doing xvid conversion to kdvd 720*576 and i wanted to know whether to have overscan=1 or overscan=2

Nothing to do with xvid or not.
TO achieve best quality, use 704*576 (and not 720*576) with overscan 2.

thanks dialhot, il do that,

dialhot, i want to convert a ntsc video to kdvd, and keep the ntsc signal

the template i got is 720*480, to achiev the best quality possible on this conversion should i change the resolution to anything else?

Can't you realize yourself that your question doesn't have any answer the way you post it ?

What can I tell if I don't even know the length of the movie ?

Stop asking "what's better, what's better ?". Do some tests by your own and let your eyes decide !

the length of the movie is 3 hours, your right il go off and do some tests,
just asking cos on the post above you said to use 704*576 ( instead of 720*576), i thought there may be a better resolution setting for ntsc, but il try

For sure the answer can't be the same for a 3h movie that it would have been for a 2h one ! I hope you can understand that.

Okay, if that was your question, the answer is : 704*480 will be better.
(but bo not resize ! Cut off 8 piwel on each side. If you resize you will screw up the A/R)

what do u mean by do not resize?

this is what iv been doing

in my script i put in

GripCrop(704, 480, overscan=2, source_anamorphic = false )

and then after i have loaded up my template in tmpg i have changed
the output video size from 720* 480 to 704*480 ( this is because none of the templates have a res. of 704*480

is that okay?

sorry for the hassle

That means do not resize. There is no equivalent

Unfortunally, that will resize.
Start you script with :
And then put the Gripcrop line you have with no change.

Never load the template after the source. Always do the opposite. This way you will control if the avs produce exactly what you want, and you won't let the template take the control.

okay heres my script with the addition u suggested:

AviSource("C:\chan.avi",false)
BlindPP(cpu=4)
Blockbuster(method="noise",detail_min=1,detail_max =3,variance=0.1,seed=1)
Convolution3D(1, 6, 12, 6, 8, 2.8, 0)
Crop(8,0,704,480)
GripCrop(704, 480, overscan=2, source_anamorphic = false)
GripSize(resizer="LanczosResize")
Undot()
TemporalSoften(2,7,7,3,2)
DCTFilter(1,1,1,1,1,1,0.5,0)
GripBorders()
TextSub("C:\chan.srt")

after this i open up tmpg, load up my template, load up my script, then i change the output video to 704*480, is that cool?

what exactly am i resizing here , the source video or the template in tmpg?

how do i avoid resizing? iv had 'GripCrop(XXX, XXX.... in all my scripts

1. Thats not cool, thats right, as you do encode in the same res as the res your avs script does output! Thats the sense that you do ONLY resize/crop etc in avisynth ... TmpgEnc ONLY is purposed to encode but no more to resize!

2. The source Video for shure!
And if you wanna be more exact, by this you do still resize within the 704x576 incl. 16-16-16-16px on each side added so that this will again fit the 704x576, means the effect. Movie area will be downsized from 704x480 to 672x448 and with the above 16pix on each side you'll end up again in 704x480, .... thats the way of Overscan in Grip Crop as it does a resized! overscan.

3. As you do apply to all your Movies an overscan area you cant' avoid a minimal resizing! as explained above in "2."

okay, im a bit confused

so basically if im going to change the res. size in tmpg, i have to make sure iv evened it out in avisynth first,

e.g if my output res. is going to be 704*576,
in my script i should have:

Crop(8,0,704,576)
GripCrop(704, 576, overscan=2, source_anamorphic = false)

that will allow me to change the the video size in tmpg


but if my output is going to be 720*576
(a resolution available in kdvd template) this is not resizing so i just need

GripCrop(720, 576, overscan=2, source_anamorphic = false),

overall by saying 'do not resize' you mean dont change the out put resolution in tmpg, only use whats avalable in the templates, if i do change it i have to do the above



sorry again for being the nusance

Did You read what Dialhot wrote?
Don't make that. Just open up tmpg, first load up your script, then load up your template, and last, in settings, put the output video res (704*480).

Predator DID YOU READ WHAT DIALHOT wrote, he clearly says load up the template, then the script, then the video res. settings

Dialhot

By loading first the template, and then the avs, the settings in tmpgenc change themselves to the ones provided by the script. In other words, you don't have to change anything, it will be done "automatically".
Thas is why I said to you taht this method will assure you that this is THE SCRIPT that controls the things and not the template.

But as always, if you just tested the things before to post, you should have seen them by yourself

No !
I wanted to say DO NOT CHANGE THE SIZE OF THE SOURCE BY MAKING A DIRECT RESISIZING FROM 720 -> 704 !

The crop command CUT the picture, it didn't resize it. You see the difference ?
If you do Crop(704,480) and then Gripcrop(704,480) and then Gripsize, you just CUT the image, and you don't screw up the A/R.
If you do directly GripCrop(704,40 + Gripsize, you will resize the picture (because the source you provide to Gripcrop is 720x480)

I actually wrote the opposite, Neo

Excuseme, is for my bad english. I'm sorry so much.
I'l go on carefuly from now.

finally i think it makes sence to me.

1. if im just doing a normal convversion of 720*480 i dont need to have the crop command, yes?

2.and if i do want to resize to say 704*480, i use the crop command before gripfit, and the resizing is done and i dont have to anything in tmpg beside loading template and my script , yes?

thanks for all the help dialhot, i know i can be a bit much, but thanks, i shall be experimenting 2morow

Yes.

Yes.

thanks Phil