nikolai
11th September 2008, 01:52
Using firewire I captured an HDTV 1080i transport stream and want to convert it to an NTSC anamorphic DVD. After many false starts and a lot of reading, I stumbled on this thread (http://forum.doom9.org/showthread.php?p=701006) from 3 years where scharfis_brain (:thanks:) suggested this script:
mpeg2source("hd.d2v")
crop(0,0,0,-8)
TFM("hd.d2v")
TDecimate()
converttoyuy2()
lanczosresize(704,480)
to convert to 23.976 fps progressive. In an amazing coincidence, the show in question in that thread is the same one I am trying to convert. Too weird. Anyway, I tweaked scharfis_brain's script to this:
LoadPlugin("E:\Video\DGIndex\DGDecode.dll")
MPEG2Source("G:\sclip\v01.d2v")
AssumeTFF()
Crop(0,0,0,-8)
TFM(d2v="G:\sclip\v01.d2v")
TDecimate()
LanczosResize(704,480)
I encode with this script and HCEnc, run DGPulldown on the result and let DVDLabPro re-mux the video with the AC3 audio. The end-result DVD looks really good. As with the previous seasons of this show I'll eventually buy the DVD box set of season 4 whenever it's finally released, so I'm not necessarily looking for perfection, just a way to tide me over until then.
I have a couple of questions about the scripts. Am I correct that the ConvertToYUY2() is no longer necessary? During my experiments I didn't use this before LanczosResize() and not using it in my script here doesn't seem to cause any problems. If I read correctly, the reason would be that LanczosResize() was updated since 2005 to support non-YUY2 colorspaces.
My other questions is about LanczosResize(704,480). Originally I was using those values based on something I had read a long time ago. But in the last 2 weeks I only saw samples that used 720x480. When I found scharfis_brain's script I got confused again. I'm pretty sure it has to do with overscan but I'm not certain. 720 would preserve more detail coming down from 1080, but if it's lost in the overscan there's probably not a lot of benefit to it. Which should I use, or why might I choose one over the other?
Many thanks for any advice and a big Thank You to scharfis_brain. I was stuck on older filters for the IVTC steps.
mpeg2source("hd.d2v")
crop(0,0,0,-8)
TFM("hd.d2v")
TDecimate()
converttoyuy2()
lanczosresize(704,480)
to convert to 23.976 fps progressive. In an amazing coincidence, the show in question in that thread is the same one I am trying to convert. Too weird. Anyway, I tweaked scharfis_brain's script to this:
LoadPlugin("E:\Video\DGIndex\DGDecode.dll")
MPEG2Source("G:\sclip\v01.d2v")
AssumeTFF()
Crop(0,0,0,-8)
TFM(d2v="G:\sclip\v01.d2v")
TDecimate()
LanczosResize(704,480)
I encode with this script and HCEnc, run DGPulldown on the result and let DVDLabPro re-mux the video with the AC3 audio. The end-result DVD looks really good. As with the previous seasons of this show I'll eventually buy the DVD box set of season 4 whenever it's finally released, so I'm not necessarily looking for perfection, just a way to tide me over until then.
I have a couple of questions about the scripts. Am I correct that the ConvertToYUY2() is no longer necessary? During my experiments I didn't use this before LanczosResize() and not using it in my script here doesn't seem to cause any problems. If I read correctly, the reason would be that LanczosResize() was updated since 2005 to support non-YUY2 colorspaces.
My other questions is about LanczosResize(704,480). Originally I was using those values based on something I had read a long time ago. But in the last 2 weeks I only saw samples that used 720x480. When I found scharfis_brain's script I got confused again. I'm pretty sure it has to do with overscan but I'm not certain. 720 would preserve more detail coming down from 1080, but if it's lost in the overscan there's probably not a lot of benefit to it. Which should I use, or why might I choose one over the other?
Many thanks for any advice and a big Thank You to scharfis_brain. I was stuck on older filters for the IVTC steps.