zoinbergs
30th July 2007, 02:28
Hey folks!
Love this forum -- just wanted to say thanks first for all your wonderful help!
Now I have a quick question if that's okay..
Right now I'm cleaning up some DVD footage, and I need to use avisynth for some filtering. I'm pretty sure I've got the process down, but could somebody please confirm that what I am doing is in fact, the correct thing to do?
(Assume I'm encoding frame-based, interlaced video, with no flags, using an alternate scanning mode and non-linear quantisizer scale)
1. Use DGIndex to build a *.d2v file
2. Use avisynth and the lines:
LoadPlugin("...\DGDecode.dll")
MPEG2Source("...\example.d2v")
<filters go here>
ConvertToRGB(interlaced=true)
3. Import *.avs file into TMPGEnc 4 Xpress
4. Encode
Now I'm pretty sure I don't have to mess with any levels or their expansion (TV-->PC or vice versa) -- but I could be wrong. Could somebody confirm this? It's what I'm most concerned about.
Also, I noticed several arguments I can set for ConvertToRGB(). They have to do with "matrix" settings such as rec.601 / pc.601 / rec.709 / pc.709 -- do I need not mess with these settings either? Is the default okay?
I always thought YV12 used a luminosity scale of (16,235) -- that was stretched to (0,255) using ConvertToRGB() -- that was then shrunk back down to (16,235) via TMPGEnc 4 Xpress during the encode. Could somebody confirm if this is indeed happening? Or do the levels always stay at (16,235) all throughout the process?
Both before and after clips have identical luminosity, which eases my worrying.. but I'd love to know what goes on behind the scenes to better prepare me for more advanced encoding.
Thanks so much to anybody kind enough to shed some light (pun intended) on the subject! ;)
Love this forum -- just wanted to say thanks first for all your wonderful help!
Now I have a quick question if that's okay..
Right now I'm cleaning up some DVD footage, and I need to use avisynth for some filtering. I'm pretty sure I've got the process down, but could somebody please confirm that what I am doing is in fact, the correct thing to do?
(Assume I'm encoding frame-based, interlaced video, with no flags, using an alternate scanning mode and non-linear quantisizer scale)
1. Use DGIndex to build a *.d2v file
2. Use avisynth and the lines:
LoadPlugin("...\DGDecode.dll")
MPEG2Source("...\example.d2v")
<filters go here>
ConvertToRGB(interlaced=true)
3. Import *.avs file into TMPGEnc 4 Xpress
4. Encode
Now I'm pretty sure I don't have to mess with any levels or their expansion (TV-->PC or vice versa) -- but I could be wrong. Could somebody confirm this? It's what I'm most concerned about.
Also, I noticed several arguments I can set for ConvertToRGB(). They have to do with "matrix" settings such as rec.601 / pc.601 / rec.709 / pc.709 -- do I need not mess with these settings either? Is the default okay?
I always thought YV12 used a luminosity scale of (16,235) -- that was stretched to (0,255) using ConvertToRGB() -- that was then shrunk back down to (16,235) via TMPGEnc 4 Xpress during the encode. Could somebody confirm if this is indeed happening? Or do the levels always stay at (16,235) all throughout the process?
Both before and after clips have identical luminosity, which eases my worrying.. but I'd love to know what goes on behind the scenes to better prepare me for more advanced encoding.
Thanks so much to anybody kind enough to shed some light (pun intended) on the subject! ;)