-
what colour space do i use for lagarith intermediate compression?
Hello technical wizards
My camera, jvc gz-hd40, records in mpeg2-ts using a 4:2:0 chroma sampling which i believe is YCbCr (correct me if i’m wrong) now i deinterlace my footage in virtualdubmod and then compress it with lagarith to be used in sony vegas, after effects, premiere pro etc. but what i am wondering is what colourspace do I select in lagarith (rgb (default), RGBa, YUV2 or YV12) so as to retain the original quality as much as possible?
ps: the camera can also record in x.v.colour which is the xvYCC colourspace, but as far i know, few editing programs or codecs support this
Cheers
-
3 Replies
-
YV12 is 4:2:0.
I think it depends on what your output format will be as to what colorspace. If you’re in YUV2 (which is 4:2:2), you’ve given the data more room – if you’re in YV12, you’ve given it identical room, as long as nothing more was done to it.
RGB goes through a transform…
So, just off the top of my head, I’d suggest YU2.
Best,
Jeff G
Apple Master Trainer | Avid Cert. Instructor DS/MC | Adobe Cert. Instructor
————
You should follow me (filmgeek) on twitter. I promise to be nice.
New- my book (with Richard Harrington and Robbie Carman)- An Editor’s Guide to Adobe Premiere Pro
Compressor Essentials from Lynda.com
(older but still good) Marquee, Media Composer (3.5) and Basic/Advanced Color DVDs (1.0) from Vasst.com
Contact me through my Website -
thanks for the quick reply
am i right to say that YV12 will be the most similar to my camera’s footage and therefore the best for an intermediate codec? after i have edited i usually compress to AVC/H.264 for internet or burn direct to DVD. what do you mean when you say “if nothing else was done to it”? i am just converting to lagarith between camera and NLE but within the NLE i will be doing a lot to it: colour correcting etc is this what you mean?
-
[Adam Prest] “am i right to say that YV12 will be the most similar to my camera’s footage and therefore the best for an intermediate codec?”
Maybe. Maybe not. You’re in the same color space – but if you do any transforms in Vegas, probably being 4:2:2 means less further damage. 4:2:0 is fairly compromised at the start. (PS I had to research YV12 – let’s call it what everyone else does 420). I’d probably go to 422 space to give it breathing room.
If you’re working in an intermediary codec, the big issue is how much extra bandwidth did you give it? (for HD 100+mbs is decent) – after that, it’s irrelevant where it goes – it was compromised to begin with, you put it in an intermediary codec.
Do your own comparisons (and report back here for others!)
[Adam Prest] “what do you mean when you say “if nothing else was done to it”? “
I mean this; if you just convert it (into a higher bitrate) you’ve done minimal to it. Each transformation does more damage – try to do them all at once; I don’t know about how Vegas handles it’s CC – is it YUV? Is it RGB?Best,
Jeff G
Apple Master Trainer | Avid Cert. Instructor DS/MC | Adobe Cert. Instructor
————
You should follow me (filmgeek) on twitter. I promise to be nice.
New- my book (with Richard Harrington and Robbie Carman)- An Editor’s Guide to Adobe Premiere Pro
Compressor Essentials from Lynda.com
(older but still good) Marquee, Media Composer (3.5) and Basic/Advanced Color DVDs (1.0) from Vasst.com
Contact me through my Website
Reply to this Discussion! Login or Sign Up