-
XDCAM EX footage 1080p30 equivalent to 1080i60?
Okay, I’m new at HD video editing but I am just starting to experiment with the various formats of Sony XDCAM EX, which is the format of my JVC GY-HM100 camera. I can easily import these files into Adobe Premiere, and editing with them seems to be very smooth. What I cannot understand, however, is why the 1080p30 and 1080i60 formats seem to be identical, as far as Premiere Pro is concerned.
Let me explain: I create two different sequences in Premiere Pro (CS4) — one sequence is in the 1080i 60i format; the other uses 1080p 30p. I will then import footage recorded in both these formats, and when dropped in their respective timelines, everything goes well (no rendering is required, as you would expect).
However, when I drop the 1080p 30p clip into the 1080i 60i sequence, still NO RENDERING is required, despite not being in its native timeline format. Why is this??? I’d have thought that rendering would be required due to the discrepancy between the clip and the timeline settings, but clearly not.
-
4 Replies
-
I suspect it’s because the EX cams (I use the Sonys myself…) run on PsF…Progressive Segmented Frame.
In interlaced video, (60i or 50i) you have “fields” that each represent an instance in time. If you have a shot of someone running across the frame and you freeze a frame…and the freeze actually includes both fields (many VTRs used to drop to one field when in shuttle/pause), the frame will probably show some “sawtoothed” edges as you are seeing the runner in two positions…a 60th of a second apart, interlaced into one frame…which is usually noted as half the field rate-two fields=one frame.
In progressive video, obviously an entire frame is composed of one instant in time. No interlacing.
However, many systems (and CineAlta has traditionally been one of these) will still structure the image data as interlaced for reasons of technical ease in various hardware configurations outside of a computer. In this case, the progressive frame is segmented into two fields, even though each field represents the same instant in time. The progressive footage is reconstructed technically in the same way that any interlaced footage is…PPro, or an HDcam deck or an SDI pipeline doesn’t know the difference, or care.
I suspect that the “engine” in PPro sees data structured in the same way whether you have 29.97p or 29.97i (59.94i for those who like to arbitrarily mix up their nomenclature by trying to quote fields as one value and frames as another…).
If the frame rates are not synchronous…say you mixed in some 25i…I’m sure the red line would make an appearance.
Your 29.97p footage will still -look- substantially different than your 29.97i in any case, which is of course, why most videographers use progressive.
Hopefully that’s helpful?
TimK,
Director, Consultant
Kolb Productions, -
Thanks Tim, yes it’s quite helpful and explains a lot. And it also makes sense that although PPro doesn’t care if you place 29.97p footage into a 29.97i timeline, it does “care” if you do the reverse — i.e. placing 29.97i footage into a 29.97p timeline does require rendering.
I have heard that it is better to do chroma keying using progressive footage, but with what you’ve just said, maybe it makes little difference whether one uses 29.97p or 29.97i?
-
Well, progressive footage can be easier to work with, yes.
If given a choice, I would typically choose progressive for an effects job.
Those sawtoothed edges can be a pain in a keying/compositing environment.
TimK,
Director, Consultant
Kolb Productions, -
A little Canon HF200 seems to say that its 1080p30 is captured as 1080i60 anyway. Certainly the images look good in a EX(HQ) 1080p project. The avchd files are going thru avisynth because CS3 does not support avchd, and everything needs rendering.
George Socka
BeachDigital
http://www.beachdigital.com
Reply to this Discussion! Login or Sign Up