Why does 25f HD look better than 50i when downconverted to SD?

[Geoffrey Cox]

As a recent purchaser of a (PAL) XH-A1 I've found the learning curve steep but am getting there thanks to this amazing forum! I shoot in HD but am in the middle of an SD project so have to downconvert all footage. I think I've determined beyond reasonable doubt (in my own mind) that:

a) Footage definitely looks better if shot in HD then down converted to SD rather than shot in SD. I do wonder if this has something to do with what I consider the rather problematic SD mode - I've played endlessly with the custom presets to get a good SD image but always come up against an unsatisfactory trade off between fuzziness and noise; the coring parameter seems key to getting clarity (at -9) but adds a lot of noise. Anyone have good custom settings for the best SD image as sometimes shooting in SD is needed?

b) Downconverting in-camera is worse quality than post conversion - I use Quicktime and found, despite numerous tests, that DV compression was just as good as anything else; does anyone have any thoughts on best compression?

c) (And this baffles me) Footage shot in 25p looks sharper when downconverted than 50i and I don't think this is a de-interlacing problem as both frame rates look equally good in HD. Any ideas why this should be (a very recent thread seemed to offer a partial explanation but not for downconverting)?

Again many thanks to all posters for the invaluable knowledge and help this forum provides!

Geoff

[Jo Potts]

Originally Posted by Geoffrey Cox

As a recent purchaser of a (PAL) XH-A1 I've found the learning curve steep but am getting there thanks to this amazing forum! I shoot in HD but am in the middle of an SD project so have to downconvert all footage. I think I've determined beyond reasonable doubt (in my own mind) that:

a) Footage definitely looks better if shot in HD then down converted to SD rather than shot in SD. I do wonder if this has something to do with what I consider the rather problematic SD mode - I've played endlessly with the custom presets to get a good SD image but always come up against an unsatisfactory trade off between fuzziness and noise; the coring parameter seems key to getting clarity (at -9) but adds a lot of noise. Anyone have good custom settings for the best SD image as sometimes shooting in SD is needed?

b) Downconverting in-camera is worse quality than post conversion - I use Quicktime and found, despite numerous tests, that DV compression was just as good as anything else; does anyone have any thoughts on best compression?

c) (And this baffles me) Footage shot in 25p looks sharper when downconverted than 50i and I don't think this is a de-interlacing problem as both frame rates look equally good in HD. Any ideas why this should be (a very recent thread seemed to offer a partial explanation but not for downconverting)?

Again many thanks to all posters for the invaluable knowledge and help this forum provides!

Geoff

Hi Geoff and welcome to the forum! Not sure about your progressive vs interlaced issues, but it's always best to shoot in HD then downconvert (in post) to SD if needed - this way you have a higher res version for future proofing etc.

I'd covert in post that way it's on your computer in HD if needed - the only issue being if you've not got enough horsepower to work with HDV vs DV (HDV uses long-GOP compression, which means it compresses the frames in a 15 frame group and thus has to decompress more info at once and therefore demands more processing power over DV, which compresses each frame internally...well that's the basic and probably not wholly accurate explanation!)

You'd have no issue with storage over DV as they are exactly the same data rate at 25MBps.

Basically i wouldn't bother with custom presets for SD, just shoot in HD with the custom presets that exist and then down res in post (I personally edit all in HDV, then output using compressor to SD if needed).

Hope that helps somewhat...sorry for the ramblings!

[Annie Haycock]

If you are mixing existing SD with new HD, should the down-converting still be done in post? As I haven't had much time for playing around with this, I just took the most obvious way out and downconverted in camera - the results still looked better than the old SD footage.

[Geoffrey Cox]

Annie - though I have very limited experience, the in-camera conversion process was definitely less sharp than doing it in post in all tests I did, though the difference was not that great and both looked better than shooting in SD.

Jo - thanks for the advice. I'd love to do more editing work in HD but unfortunately I've only got Final Cut 4.5 which doesn't recognise the HDV codec so have to convert first. I've thought of moving up to FCS but apart from the financial side, I've only just learned FCP and was worried about FCS being different enough to affect the working process of the project I'm in the middle of.

Geoff