Is the XL H1 4:2:2

macgregor

macgregor

Veteran
Hello.
Since the H1 set at 24/25f duplicates/interpolates one field to create a progressive image, could we say it is 4.2.2 color sampling since both fields come from the same one?

Is this correct? 1440x540 at 4.2.2?
 
Hard to say. If nothing else, you could always shoot 50i and convert to 25p in post. That way you're sure to have 4.2.2 to start with.

I have noticed a slight chroma increase when capturing through SDI for greenscreen, even from 24F. It's slightly easier to pull keys from SDI than from tape originals, but tape is still workable.

Scott Billups has given a big thumbs up to the H1 when it comes to SDI acquisition for greenscreen. He keyed 24F stuff, btw.
 
The Canon rep at the Birns & Sawyer seminar stated that the HD-SDI output from the Canon is full raster 1920x1080 4:2:2. Now that's unsqueezed from the CCD's 1440x1080 chip.
 
Yes but we are talking about the HDV recording. HDV is 4.2.0 but since one of every two fields come from the same field, we should have the same information, so ...
 
I don't think that's correct. In all likelihood, the reason 24F looks so good is that the vertical resolution is coming from two woven separate fields to make a progressive looking frame in a smart deinterlace process. It's probably something like 48i stitched back together to be a 24p recording. And it IS a 24p recording to tape!! They've gotta be doing something right otherwise people would dismiss the mode as too low rez, like the Cineframe stuff from the Z1U.

This is still just reasonably informed speculation, at best. I've learned to stop worrying and just enjoy the results. Whatever it is works for me.
 
Last edited:
Hi Macgregor,
I would assume that the field interpolation happens in RGB before the video encoding. The image from the chip gets the gamma and other video processing before being sent out to HDV or SDI in YUV.
There is probably a way to test this with some kind of chroma test chart. It would be interesting to print out some color circles and shoot it to HDV tape and digitize it through a SDI card to a computer for comparison. I would bet that the diagonal color resolution is better through SDI.
There is an excellent discussion about color sampling on the DVD player benchmark page. I think that if you read this you could design a test that would satisfy your curiosity about the camera.
http://www.hometheaterhifi.com/volume_8_2/dvd-benchmark-special-report-chroma-bug-4-2001.html
Have fun!
Sam
 
Bottom line, Scott Billups (The Man!), is quoted as saying that the H1 keys greenscreen footage BETTER than the F900 and that the filmout tests looked spectacular compared to the F900.

That's good enough for me.
 
Too bad those Star Wars fan geeks didn't have access to this camera in the middle of all that prequel hoopla. They might've made better movies!
 
You must log in or register to reply here.
Back
Top