One-pass VBR vs. two passes w/CCE?

[Bret34]

I very recently discovered DVD Rebuilder and in total love with it! Having owned or tried a number of DVD compression s/w, I wasn't expecting the discernible improvement the DVD-RB/CCE combo provided.

That said, my question for the extremely savvy and experienced users on the forum is that, in the interest of saving encoding time (average 57 minutes vs. 2 hours on 7-8 gig discs), have been setting CCE SP ver. 2.67.00.23 in DVD-RB for "one-pass VBR w/analysis " instead of the conventional two-pass method. I've tried it both ways, but on my Panny 42" Plasma haven't seen any qualitative difference I can really identify. I've read most of the FAQs and guides (as well as posts here) on CCE and DVDR I can find, but don't recall having seen much regarding this subject. Would appreciate some input or being pointed to a link or two that might address this.

Thanks to all for helping me to video nirvana! The only problem is that I'll have to re-rip my previously ripped DVDs '-)

[drob]

After two year of using multipass vbr, I myself started using one pass vbr, to my best observation on a 34" tv via m not able to tell the deference.
There is allot of info on the board regarding this issue, it took me awhile to get to it thaw, run a search using the key "opv" that’s what most members call it.

[leglover]

Originally posted by Bret34
I very recently discovered DVD Rebuilder and in total love with it! Having owned or tried a number of DVD compression s/w, I wasn't expecting the discernible improvement the DVD-RB/CCE combo provided.

That said, my question for the extremely savvy and experienced users on the forum is that, in the interest of saving encoding time (average 57 minutes vs. 2 hours on 7-8 gig discs), have been setting CCE SP ver. 2.67.00.23 in DVD-RB for "one-pass VBR w/analysis " instead of the conventional two-pass method. I've tried it both ways, but on my Panny 42" Plasma haven't seen any qualitative difference I can really identify. I've read most of the FAQs and guides (as well as posts here) on CCE and DVDR I can find, but don't recall having seen much regarding this subject. Would appreciate some input or being pointed to a link or two that might address this.

Thanks to all for helping me to video nirvana! The only problem is that I'll have to re-rip my previously ripped DVDs '-)



I have experimented with up to 4 passes and the one pass VBR method using both 2.67.00.23/27 and can't honestly can't see any difference on heavily compressed (185 minutes) DVDs. My reference is a 50-inch Pioneer PDP. I think the reason you haven't seen more than one response to your query is because the so-called "experts" here don't have a clue, preferring to discuss arcane tweaks that have no bearing on reality. Just my 2-sou worth.

[RB]

Each method has its merits. Multipass VBR gives exact size and allows you to tweak quality of high action vs. static scenes by modifying VBR Bias. OPV is real fast and gives excellent results too and there are several tools that can automate the tedious task of finding a Q factor that results in a specific target size. Just use what you like best
Originally posted by leglover
I think the reason you haven't seen more than one response to your query is because the so-called "experts" here don't have a clue, preferring to discuss arcane tweaks that have no bearing on reality.Welcome to the forums. Leglover, this is not a good start. What do you think you gain by offending the forum members in your first post? Please don't wrongfully accuse people here, I can't remember a single thread that discussed "arcane tweaks". Please stay reasonable, OK?

[Bret34]

Originally posted by RB
Each method has its merits. Multipass VBR gives exact size and allows you to tweak quality of high action vs. static scenes by modifying VBR Bias. OPV is real fast and gives excellent results too and there are several tools that can automate the tedious task of finding a Q factor that results in a specific target size. Just use what you like best.

Thanks for the response, RB. When you say that OPV has several tools that automate the task of finding the Q factor, can I assume you're referring to DVD Rebuilder's One-Pass w/analysis option?

Also, when using Multipass VBR, I follow the CCE SP & Rebuilder guides, using the generally recommended settings, going with 3 passes. I know there are tools for finding the Q factor for this process, but haven't had the time for mastering the (for me) difficult learning curve along with the trial & error of tweaking high action and static scenes, I'm not that sophisticated. It sounds really really tedious.

Having used both methods with 6 movies, I've yet to see any real difference between either. I just want the "comfort" of knowing the optimal way, especially with favorite films that I want to match the quality of the original as closely as possible.

Best,
Bret

[tylo]

When you say that OPV has several tools that automate the task of finding the Q factor, can I assume you're referring to DVD Rebuilder's One-Pass w/analysis option?If you can live with movie-only backups, you can try DVD2DVD (DVD2SVCD) + the D2SRoBa Plugin that I made. Personally, I prefer movie-only for multiple reasons:
1. The movie starts immediately by just loading the disc (hate menus).
2. The movie will get higher quality because it fills the full disc.
3. I never take the time to look at the extras anyway

In addition, with DVD2DVD, you have more options, and it's a tad faster because it only does the movie.

If you insist on making full DVD backups, DVD-Rebuilder is second-to-none. Still, DVD Shrink is much faster, and comes really close in quality if the movie is easy compressible, or when it doesn't need to compress much. Many choices...

Cheers.

[Bret34]

[QUOTE]Originally posted by tylo
If you can live with movie-only backups, you can try DVD2DVD (DVD2SVCD) + the D2SRoBa Plugin that I made. Personally, I prefer movie-only for multiple reasons:
1. The movie starts immediately by just loading the disc (hate menus).
2. The movie will get higher quality because it fills the full disc.
3. I never take the time to look at the extras anyway

In addition, with DVD2DVD, you have more options, and it's a tad faster because it only does the movie.

If you insist on making full DVD backups, DVD-Rebuilder is second-to-none. Still, DVD Shrink is much faster, and comes really close in quality if the movie is easy compressible, or when it doesn't need to compress much. Many choices...[/QUOTE}

I appreciate your input, but of the many, many "one-click" solutions I've used (Shrink looking the worst to me), Pinnacle's IC7 or 8 come the closest in quality to what I've gotten from the DVD-RB/CCE combo -- whether it's feature-only or an entire disc. My rule-of-thumb is to use CCE for heavy compression of full backups and IC8 for down to 70%. As a matter of fact, there aren't any noticeable differences...to these eyes at least on my Panasonic 42" plasma...between 70+% with either software.

That said, I'd still love to get some clarification from RB's response to my most recent query here while I give your program a try -- if its learning curve is manageable for my feeble mind
:D.

Many thanks!

[RB]

Originally posted by Bret34
That said, I'd still love to get some clarification from RB's response to my most recent query here while I give your program a try[/B]
Not sure what you are after :). As I said, it's mostly a matter of personal choice and what method fits best. I for instance use both VBR and OPV. OPV with Q around 50 for DVD extras. I then reauthor only the extra parts of the DVD in Scenarist, compile and use a bitrate calculator to calculate the bitrate for the main movie from the space left. Main movie then gets encoded with VBR mostly 1 pass (VAF + 1 VBR pass that is), rarely 2 VBR passes.

[r6d2]

Originally posted by Bret34
I've tried it both ways, but on my Panny 42" Plasma haven't seen any qualitative difference I can really identify.This is the whole point. Trust your eyes.
Would appreciate some input or being pointed to a link or two that might address this.If you want to get a glance understanding why you see no difference at all, I humbly invite you to read The Idiot's Guide on my sig.

[Bret34]

Originally posted by r6d2
This is the whole point. Trust your eyes.
If you want to get a glance understanding why you see no difference at all, I humbly invite you to read The Idiot's Guide on my sig.

Thanks so much for your input and the guide to which you directed me. I found it interesting, informative and agree with your overall point about trusting one's eyes. But not having the time to do extensive burning and visual experiments with various combinations of settings, I thought asking my original question would evoke more specific information and tips regarding the more historically successful (notice I didn't use the verboten word "optimal" :sly: ) settings in the CCE/DVD Rebuilder -- thus taking advantage of the vast storehouse of knowledge and experience in the group. I suppose I'll have to keep searching and reading to come up with a satisfactory group of answers.

Cheers!

[leglover]

Originally posted by Bret34
Thanks so much for your input and the guide to which you directed me. I found it interesting, informative and agree with your overall point about trusting one's eyes. But not having the time to do extensive burning and visual experiments with various combinations of settings, I thought asking my original question would evoke more specific information and tips regarding the more historically successful (notice I didn't use the verboten word "optimal" :sly: ) settings in the CCE/DVD Rebuilder -- thus taking advantage of the vast storehouse of knowledge and experience in the group. I suppose I'll have to keep searching and reading to come up with a satisfactory group of answers.

Cheers!

I think the guy's right and the only thing you can rely on are your eyes, and from what you say, your hi-def top-of-the-line video. I've read as much as I can about optimal, correct, preferred, yatta, yatta , yatta settings. And unless you have the inclinations of a hobby math geek who likes to diddle with the software IN SPITE of the movie, use the default settings or those found in the DVD-R and CCE FAQs. Sit back and enjoy. Toy around with the tweaks, if you like, while I'm enjoying my collection of over 1500 DVD "backups" while others seek the ideal "Q" factor (which doesn't make a darn bit of dif).

I'll probably get my hand slapped...again...but that's the most "optimal" advice you'll get on this forum if you're a movie lover. Trust me.

Best,
LL

[kainis]

As everybody I want to know and use the best setting the DVD-ReBuilder (CCE) and reach the closest output to an original movie with the best quality of a picture and non-noticeable errors of a picture.
I use DVD-ReBuilder v0.56 + CCE SP (v2.50).
I used CCE Customized Settings before with these values.
VBR_Bias: 25
Quality_Prec: 16
VBR_Passes: 5 (it took me on PC about 8.5 hours)
I thought that the more VBR_Passes I would use the better (although some voices say that more than 3 passes or even 2 are worthless) and 5 VBR_Passes seemed to me as maximum what I could manage overnight.
But now I have done the very long film Spartakus (3:09) with 5 languages – so that it was necessary to compress it up to 45 % and overall bitrate was 1950 Kb/s.
When I play so encoded movie on a PC people have „ghosts/aura“ round themselves, everything is as though vibrating. But on a normal TV (60 cm) I have not noticed any problems.
After that I tried to set up the ReBuilder with One Pass VBR (w/analysis) /OPV/ and saw that encoding took only 2.2 hours (the same as with Shrink 3.2 AEC-Sharp).
Furthermore I compared quality of a picture on a PC and it was better than with 5 passes! In compared scenes even higher bitrate was reached. It was by 10 – 40 % higher that with 5 passes and a size of files fitted to DVD-5. These scenes were more colored, lighter, more action but on the contrary in darker and less colored scenes the bitrate of One Pass VBR was cca 5-10% less than with 5 passes.

Because I am a little confused by all these facts I would like to discuss some my questions with somebody who has more knowledge and experience in that and can advise what to choose. My questions:
1) Is any explanation, that One Pass VBR (w/analysis) reach better quality than the 5-pass mode? It is generally valid that the more iterations are used the better distribution of bitrate is performed. But these iterations result in effect that bitrate in all scenes is getting closer to a mean/average value of bitrate, which is very low for very long movies and complicated (a lot of colors and actions) scenes. It causes that with many passes (e.g. 5) spot values of bitrate are close to mean and in more complicated scenes are not sufficient and in easy scenes are maybe uselessly high. Whereas using the One Pass VBR there is no mysterious averaging and bitrate is therefore better distributed and picture has better quality. Is it a correct opinion?
2) Concerning my example of Spartacus where the overall bitrate was 1950 Kb/s, I reached in more colored scenes with One Pass cca 4 Mb/s but with the 5-pass method 2.5 Mb/s however in less colored and darker scenes with 5-pass method 1.8 Mb/s but with One Pass only 1.6 Mb/s. What is better to prefer, brings better quality of picture, causes that it will be closer to the original movie and you will not notice any problems in a picture? To have higher bitrate in more difficult scenes and less bitrate in easier scenes (OPV method) or to have less bitrate in more difficult scenes and higher bitrate in easier scenes (Multipass method, 5 passes)?
3) Which setting do you recommend? One Pass VBR (w/analysis) /OPV/ or Multipass (5 passes)?
4) Is it really true that more passes bring you better quality of a picture? For example 5 passes (generally agreed limit of quality, after that increases of quality are too small) are better that 2 passes (implicit setting)?
5) Are any differences between settings for differently long movies (e.g. 1-hour movie, 2-hour movie, 3-hour movie)?
6) Under what compress not to go with CCE (ReBuilder)? Or under what minimum overall bitrate (I know that it depends on quality of an original but some referential values maybe exist)?

For visual comparison I can send you these variants: Original, Shrink, 1pass and 5 passes.

Thank you very much for your comments.
Kainis

[Bret34]

[QUOTE]Originally posted by kainis
As everybody I want to know and use the best setting the DVD-ReBuilder (CCE) and reach the closest output to an original movie with the best quality of a picture and non-noticeable errors of a picture.
I use DVD-ReBuilder v0.56 + CCE SP (v2.50).
I used CCE Customized Settings before with these values.
VBR_Bias: 25
Quality_Prec: 16
VBR_Passes: 5 (it took me on PC about 8.5 hours)
I thought that the more VBR_Passes I would use the better (although some voices say that more than 3 passes or even 2 are worthless) and 5 VBR_Passes seemed to me as maximum what I could manage overnight.
But now I have done the very long film Spartakus (3:09) with 5 languages – so that it was necessary to compress it up to 45 % and overall bitrate was 1950 Kb/s.
When I play so encoded movie on a PC people have „ghosts/aura“ round themselves, everything is as though vibrating. But on a normal TV (60 cm) I have not noticed any problems.
After that I tried to set up the ReBuilder with One Pass VBR (w/analysis) /OPV/ and saw that encoding took only 2.2 hours (the same as with Shrink 3.2 AEC-Sharp).
Furthermore I compared quality of a picture on a PC and it was better than with 5 passes! In compared scenes even higher bitrate was reached. It was by 10 – 40 % higher that with 5 passes and a size of files fitted to DVD-5. These scenes were more colored, lighter, more action but on the contrary in darker and less colored scenes the bitrate of One Pass VBR was cca 5-10% less than with 5 passes.

Because I am a little confused by all these facts I would like to discuss some my questions with somebody who has more knowledge and experience in that and can advise what to choose. My questions:
1) Is any explanation, that One Pass VBR (w/analysis) reach better quality than the 5-pass mode? It is generally valid that the more iterations are used the better distribution of bitrate is performed. But these iterations result in effect that bitrate in all scenes is getting closer to a mean/average value of bitrate, which is very low for very long movies and complicated (a lot of colors and actions) scenes. It causes that with many passes (e.g. 5) spot values of bitrate are close to mean and in more complicated scenes are not sufficient and in easy scenes are maybe uselessly high. Whereas using the One Pass VBR there is no mysterious averaging and bitrate is therefore better distributed and picture has better quality. Is it a correct opinion?
2) Concerning my example of Spartacus where the overall bitrate was 1950 Kb/s, I reached in more colored scenes with One Pass cca 4 Mb/s but with the 5-pass method 2.5 Mb/s however in less colored and darker scenes with 5-pass method 1.8 Mb/s but with One Pass only 1.6 Mb/s. What is better to prefer, brings better quality of picture, causes that it will be closer to the original movie and you will not notice any problems in a picture? To have higher bitrate in more difficult scenes and less bitrate in easier scenes (OPV method) or to have less bitrate in more difficult scenes and higher bitrate in easier scenes (Multipass method, 5 passes)?
3) Which setting do you recommend? One Pass VBR (w/analysis) /OPV/ or Multipass (5 passes)?
4) Is it really true that more passes bring you better quality of a picture? For example 5 passes (generally agreed limit of quality, after that increases of quality are too small) are better that 2 passes (implicit setting)?
5) Are any differences between settings for differently long movies (e.g. 1-hour movie, 2-hour movie, 3-hour movie)?
6) Under what compress not to go with CCE (ReBuilder)? Or under what minimum overall bitrate (I know that it depends on quality of an original but some referential values maybe exist)?

For visual comparison I can send you these variants: Original, Shrink, 1pass and 5 passes.

Thank you very much for your comments.
Kainis [/QUOTE}

Kainis, you have superbly articulated my own dilemma (as originally posted) with OPV vs. the multi-pass methods. Try as I may, using CCE 2.67.00.23 with DVD-REbuilder plus the AVS scripts posted on this Forum, I have yet to see any differences -- either with large compression movies (50% or less) on my PC or Panasonic plasma unit when burned to DVD. This might interest you, as it seems to have stirred up something of a thread when recently posted, but when using Pinnacle's InstantCopy on the same movies ("Captain and Commander", "Kill Bill" 1 & 2 -- all with menus and extras intact), there were no visable differences using any of the methods or alternate software.

I am beginning to think this whole controversy about "best" settings and the "transcoders vs. encoders" issue is more a sport rather than anything useful in helping film buffs enjoy better quality copys.

Just my 2-sou worth :p

PS: I'd like very much to see your comparison variants. Can you upload them to a link we can all see? Thanks.

[kainis]

PS: I'd like very much to see your comparison variants. Can you upload them to a link we can all see? Thanks. [/B][/QUOTE]
---
As I am not skilful in forums I prefer to send the pictures to you and you can place it on this page. Pls send me your address where I can deliver these pictures and you can see the differences and place them here.

I discussed this problems at some forums before and decided to use only OPV method because it gives me the same or even better results than Multipass method and it is 4-times quicker (than 5-pass)! I just want to have time for seeing movies and not for testing different passes still more and more when results from OPV (not the same as 2-pass of Multipass method!) are preferable for me.

Here are some questions with answers I sent to different forums or people which I think can help you decide which method to use:

Mine:
Overall bitrate of the OPV method was absolutely the same, i.e. 1950 kbs.
Maybe I could describe as all my troubles arose.
I made first encoding by the 5-pass method but when I saw the first scenes on my PC, I was shocked how it is vibrating, people have fuzzy outlines, etc. But it was on the PC with only 15” monitor and when I played it on my TV with 60 cm diagonal I was not able to notice anything.
So I tried OPV (One Pass) method and was surprised that these first scenes had better pictures and better bitrate and started thinking how it is possible that so quick and generally mentioned that OPV method gives less quality than Multipass can give me more quality of picture and it is closer to the original film!
As the overall bitrate of both method is absolutely the same, only amount of data of files are higher in OPV, but still possible to burn, I also tried a bitrate in darker scenes and found what I had mentioned: OPV method - in more difficult scenes higher bitrate than Multipass and in easier scenes less bitrate than Multipass; Multipass method, 5 passes in more difficult scenes lower bitrate than OPV and in easier scenes higher bitrate than OPV.
- My opinion is that to give more bitrate in more difficult scenes and lower in easier scenes, which OPV gives, is better than contraposition, which Multipass gives. (I hope that this is generally valid thing and not only for my case, but I am not 100% sure). Do you agree or not?
- So can we generally agreed which method brings us better quality: OPV or Multipass?
- Does it depend on a length of film which method to choose and if Multipass is better, is the setting for different lengths of films the same (e.g. VBR_Bias: 25, Quality_Prec: 16, VBR_Passes: 4)?
Kainis
---
Mine:
In spite of the fact that I have read a lot of forums about the issue Rebuilder/CCE Setting One Pass VBR (OPV) versus Multipass VBR I have not learnt any clear answer.
As I mentioned in the forum http://club.cdfreaks.com/showthread.php?p=667817#post667817 my experience, but only with one movie where I noticed it and analyse it, is that One Pass VBR gave me better quality (better outlines around people, not so pixelled, finer picture) than Multipass VBR, 5 passes.
However I do not know if it is only for this specific case or if it is generally valid.
Does anybody know if using OPV where Q-factor is determinated better than Multipass where you should change all parameters according to the particular quality of film, CCE version, compression level, etc when you want the best possible quality?
Does OPV only hardly determinate size of files or also reduce quality of picture?
And basically if OPV gave better quality than Multipass, why Multipass was used in CCE, when it takes much longer? It would not give a sense.
So advantage of OPV could be that it should give you better quality (duration time is not considered) than Multipass but you risk that it will not fit to DVD-r and you will have to compress it again or it is not true?
Can anybody help and explain it?
---
Answer:
Firstly, the OPV method of encoding is "relatively" new. 2 guys over at the Doom9 forum came up with the process at the same time and so was always refered to as the RoBa method. Others then have applied this concept to various programs, including RB. The best known one is the D2SRoBa pluggin for DVD2SVCD. For some background to all this, check out DVD2SVCD Advanced forum at doom9 http://forum.doom9.org/showthread.php?s=&threadid=73243

Traditionally the input to CCE was VBR encoded taking the raw data in and with more and more passes it improved the VBR allocation. With 4-5 passes, it approachs the original VBR patern of the original DVD and achieving the "best" quality. Final file size can be very acurately predicted here so it was used widely. The RoBa method, or OPV method, simply shrinks and maintains the existing VBR pattern of the original to fit to final required size. This cannot be simply calculated (thats why it had not been used in the past), but can be achieved by doing small (1% of the the number of frames) test encodes, adjusting the Q value up or down until file size is achieved. This found Q value is then applied to the whole movie. Generally its pretty good in finding the desired file size, but with some DVD's this is not. Sometimes under (Lowest size I've seen is 4.19gb) and sometimes over (Largest size I've made is 4.38gb) but generally around the 4.28-4.32 gb size (well for me that is).
---
Mine:
Thank you for your answer, but the basic and main questions still remain:
Does the OPV method really have the same quality as Multipass and does the OPV not have any disadvantages in processing of video or picture quality? I mean e.g. blocking somewhere in a movie, pixelisation, bad colours, unfocused view, etc?

I also find at this addresses http://www.mpegit.net/dvdr_opv.php and http://www.videohelp.com/forum/viewtopic.php?t=175302 the opinion that OPV gives you the same quality as Multipass 4-5 passes.
I understand that it is not easy question because conditions are various and compared files from both method can have different sizes, but you agree to this opinion, you use the method OPV for all the movies (also for short ones and for very long ones), you have not had any problem with OPV quality of movies when you played them till now?
If your answer is positive, I will be happy (at last) and will do all the movies with this method in the future.
Thanks again and sorry that I am so persistent.
Kainis
---
Answer1:
In my opinion, yes the OPV 1 pass RoBa method will/does achieve the quality output of multi VBR passes with no disadvantages except final size prediction. Sorry, I dont have any pics to compare as I did all my testing some time back and that was with the RoBa pluggin with DVD2SVCD. But results will vary from DVD to DVD, take the 2x DVD9 version of the classic Ben-Hur and encode down to a DVDR. Even the original DVD is far from perfect but the OPV method still gives pixelisation free output except for some very minor content in the extreme action scenes.

All I can suggest is to listen to everyone elses opinions and their findings and then do your own testing. We all have different playback equipment, we all have different expectations and we all are backing up very different DVD's. Only YOU can answer your own question.

9 out of 10 people are using Transcoder (Shrink/CloneDVD/etc) to backup their DVD's and only 1 out of 10 use Encoders (like CCE, TMPGEnc, etc), so does that suggest Shrink etc are better than CCE for example? At least 3 people have expressed an opinion saying OPV generally give equal results to 4-5 VBR CCE encodes, 10,000,000 and more have not, does that make OPV better than VBR multi pass ?

If your looking for some finite, absolute, chiselled in stone, technical reason why one way is better than the other, then I am unable to help you, nor do I think anyone else including Robshot and Bach who came up with the idea in the first place.
---
Answer2:
Main idea:

MultiPass VBR:
size prediction over quality

One Pass VBR:
quality over size prediction

First there were no programs like D2SRoBa or DVD ReBuilder that really helps for
finding the right Q-value. That is why everyone liked the multipass method.
Nowadays the D2SRoBa and the DVD ReBuilder are so simple to use that OPV method
is coming more and more popular. People that are saying "OPV can't match the
quality of multipass" are just ignorant. They are mixing two absolutely
different things. "One Pass VBR" is not the same thing in the CCE SP as "1-Pass
Multipass VBR" is. Those are not the same thing! That is the main reason why
most of the "sceptics" are wrong :) They just don't realize the difference
between those two CCE SP settings.
---
Rds
Kainis

[Bret34]

Thanks for such an exstensive and informative response, Kainis. I've come to believe, through reading and experimentation, that there is no "ultimate" answer to my original question. And furthermore, having backed up more than two dozen movies since I first posted it, I've used both recommended methods for CCE and DVD-R along with IC8 and have found that, for the vast majority of the flicks, IC8 is superior in the end result. It's probably mostly subjective, but whether previewing the results on PC or standalone DVD, Instant Copy is the winner -- and much much easier to work with. So I'll leave it to the dedicated hobbyists to tinker with CCE settings. As one poster put it, my first love is the movie itself.

[ducati9x]

hey fellas,

after succumbing to the temptation to go for an all-out slugfest on test comparisons doing the same choice movies (terminator 2, matrix revolutions, shawshank redemption, and casablanca) i have come to the following conclusions:

for the newer released movies that have been mostly shot on HD or digital - the encoding of the actual 8.5gig dvd is well enough to let shrink handle the job with extremely little conversion loss. I mean that the movie still looks as amazing as the original even down to 78% compression.

for earlier movies that have been re-encoded and re-grained, etc... one pass vbr works wonders using cce & rebuild. My new WS star wars trilogy is indistinguishable from the originals using one pass vbr.

for super long movies that would normally require to be compressed to 70% or less... one pass vbr still does a perfect job. (ala shawshank)

for other movies they are just nearly impossible to backup - like the entire AKIRA KUROSAWA collection from criterion. it doesnt matter which program i use... nothing comes out good at 50-65% compression.

I should note however that i usually rip the entire movie (uncompressed) using dvdshrink. I never needed menus or extras anyways - or multi-language subtitles - or multilanguage audio - or DTS audio... 5.1 channel surround is always sufficient. then encoding with cce is done.

My viewing pleasure is accomodated by a samsung 42" HD projector using a Sharper image progressive scan dvd player. All my conclusions are objectively related to my visual perception on my big ass television.

Much love to those looking at this screen
:)