Log in

View Full Version : Problem using the Tweak filter...


lovelegend
22nd July 2002, 02:19
Since I found that using TMPGEnc to convert DVDrips(avi) to VCD(mpg-1), the output would be darker then the input. Therefore, I was trying to use avisynth + tweak filter (in order to increase the brightness of the source) and then frameserve it into TMPGEnc. However, the output looks like having very low color depth... It's hard to explain, so I captured a screen:

http://fukuoka.cool.ne.jp/lovelegend/TweakResult.bmp

Here is my script:
AVISource("path1\movie.avi")
LoadPlugin("path2\Tweak.dll")
Tweak(bright=10, cont=1.0)

Do you know what did I do wrong and how could I solve this problem? Thanks anyway! :)


P.S. I did play the avs file in windows media player and it had no problem. Also, I also tried another way: Encode the file first, and then write a script to increase brightness of the output file (i.e., the mpg file):

DirectShowSource("path1\movie.mpg")
LoadPlugin("path2\Tweak.dll")
Tweak(bright=10, cont=1.0)

and then use the TMPGEnc's mpeg tools to combine it with the audio. However, TMPGEnc (and bbmpg) didn't accept it as the source for mpg... :(

droolian01
22nd July 2002, 02:39
Hi there, why don't you just increase the colour saturation, as it seems that by brightening the image you have washed the colours out.

try:
tweak(bright=10, sat=1.2)

n.b. cont=1 is the default figure and does precislly nothing to the contrast, so just omit it.
Just play with 'bright' and 'sat' only, 'cont' can easily mess up things.
Good luck

lovelegend
22nd July 2002, 05:13
droolian01, I did try what you told me to do, and it still couldn't solve the problem.... :(
But still thanks for your reply and help! :)

dividee
24th July 2002, 19:42
Have a look at this thread:
http://forum.doom9.org/showthread.php?s=&threadid=27932

lovelegend
27th July 2002, 10:54
Thanks dividee for replying... but new question's arised... :D
Since I'm a newbie, I'm not familiar with colorspaces (I've read the link provided in Q&A, but it's just the history and telling you how to convert YUV from RGB). My question is, what's the difference between encoding in YUV space and encoding in RGB space? (I mean in general, for my case converting to RGB of course could solve the color depth problem...). Or there is no difference, it just depends on the colorspace that the program can work on?
And is the output of TMPGEnc in YUV or RGB space?

I would appreciate if you could provde me some threads or websites discussing about this, because I know I don't have much knowledge on this aspect...

P.S. And what's the difference between YUY2 and YUV? :rolleyes: