Set the CQ value to 80 and encode the sample strip. If it's too big, lower the CQ value. If it's too small, raise the CQ value.
Sample strip is = sample clip. Sample prediction clip. Sample test file.

-kwag

Imoa

Oke, now is that clear for me
But should I add something to my script, now when I'm useing file predicion:

LoadPlugin("C:\Documents and Settings\Lars\Skrivbord\Tobias\Program\Blandat\Fit CD\MPEG2DEC.dll")
LoadPlugin("C:\Documents and Settings\Lars\Skrivbord\Tobias\Program\Blandat\Fit CD\GripFit_preview.dll")
LoadPlugin("C:\Documents and Settings\Lars\Skrivbord\Tobias\Program\Blandat\Fit CD\LegalClip.dll")
LoadPlugin("C:\Documents and Settings\Lars\Skrivbord\Tobias\Program\Blandat\Fit CD\Sampler.dll")
LoadPlugin("C:\Documents and Settings\Lars\Skrivbord\Tobias\Program\Blandat\Fit CD\DustV5.dll")
LoadPlugin("C:\Documents and Settings\Lars\Skrivbord\Tobias\Program\Blandat\Fit CD\TemporalCleanerOld.dll")

AviSource("C:\Documents and Settings\Lars\Skrivbord\Tobias\AmericanPie2.avi")

LegalClip()
GripCrop( width= 352, height=576, overscan=1 )
GripSize()
SpaceDust()
TemporalCleaner()
GripBorders()
LegalClip()
#Sampler(length=24)
## MPEG size = ((Total frames/MovieTimeInMinutes)/24) * MPEG sample file size ##

Or should I just high the CQ and encode with this script

The line that reads:
#Sampler(length=24)

Edit that line and remove the #, so it will read:
Sampler(length=24)

After you get the sample size you want by playing with prediction, edit again and insert the # again. Then you'll be ready to encode the complete movie.

-kwag

I really dont know what to write in "Sample Points" and "Error Margin". They are now on Sample points-->100 and Error margin-->5%. Havent touch them since I installed File Prediction. And I've allready removied "#" from the script and followed the "mini how to" guide as you made to find my sample size. The sample size came out as 23.09mb as I've already told you. But the File prediction as I havent used before now, gives me a different sample size, 26.15mb I selected my source movie and my audio file. Then I set the "Quantity" on 2x80 (because I'm gonna fit my movie on 2 cd's).
And left the "Sample Points" on 100 and the "Error Margin" on 5 % because I dont know what will happend if I change it

Maybe it's just so that the File Prediction is more exactly when it gives me the sample size (26.15mb ). So is it just to ignore the mini how to manual and take the 26.15 sample size instead of the "manual doing samle size" 23.09?

It seems to me that the File Prediction is faster and more exactly when it gives me the sample size, or have I misunderstand it all?

But the big question is what I should write in the "Sample Points" bar and the "Error margin" bar...

//Wolfi

I think that the sample points value should equal minutes-in-movie and error margin should be 2% as the compensation factor is 0.98 at the moment.

I think that make sence Boulder But...what is the compensation factor havent heard obout it...

//Wolfi

Oke, my sample size is 24,92 mb. I took away "#" from sampler() and had the CQ on 97. I ran the avs. in TMPGEnc, encoded, and my test file came out a bit lower then 24,92 mb, so I highed the CQ to 100 and the result were 24,15 mb. So I suppose I've done my first file prediction now I'm gonna put back "#" to sampler() and encode with the CQ 100 :P

I hope I've done everything right now

//Wolfi

Yippie ka yeah I hit the target on 1,44gb exactly

KVCD RoX