latest optimal script and DVD2SVCD CCE multipass?
 

Anyone try this?

I'm testing it out right now. I just set DVD2SVCD up to make a 2 hour movie on 2CDs the same as I normally would before I discovered KVCD, except I am using Kwag's latest script.

I'll post results when I'm finished or if anyone else has tried this please post.

I'll tell you one thing it is slowwwww. I have a 2.8ghz P4 with tons of RAM and 2 7200RPM 8MB cache drives and my normal CCE speed is 2.4, but with the optimal script it went down to 1.03!

Oh well we'll see the results in about 6 hours.

show more details please audi2honda,like:
matrix,resolution..everything.

see this post:

http://www.kvcd.net/forum/viewtopic.php?t=4424
:wink:

Hi Jorel,

I read that thread and the wierd thing is I just cut and pasted the same script I am using for KVCD in ToK into DVD2SVCD. I didn't need to change the format of the script, variable, comments, or anything. Wonder why? Only thing I removed is the source project which DVD2SVCD puts in for you.

About other parameters. I simply used the standard CCE matrix and SVCD resolutions 480x480. I used the same crop, border, and overscan sizes as I did with KVCD.

My goal with this test is to compare my 704_480 optimal script ToK KVCD output with the same movie encoded with CCE multipass and the optimal script.

Reason for this is that up until a few days ago when I found this site I exclusively used DVD2SVCD. I want to see what effect the optimal script has on this process, because I think CCE multipass is better then TMGEnc.

Well see how it goes. At the end I'm going to compare the same movie encoded both ways and make a decision on which I want to use moving forward.

I suspect the KVCD will look better because of the higher resolution, but the CCE will produce less artifcats and mosquito noise.

Hi audi2honda,

Why didn't you encode to 704x480 with CCE in 2-pass :?:
That way you could compare 1 to 1 resolution quality, and then see the difference :cool:

-kwag

do you think MPEG2 SVCD can handle that resolution at the same bit rate and not look like poop?

That's exactly what I think, that MPEG-2 at low bitrates even with CCE will look like caca at 704x480, even at 100-passes :mrgreen:
But then that's the real way to compare it to TMPEG's MPEG-1 at low bitrates :cool:

-kwag

Ok well it's running. It's pretty slow, but I'll have something by tomorrow. I changed it to 704_480 so that the only difference between the two movies are one is Mpeg2 SVCD and the other Mpeg1 kvcd. i used the optimal script on both.

Hopefully your right and it doesn't look any better, because it takes way to long to encode. :) stay tuned

yes,yes, :wink: :wink:
i'm with you 2 ,great observations in the last 2 posts.
Kwag is right,you is right.

waiting... :D

Ok I'm done and did a couple of encodes because I have two machines running.

Well here are my observations. Note I didn't have time to watch on the TV yet only the monitor.

DVD2SVCD using the optimal script and encoding as SVCD at 740_480 using the standard CCE matrix
Clearly less banding in low detail areas. CCE seems to break up the banded areas. Not neccessarly putting more detail, but the waves of banding are much less evident. Encoding time very long even on 2 pass

DVD2SVCD using the optimal script and encoding as VCD at 740_480 using the KVCD matrix.
Also better at reducing banding then ToK. I don't really see much difference from the SVCD encode

ToK using the optimal script and encoding as VCD at 740_480
ToK shows more banding then the others, but the colors especially skin tones look warmer and more natural. The image is also smoother and this is a good thing. When I say smoother I mean I notice slightly less jagged edges and noise on high contrast transitions compard to the CCE DVD2SVCD method. Thumbs up to ToK for this.

Overall my conclusion is that CCE did somehow do a better job at breaking up the annoying banding you get in vary dark areas. Allmost like the blockbuster plugin but better. However I think ToK gave a more natural and smooth film like picture. Not to mention ToK was about 2.5 to 3 times faster to encode because there was no multi-pass.

I guess I'll live with the banding for now and am definetly going to stick with ToK. Aside from the subtle differences in low detail solid areas ToK produced a better image and took way less time to encode.

Note that I am probably very patricular about quality so I'm pretty critical of the images. I have screen shots of all 3 to compare, but don't have a host for them. If someone cares to post them you can see for yourself.

Hope this insight is helpful to someone :)

great results and observations audi2honda
:wink:

"Note that I am probably very patricular about quality so I'm pretty critical of the images."

after worked with televisions,videok7,cameras,for 30 years,
i can speak that my eyes can see little differences.
i'm really "bored" with audio and video quality! :roll:

i hate "more or less" results. :evil:

i post in the forum and in pms that with ToK,
i got better results with kvcds and the Kwag's script
then using cce in dvd2svcd.

dvd2svcd is easy to use for who know the prog but,
if follow the guides(great guides) of Red-M,"friendalunit"(where are you?),
black prince,GFR,muaddib,hedix...and tons more guides and helps from very wise friends here with the Kwag's script,
the result is fantastic. :wink: :wink: :wink:

well,this is just my opinion and i stay in front of the pc testing
everything tons of times all day long....every day!!!

then my friend audi2honda
your results are really important for us all.

thank you very much.

:D

Hi audi2honda,

Can you convert the screenshots to .PNG and E-Mail them to me :?:
I'll post them here.

-kwag

Sure, but I can't find your email. What is it?

I PM'ed you.

-kwag

Here they are:

TOK 704_480.png
http://www.digitalfaq.com/archives/error.gif


D2S VCD with KVCD matrix cce 2 pass.png
http://www.digitalfaq.com/archives/i.../2003/06/5.png


D2S_SVCD_704_480_cce_matrix_2_pass.png
http://www.digitalfaq.com/archives/i.../2003/06/6.png

They look so dark. That is actually a persons shoulder there where the arrow is. I can't see any differences in the detail here. Oh well I guess converting them as PNY lost the color or something.

The details are there if you blow them up in Windows Paint. But the difference is very small.

-kwag

Ok I see you are right. Subtle differences yes, but the good thing is ToK KVCD took way way less time to encode. :D

yes,
and we don't need to see this "gay" again.... :roll:

ps:
read it faster,i will edit it.
:lol:

LOLROTF :mrgreen:

:lol:

now i can't edit cos you quote this post..... :lol:

but really, using ToK is better...no doubts :!:
i will send this thread to everybody that ask me
about dvd2svcd with cce and ToK with best script!

...
then all can see this "g**" face to compare
and maybe the "boy" got a boyfriend.
:lol: :lol: