Okay, I'll try it out. Thanks for you help, Boulder. BTW, should I use 24 for sample lenght also - on all source material (NTSCFilm, PAL & NTSC)?

Yes, it would make sense..until someone comes up with a sample length that gives a better prediction.

Thank you again, Boulder! You've been a great helper indeed...
:wink:

okay, with all the latest talk on motion adaptive scripts and detecting scene changes, blah blah blah... got me thinking, should we be making any more changes to our tmpgenc settings than those in the old kvcdx3 template? I've read to change the motion search precision to fast (motion estimate search). But what about Detect Scene change (in the GOP tab) and the Special Settings (Quantize Matrix tab)? I just noticd there is an unchecked box for "No motion search for still picture part by half pixel"... Thanks guys!

That was already changed and SET in the http://www.kvcd.net/video.en1 file, ready for ToK.That actually makes the picture worse :wink:

-kwag

right Boulder :!:

for the first time from days this is the first time
that i got high CQ using the Boulder recomendations :!:
maybe somebody remember a weeks ago when i post about low CQ.

i did the test twice and ToK give me
CQFinal CQ: 70,349 :D :D :D
and before was CQ: 56,467 :arrow: huge difference,don't?

i only uncheck the "Auto GOP" and wrote:
1,24,2,1,24!
uncheck the "new faster prediction" and 2nd group of passes
than i wrote this values:
precision% 1,00 and speed up by 10 !

new video.en1 (thanks Kwag) only changing this line to:
Job.MPEG.Video.VideoEncodeMode=MPEGVideoEncoder_Vi deoEncodeMode_interlace
cos my source is 29,970,ntsc.

the CQ: 70,349 is perfect for my desired size now!
:D
i'm happy and encoding now!

thank you all friends, in special Kwag and Boulder for this result!
:D :D :D

should NTSC film users be unchecking auto GOP too? and manually entering a value?

For film material the auto GOP length should be 24 already as its 23.976fps.

@Boulder,
You wouldn`t use that GOP (24) with Anime/Cartoon...would you???


*******************************
The Devil`s always.....in the Details!

After some more tests a sampler length of 15 isn't accurate for me :evil: .
I had to reinstall my Windows (even manual repair failed) so i've only done 2 more tests, but i had an idea:
Why not taking a longer sample size to make a better prediction with the adaptive script :?: I had good results with a Sampler length of 50, maybe 48 would be better. I thought that a bigger sample size than GOP would screw up Prediction or Sampler() but it didn't.

Kwag, do you know why TMPGEnc encodes with a GOP of 24 instead of 25 with PAL-Sources :?:

EDIT: Here is a comparison on a short movie (PAL):

I just finished encoding "Enough" (PAL) with GOP and sample size set to 24. And this is what I got:

Wanted size: 726 XXX XXX
Encoded size: 721 XXX XXX
(Don't have the exact numbers. I'm at work now)

AMAZING! A difference less than ~0.7% :D !!! Final muxed movie was 795MB. I changed ToK's setting to what kwag recommended in an earlier post and used kwag's modified video.en1 file. Try it out all you PAL encoders out there. Thank you very much guys! Maybe I'll be able to sleep at night again now :wink: !

Hi, all
My way for PAL prediction is not fully tested (only two movies so far - an AVI and a DVD) but gave me for first time correct file sizes.
Sample=1% of movie with the line in the script SelectRangeEvery(2500,25)
run a sample at CQ100 without black borders. Mark the file size. Run the sample at CQ100 borders added. Mark the file size. Find the difference of the file sizes in %.
Calculation of desired sample size
samplesize=((fullvidoesize/totalframes)*sampleframes) + (% of the difference you found before). Hmm, sorry for the bad english.
An example:
A movie has 137145 frames. Final video size is 740912KB. Sample has 1375 frames. Sample size = (740912/137145)*1375 = 7428.
CQ100 without borders = 15275 KB
CQ100 with borders = 14569 KB
Difference = 4.6%
4.6% from 7428 = 341
7428+341=7769 -> this is your real size of the sample.
The results. AVI - required 730xxx KB, final encode 723xxxKB.
DVD (the example above) encoded with CQ57 with sample size 7785 (a bit higher than the desired 7769). Required size 740912, final encode 742xxx (a bit higher).
Well, it need to be more carefully tested.

I can't say because I don't encode any anime or cartoons. Surely Kwag knows this stuff better than me so you're better off with his recommendations.

No, becauese I don't encode PAL :)
But I'll take a look :!: Maybe it's a bug in TMPEG :?:
BTW, have you tried encoding with ToK with the screenshot I posted earlier :?:
Here it is again:

http://www.digitalfaq.com/archives/i.../2003/06/7.jpg

-kwag

Well I just have to post this one :mrgreen:

Here's my encode of K19, which just finished 5 minutes ago. Look at the log, and see for yourself :D

A difference of 0.39% for a movie that is 138 minutes long :mrgreen: :mrgreen: :mrgreen:
Thanks again to Jellygoose for suggesting to change asharp to unfilter on the adaptive part of the script.
I'm updating the current script now. BTW, the settings I used for this movie are exactly as the screenshot above this post :!:

I'm uptating the

Thats great Kwag :D ,just a quick question with the new change in the script do you think we will be able to go back to the faster prediction in tok or still use the longer one.

Thanx

Hey Bigg, I just tried Kwags settings vs tenras Fast prediction

pretty much close dont you think :idea:

I'm doing a fast "Tenra's" CQ prediction right now to see what CQ it finds, and compare that to the CQ I got with the full prediction :)
I'll post the resulting CQ here in a while...

-kwag

Yes ovg64,i think we all could live with that :lol:

Looks like it might be a good night :D

YES IT IS :!: I'M GOING TO BUY SOME BEER :!: :mrgreen:

The script is now "FREEZED", and file prediction is again considered STABLE :!: :mrgreen:

-kwag

I guess we skin the pig ehhh :mrgreen:

Edit: you all did :!: :mrgreen:

[quote="kwag"]It's Miller Time :!: woot. Starting to run my tests and I can't wait to see the results. :D :D :D

Here you go guys :)
Here's the small sampler of K19 (~1MB), so that's how the whole 138 minutes looks like on one CD-R :mrgreen:

http://www.kvcd.net/k19-fast-predict...-cq-63.209.mpg

Enjoy :)
-kwag

So do we continue using tok 0.0.5.2 or can we go back to 0.0.5.3 and what settings should we use?



David

Hi David,

I havent tried that yet :!:
I'll give that a shot too, and see if the CQ matches a very close CQ as 0.0.5.2.
If it does, then we can use 0.0.5.3 :)
I'll run both predictions. The full, and Tenra's "Fast" prediction.
Only if both match, I would then use 0.0.5.3.

-kwag

Here's the result with ToK 0.0.5.3, which uses "SelectRangeEvery()" instead of "Sampler()" for prediction.
This was on my "K-Pax" movie.

CQ found using "Tenra's" Fast prediction: 70.138
CQ found using normal (long) prediction: 69.940

The CQ found by ToK 0.0.5.2 was 65.51 using Tenra's "Fast" mode, and the full prediction was almost the same number. And that is the correct CQ, because I left the movie encoding last night, and the result this morning was +1.73% accurate :)

This proves that the "SelectRangeEvery()" is a problem for prediction, but SansGrip's method of selecting frame range is excelent :)
So I would stick to 0.0.5.2. Other than the file prediction method used, there's no difference from 0.0.5.2 and 0.0.5.3

-kwag

Hi Guys, last night i encodes (A guy thing) 1:41 min 70.61% CQ on one CD output video was 710MB, audio was 83MB for a total file size of 793MB.
This was using tok 5.2 n tenras fast prediction, so i guess this turkey is really cooked now. :D

Hi ovg64,

What was your % accuracy on that encode :?:

-kwag

and please Osvaldo,
write all details,..if is mpge1 or 2,the "resize",time of the source,
everything cos i loose the high prediction again,
then i need of all informations.
my CQ back to 56 "house" again.

Hey Guys, here you go; :D

Mpeg 1
712 mb video wanted 710 obtain
83 mb audio at 112 kbps
1, 24, 2, 1, 24 gop

Fast Prediction *
Factor 1.00
Sample Auto *
# sample/min. 1

1 Group Pass
presicion % 0.50
speed up by 10
10:00 % presicion

80min. CD -1

ovg64: That's great that you got such an accurate prediction! I've been wondering lately at what point should we predict using 2 samples/min instead of the default 1? I've pretty much been raising it to 2 whenever the movie is shorter than 2:00:00 long, but I see this one is 1:40:00. Any opinions?

I dont know but it sims like sample() is more accurate than selectevery()
use in tok 5.3 at least in Avs 2.52, cause yesterday i use 3 setting mine, kwags n Tenras and they all came to around the same 70% CQ for this same movie. :)

really strange Osvaldo.
my source:

Resolution (fps):480x480 (29,970 fps)
Total Frames: 159173
Total Time : 01:28:31

Audio Size: 84.967.000
Required Video Size: 734.224.000

:arrow: and my best CQ56,255!

why i with: less resize, less time movie got less CQ :?:
:? :? :? :?

i need a BIG help :!:
:(

What are you ecoding :?: National Security :?: :D 1:28:31 time
post your script i think maybe one of your filter might be doing
this are u using DGbob :?: you are using 29fps that might be another
reason. :!:

no Osvaldo,no DGBob no more i cahnge to fielddeinterlace!

see my script:

MaxTreshold=1.50
nf=0 # Current frame.

Mpeg2Source("D:\La Luna\Segment2\DVD2AVI_PROJECT_FILE.d2v")

FieldDeinterlace()
undot()
Limiter()
asharp(2,4)
GripCrop(480,480,overscan=1,source_anamorphic=fals e)
GripSize(resizer="BicubicResize")
STMedianFilter(8,32,0,0)
MergeChroma(blur(MaxTreshold))
MergeLuma(blur(0.2))
ScriptClip("nf=YDifferenceToNext()"+chr(13)+"nf>2. 5?\
unfilter(-(fmin(round((nf/0.5)),100)),-(fmin(round((nf/0.5)),100))):\
TemporalSoften(2,7,7,3,2)")

GripBorders()
Limiter()

function fmin(float f1,float f2){return(f1<f2)?f1:f2}

thanks in adbvance!
:wink:

ps:
Jell is helping me too in the "best scripts" thread...
of course,something is wrong in my system and
i don't know what is ?!?!?
:?

@jorel

Look at the Optimal Scripts forum... :wink:

Hey Jo your script looks ok to me, i dont know maybe tok doesn't pred. well with 29 fps. Maybe you want to experiment a bit i would maybe put undot after the resize n take out the deinterlance filter just to see what kind of CQ you get or maybe try another deinterlance like tomsmocomp
for 2.52 and experiment a bit you know im not a brain like you but i some how solve my problems :wink:

I can confirm this for PAL-Sources. Prediction is very accurate for PAL using ToK 0.0.5.2, a Sampler Length of 24 and the newest adaptive script, even with Tenras new prediction. The prediction was 0,3 % accurate on a middle action movie.
Thanks to Kwag, Jellygoose and all others for helping figure this out.
Maybe we should contact Pegasys-Inc. to determine if theres a bug in PAL-GOP encoding :roll: .

Does this mean that we can continue using v0.0.5.3 and just substitute
SansGrip`s "Sampler"??? (sorry if this has been answered already)

Or do we need v0.0.5.2,<---(???) If so...where can we find that version??


*******************************
The Devil`s always.....in the Details!

@ Jell
thanks, i read all there and waiting your helper answers. :)

@ vico1
good question,i want to know the same.
:wink:

@ Osvaldo :)
tomsmocomp is worse,decrease the CQ.
i change deinterlance to just before and just after resize and change just a little...insignificant.
fielddeinterlace is in the better place in the script,
(thanks to Boulder and Phil for big explanations about deinterlace and ivtc)

you wrote:
"...i'm not a brain like you..." 8O
you are in better perspective,you have a good brain working,
mine is lost. :roll: