Thank you very much, Vhelp :!:
|
Hello Vhelp,
When I use your script I get following Message by TMPGEnc. "Convolution3D supports YUY2 color format only" Can you help with that :?: |
Bchteam,
Are you using the avisynth 2.5 version of Convolution3D? Available at http://hellninjacommando.com/con3d/ |
It's fixed.
I had to download the YV12 Version of Conv3D, instead of the regular one. |
hi Bchteam, ..TGIF 8)
The Script Challenge (SC)... I'm glad you got it all sorted out. Unfortunately, you live in German (assumed, cause you've ben assigned MOD under German forum) so, we may miss each other, in time to respond quickly enough :( Case you were wondering... I chose Convolsion3D, simply because no-one was using it at the time, and I wanted to script it w/ some feel for "orginality". I worked the script to a point that satisfied me enough for this SC, and in 'nough time too. There's so much work going on here that things can get pushed asside. I myself have pushed a few things aside, these past few weeks. It just depends on what's going on at the time, and sometimes, it can be soo excited that.. Well, anyways.. I hope that you are playing around w/ the snip portion I've used in above SC. It's for all to do as they see fit. I'm no script guru, but I'm getting a place in their somewhere's (someday) Take it easy all, and have a great weekend :) -vhelp |
My family poll results:
MrTibs_attempt2 - 3 votes MrTibs_kwag - 2 votes Comments: I was quite surprised how close each of the encodes actually looked on the TVs. On the computer screen there seems to be quite a difference but on TV, choosing became difficult. Personally, I think I prefer Vhelp's submission. |
@ MrTibs..
Quote:
When we view these VCD's, we're really suppose to view them at 352x240. And, as such, no-one would be able to tell the differrence. I did that with all 4 clips. I couldn't tell the difference :!: In my honest opinion, I think the TV's are set close to 640x480, but I could be wrong. Lots of debate on this issue, and then some, over at vcdhelp :!: IMO, I thought Kwag's and your's were the prefereable subs. I'm stopped playing around w/ 352x240 stuff long ago.. But, in any case, it was fun playing around some :D 8) Have a great day all (weekend) -vhelp |
Quote:
|
Personally, I think the cleanest and most stable of all was the last clip submitted by MrTibs :roll:
And that was the whole idea of the test. To remove the noise from the clip, and maintain best quality without loosing too much details :!: But I'd rather see some minimal noise than a blurred picture :) -kwag |
@ Wilbert..
Sorry 'bout that. I missed one. I did see your's, and yes, it was sharper too!! The difference was based on "strutney" (spelling) testing. Under normal cercumstances, we all just watch the movies we encode (I hope) ... That's 5 clips, hehe.. I was more interested in the "noise filtering" though, cause I thought that that is what MrTibs' post was all about. But, you are right :!: ..sharpness needs to be maintain, and not overlooked. Thank you for reminding me of this important fact for VCD resolution (352x240) its important to maintain as much sharpness (detail) as possible. :) and yes, after looking at your sample again :) ..IS a bit sharper :!: Noise, is in the :roll: of the beholder. Have a great day. -vhelp |
After viewing all the clips- Wilberts is the only one with the correct aspect ratio 8O The others look vertically stretched when compared to the original test.m2v file. Also it's definitely sharper. It's a bit too blocky though (IMO). As far as smoothness though I liked MrTibs_attempt2 and vhelps.
|
I agree that Wilbert's clip was the sharpest but it did have quite a bit of blockiness.
There were a few reasons for the contest. One the main reasons was to provide a uniform starting point so that everyone could easily compare script results. The "goal" I guess was to show the various ways people are cleaning video. I didn't want to start with a DVD source because the amazing work that KWAG and others have done have made it nearly perfect (IMHO) already. What I have taken from the contest are ideas on making the "perfect" script. I also have a hardened benchmark that I can compare future scripts with. Now filters can be written and used to further impove the results. I guess my dream would be to have the "perfect" filter: A motion adaptive temporal filter with varriable motion smoothing. Idealy this filter would do all the things that Kwag's (and my) script does but faster. For now, however, I will be re-writing my script to include some of the features in each of the submissions to improve the results. (Like Wilbert's sharpness and Kwag's motion smoothing). Great work to everyone who submitted! |
Hey guys i just need a little help how come i cant download these have they been taken off or something by this i mean are they no longer able to download cause i would like to have a look at them so i cant experiment myself thank u.
|
@smarty,
Please look at the dates on the threads before posting. This thread is a year old :!: -kwag |
oh lol didnt look thx m8
|
Site design, images and content © 2002-2024 The Digital FAQ, www.digitalFAQ.com
Forum Software by vBulletin · Copyright © 2024 Jelsoft Enterprises Ltd.