TEST/POLL: All different DVDtoDVDR methods

[DereX888]

THIS IS ORIGINAL SOURCE:
SOURCE


Here are the samples of 3 BEST different DVD>DVDR methods:

SAMPLE1

SAMPLE2

SAMPLE3


PLEASE VOTE ON BEST LOOKING SAMPLE
To see differences better please save each bitmap on your computer and open them with your favorite image editor and zoom views (even Microsoft Paint - which is in every Windows since Win95 - can do this).



____________________________________________
How it was done:
The source was a TV capture MPEG-2 1-pass VBR at 9Mbps,
spanning equivalent of two DVD-R discs.
Virtual equivalent of 'dual layer' DVD was created from the source file,
then it was transcoded/reencoded using:
- Cinema Craft Encoder 2.50.0.1
- CloneDVD 2.0.5.3
- DVD2One 1.4.0
- DVD95Copy
- DVDFab 0.17.1
- DVDFab 1.9.9
- DVDShrink 3.1.7.6
- DVDXCopy Express 3.2.5
- InstantCopy 8.0.5
- QuEnc 0.51
All of them were set to achieve the best possible quality (i.e. deep analysis for DVDShrink, variable ratio for DVD2One, 6-pass for CCE etc etc), and all of them were set to fit their output on a single DVD-R (between 4460MB and 4475MB, however CCE output turned out to be slightly higher than the rest of them - 4481MB, and QuEnc made for me just about 4200MB file).

I tried to use ReJig as well, alone as an encoder, and within DVDReBuilder, but I just couldn't get it to work (probably I did something wrong since it was last one I tested, and its almost midnight here - so I gave up on it).

All sample pictures presented here are screencap bitmaps taken with PowerDVD, chosen from a high-motion scene, and then cropped (to fit on this board within its 50KB pic size limit, and to trick the board to accept bitmaps I changed their extensions to .JPG - but they are bitmaps still ).

[888888]

Yuck, what is that?

[HAMP]

@ 888888

I feel the same way, im not about to scan that. what the hell is it?

[vitualis]

You cannot tell the end video quality from one still frame...

Perhaps it would be useful to post short video clips...

Regards.

[vioxx]

April 1st is the best...

[FulciLives]

Maybe if you made the images even smaller we could tell even better. :P

I mean those pics aren't even VCD resolution!

How the hell can anyone tell the difference?

- John "FulciLives" Coleman

[GKar]

What! Are you all blind? Sample 4 is clearly the best. I use DVDShrink2CryptorOne 6.66 all the time and get the same outstanding results. Many thanks to the author Michael Moore.

[lordsmurf]

Is that a PENIS ROCKET ?

[GKar]

Is that a PENIS ROCKET ?

Looks like a mugshot of a Penisaurus from that X-Rated Buck Rodgers movie.

[FulciLives]

Is that a PENIS ROCKET ?

What the HELL is a PENIS ROCKET?

- John "FulciLives" Coleman

[DVDDave]

this dude wants us to vote on which penis we like.....very strange

[GKar]

Who cares, assmaster's girlfriend gives me a woody.

[DereX888]

LOL!
Yes it is PENIS ROCKET hehe
Its from "Flesh Gordon 2" movie I just capped on tv last week. Stupid but funny X-rated flick from early 80's or so, kind of "Porky's" in space.

Its true its hard to judge from a single frame, and much less from a cropped part of this frame, but:
This frame was chosen from a high motion scene as I explained first - bright stars flashing on the dark sky background, the rocket fly across them, and titles fade in and out in the same time.
If this scene (and frame) is not good for comparison purposes, then I dont know what else would be better.
Save the pics on your box and zoom them 6x to see greatr deal of differences in encoding, they are uncompressed BITMAPS - didnt I have said this already?

You know better test to post here - within 50kb picture size limit - then go ahead, do it. I'll be happy to check up on your job and make some stupid remarks too :P

Below is compressed JPEG of the original (source) BITMAP frame from which the cropped part is posted above. Same part of the frame was taken from each encoding/transcoding sample.




VOTE ON THE PENIS, DAMMIT :P

[GKar]

this dude wants us to vote on which penis we like.....very strange

I think we now know why they have a serious AIDS problem in Africa.

[DVDDave]

you hit the nail right on the head--not the penis head--the aids problem in Africa---why is this guy into posting his penis multiple times??? this should not be allowed

[richdvd]

@dereX888-It looks like your avatar has a "bulge" going on too

[DVDDave]

i noticed that bulge as well---do you think he likes penis's??? if so, he needs to keep that shit to himself

[stiltman]

i noticed that bulge as well---do you think he likes penis's??? if so, he needs to keep that shit to himself

Add that up with the fact that he's capping porno and it all become too clear

[GKar]

You should not let Stiltman II see this thread....

[adam]

In case this thread is actually serious... I see posts/polls like this every week and I always say the same thing. These types of comparisons literally mean nothing.

Compressed domain transcoders like DVD Shrink, DVD2One, etc... do not compress all frames equally, and more importantly, they leave some frames completely uncompressed entirely. You could process a source in both the absolute best transcoder in existence and the absolute worst transcoder in existence and analyze any given single frame and you'd get a different result almost every time. Out of all of those tests you performed, it is highly probable that at least one of your screengrabs is an exact copy of your source, because that particular transcoder did not apply any compression to it.

Really, the same idea goes for encoders as well. That frame you pulled from CCE. Is it an I frame, or a P, or a B? You may be comparing a P frame from your CCE encode against an I frame from your transcode. Once again, take the absolute best encoder and compare it to the absolute worst transcoder, and depending on what type of frame you get, you could get different results everytime.

In short, using screen grabs for this type of comparison is about as accurate as doing eenie meenie minee moe. It doesn't matter how demanding the given scene is, a single frame simply tells you nothing about the visual quality of the video or the quality of the encoder/transcoder.

You must use clips. There's no other way to make a comparison. Ideally you would take several clips from various points in the source because, once again, transcoders don't compress all frames. It is common to have an entire scene left untouched by a transcoder.

[GKar]

Hey Cyber Janitor, we were doing just fine here until you added an intelligent comment

[stiltman]

Hey Cyber Janitor, we were doing just fine here until you added an intelligent comment

Remember, he's trying to clean up......thus, back on topic..

Bad stiltman...bad!

[DereX888]

@dereX888-It looks like your avatar has a "bulge" going on too

it is NOT any "bulge"!!
Cant you *guys* (?) recognize a PENIS??

[Treebeard]

Cant you *guys* (?) recognize a PENIS??

I dont think its a question of can we, the question is why in the world would we want to stare at other ones until we recognize them.

Capping pron is the worst use of HD space I have ever heard of.

[gitreel]

Anyone that has a bear with a penis in his avatar is someone that is a little fruity. What is up with your penis fascination Derex?

[DVD_Ripper]

Can you please go back to your original movie and pick a better image to 'test' and present .... not that I'm homophobic but this forum is scaring me.

[gitreel]

Can you please go back to your original movie and pick a better image to 'test' and present .... not that I'm homophobic but this forum is scaring me.

I agree DVD_Ripper.

[DereX888]

If I'd chose naked chick frame I'd have to add stupid black dots to cover her nipples or pussy, but dick spaceship is OK - and yes, such sad facts are not funny either. Don't you think I'd rather post a cute chick instead? Thank to the idiotic american political correctness governing this board (and most of the internet) for that.
Anyway yet - it is not a real dick, so what are you bitchin about? Feeling too insecure about your own sexuality or something? Too puritan to find it funny that someone thought of creating 'spaceship' this shape?

[stiltman]

Thank to the idiotic american political correctness governing this board

Baldrick an idiotic American

[DereX888]

@adam:

Really, the same idea goes for encoders as well. That frame you pulled from CCE. Is it an I frame, or a P, or a B? You may be comparing a P frame from your CCE encode against an I frame from your transcode. Once again, take the absolute best encoder and compare it to the absolute worst transcoder, and depending on what type of frame you get, you could get different results everytime.

Thank you for pointing it out, but I thought of that before I posted it here, and all three examples are I-frame captures (however the source frame had to be a B-frame to keep the same frame - strange, but thats how it came out).

Compressed domain transcoders like DVD Shrink, DVD2One, etc... do not compress all frames equally, and more importantly, they leave some frames completely uncompressed entirely. [...]Out of all of those tests you performed, it is highly probable that at least one of your screengrabs is an exact copy of your source, because that particular transcoder did not apply any compression to it.

Yes, it is possible. However zooming each little piece of sample I have posted would show you that each piece is slightly different from each other, and from the source frame. It is feasible to assume they were all processed, right? Also - all of them are I-frames, while source is B-frame, this should prove best they were processed.

On the other hand you made me think of some dilemma. (I have reveal now that at least one of the 3 best samples is made by transcoder). How is it possible that the processed frame became an I-frame if the original frame was a B-frame? I thought transcoders just reduce the size of each frame, without changing GOP structure, since they don't do any 'real' encoding like encoders, you know what I mean?

You must use clips. There's no other way to make a comparison. Ideally you would take several clips from various points in the source because, once again, transcoders don't compress all frames. It is common to have an entire scene left untouched by a transcoder.

Unfortunately I have to agree. Clips, from few various scenes (dark/bright scene changes, high motion and slow motion) would be the best comparison. Probably there is not much 'value' in this test, but to its defense I must add that I have viewed with few friends each final video for couple minutes before we selected three best ones (and actually it came down to 2 best ones from which we couldnt decide which one of them looks better, the third one added here was merely a 'distant second' compared to both of them).