| It is currently Thu Apr 18, 2024 8:36 am |
|
All times are UTC + 1 hour [ DST ] |
 
|
Page 16 of 20 |
[ 587 posts ] | Go to page Previous 1 ... 13, 14, 15, 16, 17, 18, 19, 20 Next |
| Print view | Previous topic | Next topic |
| Author | Message |
|---|---|
|
End of Level Boss   Joined: Fri Mar 16, 2007 5:10 pm Posts: 7535 |
  Blurry but not blocky!
Let's see. LUX-60 is doing the first pass at Quant=3 and it's using a highbitrate matrix. And the second pass has the min quants at 3. With these settings you get an undersize at Avg. DRF ~3.5, so your 4 is still good. I think there is nothing wrong with the Xvid settings, but the 2 lines of your script are compression killers!!! Code: LimitedSharpenFaster(Smode=4,Lmode=3,strength=60) mt_average(AddGrainC(2),AddGrainC(2,seed=10)) Even if the AddGrain line has a small visual impact it still needs A LOT of extra bitrate. I really use it only if the movie is very very compressible. I would say AddGrain is a no in your movie. LSF looks ok, but I would use LSF + Soothe, perhaps something like: Code: dull=last sharp=LimitedSharpenFaster(Smode=4,Lmode=3,strength=70) Soothe(sharp,dull,50) Without AddGrain and using LSF+Soothe I guess you will be near ~LUX-75 or even higher. The result will be closer to your AVS script. |
| Fri Sep 21, 2007 8:31 pm |
|
 |
|
|
Zombie Soup   Joined: Wed Aug 29, 2007 2:04 am Posts: 198 |
well, let's try it then. really like this avs2Xvid thing
 |
| Fri Sep 21, 2007 10:04 pm |
|
|
|
|
Chuck Norris   Joined: Mon Mar 13, 2006 3:42 pm Posts: 16726 Location: Shitopia |
They're giving me a headache though
 Ok that's just shitty avs scripts 
Anyways, the presets are growing bigger inside me foldah 
Added a lotsa new ones. Currently running this, I kept V3HR and for a comptest 100 (?! could be just this source) running: Code: "LUX-87","1","4","1",".",".","2","100","0","heini_mr-sap.xcm","3",".",".",".",".",".",".",".",".",".",".",".",".","Y","." "LUX-87","2","4","3",".",".","2","100","0","heini_mr-sap.xcm",".","2","3","2","4","2","4",".",".","0",".","Y","Y",".","." I wonder how the drf stats are gonna be like.... Been thinking of VHQ 4 settings on some lower ones like J2CD and heini etc. ... Guess it would be better if they also had 4,3 like the higher ones if that would be better at keeping grain/details....  |
| Fri Sep 21, 2007 11:49 pm |
|
|
|
|
Zombie Soup Joined: Wed Aug 29, 2007 2:04 am Posts: 198 |
elguaxo wrote: Blurry but not blocky! Let's see. LUX-60 is doing the first pass at Quant=3 and it's using a highbitrate matrix. And the second pass has the min quants at 3. With these settings you get an undersize at Avg. DRF ~3.5, so your 4 is still good. I think there is nothing wrong with the Xvid settings, but the 2 lines of your script are compression killers!!! Code: LimitedSharpenFaster(Smode=4,Lmode=3,strength=60) mt_average(AddGrainC(2),AddGrainC(2,seed=10)) Even if the AddGrain line has a small visual impact it still needs A LOT of extra bitrate. I really use it only if the movie is very very compressible. I would say AddGrain is a no in your movie. LSF looks ok, but I would use LSF + Soothe, perhaps something like: Code: dull=last sharp=LimitedSharpenFaster(Smode=4,Lmode=3,strength=70) Soothe(sharp,dull,50) Without AddGrain and using LSF+Soothe I guess you will be near ~LUX-75 or even higher. The result will be closer to your AVS script. Well, trid it with that script, but the comptest was exactly the same: 79. In the first page: "<"49 - LUX-40 50-59 - LUX-50 60-69 - LUX-55 70-79 - LUX-60 80-86 - LUX-70 87-100 - LUX-80 101-109 - LUX-85 110-130 - LUX-90 131-145- LUX-95 146-160 - LUX-97 160">" - LUX-99 But in the tables I can see some more. I should have used LUX 75, right? Didn't notice it so I went to LUX 60... 
maybe I'll go back to the old script and see if I get any difference.... |
| Sat Sep 22, 2007 4:55 pm |
|
|
|
|
End of Level Boss Joined: Fri Mar 16, 2007 5:10 pm Posts: 7535 |
dvdfernandes wrote: Well, trid it with that script, but the comptest was exactly the same  Something is wrong then. |
| Sat Sep 22, 2007 5:00 pm |
|
|
|
|
Zombie Soup Joined: Wed Aug 29, 2007 2:04 am Posts: 198 |
elguaxo wrote: dvdfernandes wrote: Well, trid it with that script, but the comptest was exactly the same Something is wrong then.Back to the old script. "Compchecking" Will get back soon |
| Sat Sep 22, 2007 5:02 pm |
|
|
|
|
Chuck Norris Joined: Mon Mar 13, 2006 3:42 pm Posts: 16726 Location: Shitopia |
But the old script is compression murder
|
| Sat Sep 22, 2007 5:08 pm |
|
|
|
|
Zombie Soup Joined: Wed Aug 29, 2007 2:04 am Posts: 198 |
Old script with AddGrain. Saem result... 79
 |
| Sat Sep 22, 2007 5:25 pm |
|
|
|
|
Chuck Norris Joined: Mon Mar 13, 2006 3:42 pm Posts: 16726 Location: Shitopia |
Ok post the entire .avs of the new script.
Without addgrain... perhaps you left a poop there you missed  |
| Sat Sep 22, 2007 5:30 pm |
|
|
|
|
Zombie Soup Joined: Wed Aug 29, 2007 2:04 am Posts: 198 |
Code: LoadPlugin("C:\PROGRA~1\GORDIA~1\DGMPGDec\DGDecode.dll") LoadPlugin("C:\PROGRA~1\GORDIA~1\AviSynthPlugins\UnDot.dll") mpeg2source("C:\Documents and Settings\David\Ambiente de trabalho\dd\DirtyDancing.d2v") crop(4,4,712,564) LanczosResize(704,384) LoadPlugin("C:\Programas\GordianKnot\AviSynthPlugins\RemoveGrain\RemoveGrain.dll") LoadPlugin("C:\Programas\GordianKnot\AviSynthPlugins\RemoveGrain\Repair.dll") LoadPlugin("C:\Programas\GordianKnot\AviSynthPlugins\masktools-v2.0a30\mt_masktools.dll") LoadPlugin("C:\Programas\GordianKnot\AviSynthPlugins\mvtools-v1.6.2\mvtools.dll") LoadPlugin("C:\Programas\GordianKnot\AviSynthPlugins\Masktools\MaskTools.dll") LoadPlugin("C:\Programas\GordianKnot\AviSynthPlugins\degrainmedian\degrainmedian.dll") Import("C:\Programas\GordianKnot\AviSynthPlugins\LimitedSharpenFaster\LimitedSharpenFaster.avs") Import("C:\Programas\GordianKnot\AviSynthPlugins\Soothe\Soothe.avs") dull=last sharp=LimitedSharpenFaster(Smode=4,Lmode=3,strength=70) Soothe(sharp,dull,50) |
| Sat Sep 22, 2007 5:34 pm |
|
|
|
|
Chuck Norris Joined: Mon Mar 13, 2006 3:42 pm Posts: 16726 Location: Shitopia |
Well perhaps the problem is with LOAD and IMPORT all over the script between the actual functions? Try this:
Code: LoadPlugin("C:\PROGRA~1\GORDIA~1\DGMPGDec\DGDecode.dll") LoadPlugin("C:\PROGRA~1\GORDIA~1\AviSynthPlugins\UnDot.dll") LoadPlugin("C:\Programas\GordianKnot\AviSynthPlugins\RemoveGrain\RemoveGrain.dll") LoadPlugin("C:\Programas\GordianKnot\AviSynthPlugins\RemoveGrain\Repair.dll") LoadPlugin("C:\Programas\GordianKnot\AviSynthPlugins\masktools-v2.0a30\mt_masktools.dll") LoadPlugin("C:\Programas\GordianKnot\AviSynthPlugins\mvtools-v1.6.2\mvtools.dll") LoadPlugin("C:\Programas\GordianKnot\AviSynthPlugins\Masktools\MaskTools.dll") LoadPlugin("C:\Programas\GordianKnot\AviSynthPlugins\degrainmedian\degrainmedian.dll") Import("C:\Programas\GordianKnot\AviSynthPlugins\LimitedSharpenFaster\LimitedSharpenFaster.avs") Import("C:\Programas\GordianKnot\AviSynthPlugins\Soothe\Soothe.avs") mpeg2source("C:\Documents and Settings\David\Ambiente de trabalho\dd\DirtyDancing.d2v") crop(4,4,712,564) LanczosResize(704,384) dull=last sharp=LimitedSharpenFaster(Smode=4,Lmode=3,strength=70) Soothe(sharp,dull,50) |
| Sat Sep 22, 2007 5:38 pm |
|
|
|
|
Zombie Soup Joined: Wed Aug 29, 2007 2:04 am Posts: 198 |
same shit:
 79 |
| Sat Sep 22, 2007 6:06 pm |
|
|
|
|
Chuck Norris Joined: Mon Mar 13, 2006 3:42 pm Posts: 16726 Location: Shitopia |
Try Seesaw, can't remember what you need to load right now except the seesaw.avs
Code: mpeg2source("C:\Documents and Settings\David\Ambiente de trabalho\dd\DirtyDancing.d2v") crop(4,4,712,564) LanczosResize(704,384) a= last b=a.degrainmedian(mode=5) SeeSaw(a,b, NRlimit=3, NRlimit2=4, Sstr=1.5, Slimit=1, Spower=2, Sdamplo=6, Szp=16, soothet=70) |
| Sat Sep 22, 2007 6:09 pm |
|
|
|
|
Zombie Soup Joined: Wed Aug 29, 2007 2:04 am Posts: 198 |
SeeSaw (settings above)
Somethings gotta be wrong... 79 again.  |
| Sat Sep 22, 2007 8:18 pm |
|
|
|
|
End of Level Boss Joined: Fri Mar 16, 2007 5:10 pm Posts: 7535 |
 I think it's your pITo.
Now really, something must be wrong. _________________ Old2New SBiG URL Replacement Script | SBiG searchplugin my AviSynth Plugins folder | Show Just Image 2 | Doom10 |
| Sat Sep 22, 2007 8:32 pm |
|
|
|
|
Zombie Soup Joined: Wed Aug 29, 2007 2:04 am Posts: 198 |
I think It was really my fault. Forget. Went to sleep and a light came upon me. I was doing this all wrong. Compchecking again with LSF+Soothe
 
[EDIT] 84%. I'll be at the corner shoving the keyboard up my ass. thx guys 
[EDIT2] LUX 85, right? |
| Sat Sep 22, 2007 8:36 pm |
|
|
|
|
Chuck Norris Joined: Mon Mar 13, 2006 3:42 pm Posts: 16726 Location: Shitopia |
No no, that's too high, if I remember correctly... try with LUX-70
 |
| Sun Sep 23, 2007 11:35 am |
|
|
|
|
Zombie Soup Joined: Wed Aug 29, 2007 2:04 am Posts: 198 |
Don't mind If I ask. What are the ranges for LUX 75? Got a 73 % comptest right now.
 |
| Mon Sep 24, 2007 1:14 am |
|
|
|
|
Chuck Norris Joined: Mon Mar 13, 2006 3:42 pm Posts: 16726 Location: Shitopia |
Hmmm I'd say somewhere between 85-90, kinda non-useful preset actually...
WTF how did you end up 73% now?! 
For 73% best option would be V3LR preset or perhaps heini6of9 at min quant 3.
What's your script? |
| Mon Sep 24, 2007 1:19 am |
|
|
|
|
Zombie Soup Joined: Wed Aug 29, 2007 2:04 am Posts: 198 |
Lux Delux wrote: Hmmm I'd say somewhere between 85-90, kinda non-useful preset actually... WTF how did you end up 73% now?! What's your script? Different encoding  Quote: For 73% best option would be V3LR preset or perhaps heini6of9 at min quant 3. Could you put that in LUX - something? Would that be LUX - 60? Code: Undot() backward_vec2 = last.MVAnalyse( isb = true, delta = 2, pel = 2, overlap=2, blksize=16, sharp=1, idx = 1 ) backward_vec1 = last.MVAnalyse( isb = true, delta = 1, pel = 2, overlap=2, blksize=16, sharp=1, idx = 1 ) forward_vec1 = last.MVAnalyse( isb = false, delta = 1, pel = 2, overlap=2, blksize=16, sharp=1, idx = 1 ) forward_vec2 = last.MVAnalyse( isb = false, delta = 2, pel = 2, overlap=2, blksize=16, sharp=1, idx = 1 ) dull = last.MVDegrain2( backward_vec1, forward_vec1, backward_vec2, forward_vec2, thSAD=150, idx=1 ) sharp = dull.LimitedSharpenFaster(smode=4,strength=30) Soothe( sharp, dull, 40 ) |
| Mon Sep 24, 2007 1:24 am |
|
|
|
|
Chuck Norris Joined: Mon Mar 13, 2006 3:42 pm Posts: 16726 Location: Shitopia |
I don't have V3LR in my presets, since I experiment a lot with the crappers. Open 25h and look for V3LR, I think it is LUX-60 or 65..  |
| Mon Sep 24, 2007 1:26 am |
|
|
|
|
Chuck Norris Joined: Mon Mar 13, 2006 3:42 pm Posts: 16726 Location: Shitopia |
dvdfernandes wrote: Code: Undot() backward_vec2 = last.MVAnalyse( isb = true, delta = 2, pel = 2, overlap=2, blksize=16, sharp=1, idx = 1 ) backward_vec1 = last.MVAnalyse( isb = true, delta = 1, pel = 2, overlap=2, blksize=16, sharp=1, idx = 1 ) forward_vec1 = last.MVAnalyse( isb = false, delta = 1, pel = 2, overlap=2, blksize=16, sharp=1, idx = 1 ) forward_vec2 = last.MVAnalyse( isb = false, delta = 2, pel = 2, overlap=2, blksize=16, sharp=1, idx = 1 ) dull = last.MVDegrain2( backward_vec1, forward_vec1, backward_vec2, forward_vec2, thSAD=150, idx=1 ) sharp = dull.LimitedSharpenFaster(smode=4,strength=30) Soothe( sharp, dull, 40 ) That script needs changing. No point in Undot at all and why is thsad=150? Your LSF part is also fucked up. Here ya go: Code: source=last denoised=DegrainMedian(mode=5) backward_vec2 = MVAnalyse(denoised,isb = true, delta = 2, pel = 2, overlap=4, sharp=1, idx = 1) backward_vec1 = MVAnalyse(denoised,isb = true, delta = 1, pel = 2, overlap=4, sharp=1, idx = 1) forward_vec1 = MVAnalyse(denoised,isb = false, delta = 1, pel = 2, overlap=4, sharp=1, idx = 1) forward_vec2 = MVAnalyse(denoised,isb = false, delta = 2, pel = 2, overlap=4, sharp=1, idx = 1) MVDegrain2(source,backward_vec1,forward_vec1,backward_vec2,forward_vec2,thSAD=100,idx=2) dull = last sharp = dull.LimitedSharpenFaster(smode=4,strength=30) Soothe( sharp, dull, 40 ) |
| Mon Sep 24, 2007 1:29 am |
|
|
|
|
Zombie Soup Joined: Wed Aug 29, 2007 2:04 am Posts: 198 |
Lux Delux wrote: :shy: I don't have V3LR in my presets, since I experiment a lot with the crappers. Open 25h and look for V3LR, I think it is LUX-60 or 65.. heini_lr_6o9, right? LUX-60. I was trying to reach for LUX 70, but I guess It will be hard. Wait, wrong script up there 73% Code: Undot() backward_vec2 = last.MVAnalyse( isb = true, delta = 2, pel = 2, overlap=2, blksize=16, sharp=1, idx = 1 ) backward_vec1 = last.MVAnalyse( isb = true, delta = 1, pel = 2, overlap=2, blksize=16, sharp=1, idx = 1 ) forward_vec1 = last.MVAnalyse( isb = false, delta = 1, pel = 2, overlap=2, blksize=16, sharp=1, idx = 1 ) forward_vec2 = last.MVAnalyse( isb = false, delta = 2, pel = 2, overlap=2, blksize=16, sharp=1, idx = 1 ) dull = last.MVDegrain2( backward_vec1, forward_vec1, backward_vec2, forward_vec2, thSAD=100, idx=1 ) sharp = dull.LimitedSharpenFaster(smode=4,strength=30) Soothe( sharp, dull, 50 ) New one, will check the comp difference... Code: Undot() backward_vec2 = last.MVAnalyse( isb = true, delta = 2, pel = 2, overlap=2, blksize=16, sharp=1, idx = 1 ) backward_vec1 = last.MVAnalyse( isb = true, delta = 1, pel = 2, overlap=2, blksize=16, sharp=1, idx = 1 ) forward_vec1 = last.MVAnalyse( isb = false, delta = 1, pel = 2, overlap=2, blksize=16, sharp=1, idx = 1 ) forward_vec2 = last.MVAnalyse( isb = false, delta = 2, pel = 2, overlap=2, blksize=16, sharp=1, idx = 1 ) dull = last.MVDegrain2( backward_vec1, forward_vec1, backward_vec2, forward_vec2, thSAD=150, idx=1 ) sharp = dull.LimitedSharpenFaster(smode=4,strength=30) Soothe( sharp, dull, 40 ) |
| Mon Sep 24, 2007 1:30 am |
|
|
|
|
Chuck Norris Joined: Mon Mar 13, 2006 3:42 pm Posts: 16726 Location: Shitopia |
Just copy/paste what I gave you up there
 |
| Mon Sep 24, 2007 1:31 am |
|
|
|
|
Zombie Soup Joined: Wed Aug 29, 2007 2:04 am Posts: 198 |
EDITED at the same time. Didn't really understood why that was so fucked up...
(NOOBER)
trying your thing [EDIT] thSAD=150 was defined according to the amount of noise in the source... |
| Mon Sep 24, 2007 1:34 am |
|
|
|
|
Chuck Norris Joined: Mon Mar 13, 2006 3:42 pm Posts: 16726 Location: Shitopia |
well the dull part of soothe was actually inside the mvtools part of the script, and that was a no-no.
Undot will just kill extra details without doing much since you're using mvdegrain already. Though, I dunno anything about your source actually, so I've no idea how much cleaning-sharpening it really needs |
| Mon Sep 24, 2007 1:42 am |
|
|
|
|
Zombie Soup Joined: Wed Aug 29, 2007 2:04 am Posts: 198 |
Lux Delux wrote: Code: source=last denoised=DegrainMedian(mode=5) backward_vec2 = MVAnalyse(denoised,isb = true, delta = 2, pel = 2, overlap=4, sharp=1, idx = 1) backward_vec1 = MVAnalyse(denoised,isb = true, delta = 1, pel = 2, overlap=4, sharp=1, idx = 1) forward_vec1 = MVAnalyse(denoised,isb = false, delta = 1, pel = 2, overlap=4, sharp=1, idx = 1) forward_vec2 = MVAnalyse(denoised,isb = false, delta = 2, pel = 2, overlap=4, sharp=1, idx = 1) MVDegrain2(source,backward_vec1,forward_vec1,backward_vec2,forward_vec2,thSAD=100,idx=2) dull = last sharp = dull.LimitedSharpenFaster(smode=4,strength=30) Soothe( sharp, dull, 40 ) 0.56 fps  |
| Mon Sep 24, 2007 3:53 pm |
|
|
|
|
Zombie Soup Joined: Wed Aug 29, 2007 2:04 am Posts: 198 |
Weell, I wanted to try a test encode with a small 20MB VOB. Just to have a peak at the final result
created the txt file and wrote "LUX-60 1500kbps" but came up with small .avi with bitrate=2200kbps.. Isn't this the right way to do it? |
| Mon Sep 24, 2007 8:18 pm |
|
|
|
|
End of Level Boss Joined: Fri Mar 16, 2007 5:10 pm Posts: 7535 |
dvdfernandes wrote: Isn't this the right way to do it? No!
Encode the whole movie, but add this at the end of your script: SelectRangeEvery(233, 14) It's the same thing beeing used in a comptest, but you will do 2 passes. Edit: don't play around with the SelectRangeEvery values. People smarter than you and me discovered these settings are good for this kind of tests. |
| Mon Sep 24, 2007 8:32 pm |
|
|
|
|
Zombie Soup Joined: Wed Aug 29, 2007 2:04 am Posts: 198 |
elguaxo wrote: dvdfernandes wrote: Isn't this the right way to do it? No! Encode the whole movie, but add this at the end of your script: SelectRangeEvery(233, 14) It's the same thing beeing used in a comptest, but you will do 2 passes. But... isnt' that just like doing a wider comptest? /Maybe I disnt understand what I read at Doom9 - your questio, actually )elguaxo wrote: Edit: don't play around with the SelectRangeEvery values. People smarter than you and me discovered these settings are good for this kind of tests. Don't plan on doing it, but thanks for the heads up!  |
| Mon Sep 24, 2007 8:41 pm |
|
|
|
|
|
Page 16 of 20 |
[ 587 posts ] | Go to page Previous 1 ... 13, 14, 15, 16, 17, 18, 19, 20 Next |
|
All times are UTC + 1 hour [ DST ] |
Who is online |
Users browsing this forum: No registered users and 13 guests |
| You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot post attachments in this forum |