it I'm going to try very hard to, in my next posted version, fix the errors #0003 and #0004.
Great! :) I have a project waiting for rebuild here that will be a very good exercise (currently fails with #0004). VTS_01 is 612 MB but video is just 16 still frames, the rest is AC3 and LPCM audio. A typical slide show with music tracks.

[jdobbs]

Originally posted by RB
OK, thanks for looking into this. But actually, these I-Frames are not stills, this is just a trailer on this DVD which plays sequencially and seamlessly. It just has a lot of quick scene changes at the beginning (you can really see this), so that's probably why the original encoder inserted a lot of I-Frames. I guess I may need to look for a different method of determining stills. I'm going to leave it for now until I see how often this might occur.

[rui]

A question:

Unless i am mistaken (it’s quite possible :p ), i believe that someone on the forum said that one could use a transcoder without loosing much quality, if a DVD backup wouldn't go down below 75% of the original space.

Assuming this is correct, and since DVD-RB now informs, after the PREPARE fase, the average bitrate, what would be the value that would determine if one could use Rejig without a too great loss, or should use CCE/QuEnc?

There are situations where one doesn’t mind to loose a bit of quality for the sake of speed, if the loss would remain at acceptable values.

I believe that the latest versions of DVD-RB do the PREPARE fase in such a way, that the ENCODE fase could de done using either 3 methods, right?

Sorry for my bad english

[onesoul]

@DDogg, jdobbs

I'm zipped :)

[jdobbs]

Originally posted by rui
I believe that the latest versions of DVD-RB do the PREPARE fase in such a way, that the ENCODE fase could de done using either 3 methods, right? Definitely not. You have to do the prepare for the method you will be using in Phase II and III... For example, if you choose ReJig the PREPARE takes considerably longer because all the original cells must be demuxed.

[rui]

Sorry. My mistake :o

[RB]

Originally posted by jdobbs
I guess I may need to look for a different method of determining stills. I'm going to leave it for now until I see how often this might occur.
OK. AFAIK you already told me that a still frame could be identified by looking at some timecode that exceeds the normal duration of a single frame in case it's a still.

[Djuby]

Originally posted by gopalkk
FYI,
i set TargetSectors=2386000
and got 4.32GB output from 7GB source.

I did some tests to find the MAX TargetSectors. You can use this one:
CCETargetSectors=2260007

I made two backups already and both of them came out 4.37 GB. Just a reminder, the final size is very close to DVD-R limit, so consider lowering the target sectors just a little bit. I personally prefer to be on the edge :-)

Djuby

[nwg]

I have just done Roger Waters 'In The Flesh' R1 with Rejig. I can confirm that the LPCM is now removed with 0.43b. :)

I also really like that extras at Half D1 option.

[timekills]

I know the post after this one won't make much sense after I edit this...heh.

Anyhow, in response to below - I was pretty sure that was the problem too, but no dice. I know the link is right though, since I can hit it from my other PCs, just not that one.

I should remember the adage: Try everything you can, and then try one more thing before you post.

Well. It should be an old adage.


Anyway, it motivated me to give jdobbs some $$$, so I guess it wasn't an entirely wasted post.

[lab-one]

Try refreshing your browser, hit ctrl+r, clear your browsers cache (temporary internet files) or use Internet Explorer.

[DVD Maniac]

The Half D-1 mode for extras looks really promising and I am going to give a try tonight. One problem I can see here is when you are encoding an episode based source with multiple Titles, how will Rebuilder be able to differentiate between the Titles which are the episodes (which we want to maximise the bitrate) as opposed to the extras which might be of similar size? I realise we might be trying to run before we walk here but could a feature be added for us to identify those titles which we want a Half D-1 allocation?

[nwg]

There is also the option to set which VTS files are to Half-D1 manually in the AVS options.

As long as each VTS for the episodes is known, the extras VTS can be turned into Half D1.

[DVD Maniac]

Ah yes - I missed that - cool.:cool:

Now if only those Authoring companies would have the decency to put all the extras in separate Title-Sets. Unfortunately they often don't and have that annoying habit of stashing them in the main movie title-set. That's where the one-clickers ability to specify compression levels by title comes in really handy.

Currently I am doing a two stage process for these types of structures. Using Shrink to reduce the size of the extras that I want to keep + shrinking down menus a tad first. Then feed the result to Rebuilder - works fine, I have only had one result which Rebuilder did not like.

What's the likelihood of Rebuilder eventually having the same Title specific reduction functionality? Not a major issue - just a polite suggestion for later enhancement:p

[nwg]

Currently I am doing a two stage process for these types of structures. Using Shrink to reduce the size of the extras that I want to keep + shrinking down menus a tad first. Then feed the result to Rebuilder - works fine, I have only had one result which Rebuilder did not like.

I am doing something similar. I am doing Kill Bill and reduced the extras and menus down to 60% in Shrink (with deep analysis). Then doing the disc with CCE in DVD Rebuilder and Half D1 on the extras.

I have also gained an extra 10% in disc space.

[jdobbs]

Originally posted by nwg
There is also the option to set which VTS files are to Half-D1 manually in the AVS options.

As long as each VTS for the episodes is known, the extras VTS can be turned into Half D1. Hmmm... actually I hadn't thought of that -- I don't cut the bitrate on the "normal" Half-D1. Maybe I should.

[smlong426]

Hmm.. I have my target size as TargetSectors=2270000 .. Using the Half D1 and Half Space option, my DVD size came out to only 4.03 gigs. This doesn't seem right ..

[jdobbs]

Originally posted by smlong426
Hmm.. I have my target size as TargetSectors=2270000 .. Using the Half D1 and Half Space option, my DVD size came out to only 4.03 gigs. This doesn't seem right .. Don't know how... every sector that gets removed from extra allocation gets added to the movie... maybe I'm miscalculating somewhere.

[smlong426]

I'll just up the size and see what happens ..

[plethman]

i like the new feature in .43b where "extras" are half size
another nice feature might be a VTS min size variable ... in which if the VTS size is smaller then set variable ... the VTS won't be processed. re-encoding a 100mb file .. or even smaller doesn't do much for saving space.

just my 2 cents ... keep up the good work .. i love this proggie

[nwg]

jdobbs wrote,
Hmmm... actually I hadn't thought of that -- I don't cut the bitrate on the "normal" Half-D1. Maybe I should.

I didn't realise that. I think it might be better to have the option of full bitrate or reduce it.

I did a whole DVD at Half-D1 (homemade DVD of a TV series - eight 1/2 hour episodes). I did it at Half-D1 so, I didn't get pixelation like I did with DVD Shrink. I wouldn't be able to do it with reduced bitrate at the same quality.

[Joergen]

Originally posted by plethman
i like the new feature in .43b where "extras" are half size
another nice feature might be a VTS min size variable ... in which if the VTS size is smaller then set variable ... the VTS won't be processed. re-encoding a 100mb file .. or even smaller doesn't do much for saving space.

just my 2 cents ... keep up the good work .. i love this proggie

That's already at 50MB :D

But it might be nice to have a custom variable through the options :cool:

[jdobbs]

Originally posted by Joergen
That's already at 50MB :D

But it might be nice to have a custom variable through the options :cool: The only reason I don't make it settable yet is that it would likely cause a lot more of the #0003 and #0004 errors. Most of the occurances of the source of those errors happens in those small VTSs.

[Joergen]

Originally posted by jdobbs
The only reason I don't make it settable yet is that it would likely cause a lot more of the #0003 and #0004 errors. Most of the occurances of the source of those errors happens in those small VTSs.

Yes I meant you could set it higher. :) Ie 50, 100, 150 to avoid multiangle extras or other problems.