Another quality comparison using SPIDERMAN

[UncasMS]

you're right, donpipo, the procoder2 results are darker

it is not procoder, who is to blame, though, but the missing avs option "convert to yuy2" which was unchecked in my comparison

with yuy2 on the picture is normal again in terms of brightness and contrast


and no, neither hc nor cce were used in opv mode as i'd rather consider this comparing apples to oranges

[donpipo]

it's not only darker, it's darker in dark areas and brighter in bright areas
i dont know if it is a colorspace matter
i just see details are effectively been trade

just have a look at the windows and curtains in the background (especially on the right) of this still for instance:
http://www.digital-digest.com/~uncasms/spiderman/32647_b_pro_jpg.html
compared to any of the others

[donpipo]

and, btw, thanks for the OPV answer.
i was thinking that would help making the recoding faster without hurting quality but unfortunately it's not

[UncasMS]

i havent noticed trade-ins in the sharpness but then on the other hand i only used opv with high bitrates

and yes, the procoder results were a colourspace matter

[techreactor]

i havent noticed trade-ins in the sharpness but then on the other hand i only used opv with high bitrates

UncasMS, can you share your experience on this. What threshold of Avg/High bitrates in your experience works good with OPV ???.

Also if you can remember the time savings using OPV.

[UncasMS]

since i prefer procoder for almost every job i hardly use cce or hc, which provide opv

when i use those encoders it is because the avg bitrate is rather high and procoder is not necessary

we may talk about 4500+ kbps


concerning encoding times -i have listed opv encoding times for every idct using cce + hc here:

http://forum.digital-digest.com/showthread.php?t=55813

the two pass times for this title were these:

cce 270 2 passes - 123 min
HC 016 - TR2 (profile "Best") - encoding - 213 min

Procoder 2 - 237 min

[michi90]

@ UncasMS

You wrote "shrink was used with max smoothness + aec".
Hhhhhh ?
But you have used deepanalysis + max smoothness (this is AEC) !
Correct ?

This is not fair !

You have to use shrink (always !) with maximum sharpness (this is AEC) + deepanalysis.

The names of the 4 different AEC settings are misleading.
Instead of 'maximum sharpness' I would prefer 'maximum precision'.
I've done many comparisons with the 4 different AEC settings.
Even with long extreme action DVDs (for example 'Gladiator')
'max sharp' produces the best quality.
Though it should be the other way round, 'max smoothness' produces extreme blocking artifacts ironically even in action sequences !

So, please use Shrink with 'maximum sharpness' and update your comparison !

[HAMP]

I'm new to this site. I am a CloneDVD2 user and I was looking at ProCoder2 and it shows a big difference.
I was wondering if you could add something to the final test, if you could show the original shots also.

[rendez2k]

Nice test! I notice you didn't test AutoQMatEnc, which the next version of DVD-RB is apparently going to natively support... would have been nice to see how it stacked up along side these.

[UncasMS]

thanx for your feedback and sugestion

i may run another test with another title using different shrink settings, autoqm,...

[UncasMS]

here is another comparison:

http://forum.digital-digest.com/showthread.php?p=388419#post388419

and max sharpness is NOT the way to go ^_~