DVD-RB and AutoQMatEnc

@ windtrade

Hi

If you read what I said you would of noticed that 5 posts down from that quote I explained that I did not look at the frames properly. I should of waited before posting for which I appologise.

I do these tests in my spare time which I dont have alot of.

While it would be great showing moving clips I just do not have the time to do it, by showing a still image members will get a rough idea on how well the transcoders and encoders work at high compression.

If you could point me in the right direction of a program that I could use to make short video clips then I would like to know as I could add this feature in future compression tests.

 

Very few threads, or forums have actual comparisons of the programs, so this is a real plus. pazzini you have done terrific job in doing this, I know how pressed for time people are and how long it takes to do something like this. Since I do not have ProCoder it is good to see the comparisons.

 

Great job Pazzini, keep up the good work. After doing my comparisons and looking at the ones that you have done, I am liking Procoder more and more, I'll still use CCE when I need it done a little quicker, but Procoder seems to give better results on a more regular basis. Thanks again.

 

@all,

I do appreciate the efforts here and my post was not intended as any sort of complaint. Conducting these test is time consuming, well recognized, and clearly demonstrates the superiority of reencoding rather than transcoding on highly compressed source.

My intent is to refine and add more sophistication to this type of comparision testing. For example, one can see the difference between CCE and Procoder in the still images; Procoder tends toward softing while CCE tends toward sharpening. The question is what is the perceptual qualitative difference between these two when viewing the video. Which version tends to reproduce the source more accurately? Maybe certain scenes benefit from the softening effect while others benefit from the sharpening.

Developing a few standard test scenes we all use for testing just seems logical if we want to maximize the ability to compare results from various tests from various testers. A variety of source material could be evaluated more consistently. For example, high motion scenes vs low motion, high contrast vs low contrast, the obvious flame scenes, poor quality vs high quality sources, etc.

I would be willing to develop a short test DVD with chapters having a short clip of the various test scenario. I'd like a few suggestions for specific source secenes we would agree represent what is typically encountered on mainstream commercial dvd releases.

My preference would be to post the video on usenet. Can most of you download from there? I could post it on my public website but depending on the size and demand this may become a problem.

Meanwhile I will get a quick one put together to see if this makes any sense.

fwiw - I'm not a newbie at this and have used nearly all the transcoders and encoders at one time or another as well as various process automaters such as DVD2SVCD and DVD rebuilder.

 

pazzini,

What quality setting are you using with procoder? I just got my vs set up and was just wondering.

 

I always use mastering quality

 

Thank you. Also how many passes with CCE?

 

Hi mort

I use 3 passes with CCE.

If you are doing some tests for yourself any chance of posting your results in here?

Also if you ever conduct tests with the different quality settings in procoder2 I would be very interested in your findings, as this is something I have yet to do.

 

I encoded the same folder (7.82 gb, reduction level 52.6%) with RB/CCE SP with 3 passes and with RB/Procoder2 set at highest quality (not mastering quality). CCE took 166 minutes and Procoder2 took 193 minutes (that encludes prepare, encode, and rebuild times). I might run procoder2 at mastering quality later. I might post some screen shots later too.

 

Be interested in those procoder screen shots, do not have this one so am curious as to how it does in the different settings

 

I for one would really like to see how well procoder2 on highest quality setting comes out compared to cce 3 passes and to the mastering quality setting in procoder2.

 

Procoder2 highest quality 193 minutes.



CCE 3 passes 166 minutes.



Sorry if the frames aren't exactly the same. There was a descrepancy between frames done with cce and procoder for some reason. I might run this through procoder2 at mastering quality and post it later. It's been one of THOSE days.

 

The procoder one looks very good, I am envious

 

Here's some better pics.

Procoder2 mastering quality. 189 minutes



Procoder2 highest quality. 193 minutes.



CCE 3 passes. 166 minutes.



The reason it took longer for procoder2 highest quality than it did for mastering quality is because I was multi-tasking while running highest quality.

 

They all look pretty damn good for having a reduction level of 52.6%.

 

Holy S*%t that mastering quality is fantastic, now I am really green

 

Noobs making comparisons...............

 

Thankyou mort,

Procoder2 with mastering quality is the winner for me then procoder2 with highest quality setting while comes cce last.

1st Procoder2 MQ
2nd Procoder2 HQ
3rd cce

All three pictures look excellent.

Thanks for doing the different quality setting in Procoder2 much appreciated

 

Well done ddp. Contadictory statement. A noob calling us noobs? Besides I was asked by more than one member to post those screenshots for comparison purposes.