Yes I know that is the case for versions 5 and below, but if I'm correct (and I may not be), Vegas 6 handles interlaced video more like AVISynth does in that it treats it as 60fps progressive, then drops the extra frames in some kind of a 3-2,2-3 based pattern to make 24p. If that is the case, then supersampling and a bit of motion blur might possibly mimick the effect of a shutter being held open for more than a sixtieth of a second, and the motion blur generated would follow the motion nicely and smooth out things like fast pans. I know people have been doing this trick quite successfully with AVISynth for a while now, but it looks to me like Vegas might be able to as well now. Am I right? Somebody must know!

Sorry I don't have any answers for you. But I am also interested in this general subject, starting with a far more basic question: is the 24p conversion best arrived at by simply rendering a 60i timeline as a 24p NTSC DV file?

Also, is anyone using this sort of conversion for any high-end broadcast work? In other words, is this 24p conversion capacity a bonafide tool to be utilized by professionals? Or is it more of a short-term gimmick until most of us buy a 24p-capable camera?

I wouldn't call it "high end" but the vitamin commercials we're delivering are being delivered at 8Mbps, 24p on DVD.
24p for delivery and 24p for acquisition are two VERY different things. If you want to shoot 24p, you need to learn a lot more about how film cameras shoot/operate, what you can and can't do, and shoot very differently than you shoot for 60i. There are also differences in aperture settings that you'll need to at least understand/experience. Barry Green's book on the DVX100 is great for helping you to understand the concepts.

Shooting in 60i for highspeed motion or moving detail is still the best acquisition we've got. You can shoot that and deliver in 24p just fine, and still get a "film-look." What you lose is motion blur that is only gotten in the acquisition stage. If you're shooting 24p from a static tripod or even on a slow moving dolly, I like the look of 60i resolved to 24p. I'm not a huge fan of shooting 24p, but there are obviously some significant artistic merits to 24p.

Play with what you have, and see if you like the cadence of 24p. You likely will.

Thanks, Spot. That's just the overview I was looking for. As long as the 60i > 24p conversion is considered a legitimate delivery output, then I'll definitely give it a go. Just knowing that you're delivering a few commercials in this fashion tells me all I need to know.

Just to follow-up on my other question, is rendering a 60i timeline as a 24p NTSC DV avi the appropriate workflow for this?

Well one huge benefit of 24p for me is how well it compresses to internet deliverable sizes. I hate little postage stamp sized videos, but going bigger than 320 x 240 usually means having to deal with compressing interlace artifacts. I just rendered a 24p version of my latest project, letterboxed it and rendered it out as a 640 x 352 mpeg1 file with an aspect ratio of "1". At under 200 kbps including audio, it still looks amazingly close to the original 24p DV file. I just can't squeeze a 60i file anywhere near that small and have it look decent.

Slacy, try this one;
http://www.vasst.com/resource.aspx?id=9d37ead7-6d9d-48f1-ade5-6111aba75c96

Spot. In the link you just posted, I found the following two sentences:

"Adding Motion Blur to pans or transitory elements can also make a difference in a shot. Consider using SuperSampling on Motion Blur shots."

This is exactly the stuff I want to know about. As you pointed out, motion blur is the one area where a generated 24p can be considered inferior to native 24p footage. Does adding a little SuperSampling and Motion Blur help bridge this gap? I'm doing a bunch of test renders and it looks like it might.

I don't know that I'd say it looks the same, because it doesn't, but it DOES provide a nice artistic value that if used correctly, can simulate a motion blur that's not terribly different than panning the camera. But....in my experiments, it's tough to get close to "right" and it will never be the same as blurring with a camera and a slow framerate.

There are a few professional products for adding very good quality motion blur in post (based on pixel tracking from frame to frame).

I recently got ReelSmart Motion Blur (works in AE, Combustion, RED and a few others), and it is pretty astonishing.

See RSMB for more info.

Impressive stuff from RSMB, I wasn't aware they had this particular product. One thing I can say with certainty though....they faked a couple of the images in their demo, because I own 2 of the libraries they used for demos, and they played with the footage first. The kids in the last demo are also from the Artbeats library, Kids of Summer, and that footage was shot with 35mm film on tripod.

Spot - in your attached 24p link you mention using the PD170.

In your latest experiments are you seeing a dramatic difference in this "film look" conversion with the Z1 60i vs the PD170 30i to 24p?

Also, you mention the need for a deck that can receive 24p - is this strictly for reconversion to film? I just want the separate reality of the film look - I still want to release on either BetaSP or DVD.

Your thoughts?
v

re: Previously, supersampling only benefited generated media and not video.

This page shows what the main use for supersampling is:
http://ben.orona.com/video/supersample/supersample.htm

The bottom image demonstrates how supersampling has no effect on existing video - when it matches the project frame rate. But in theory, if the media contains additional frames in between project frames, supersampling might pick them up so the motion blur can use them as extra samples, but I'm not sure how or if this works in practice. The help file only mentions track motion, event pan/crop, transitions, or keyframable effects.

Yes you're right. Supersampling is usually used on generated media and has no effect on video at a given frame rate. The reason I brought it up in this thread is because Vegas 6 seems to treat 60i as 60p with some line doubling or interpolation and I believe it achieves it's target 24p by dropping these extra frames. Rather than lose this image data altogether, it seems logical to use it to generate some of the motion blur that would be there if the footage was actually shot 24p to begin with. I've been doing a bunch of test renders with different motion blur amounts and a supersampling value of 3. So far, I would concur with Spot that a tiny bit seems to approach this, but it never seems to get as close to this effect as I had hoped it might.

After looking at a bunch of test renders of 60i to 24p using supersampling and motion blur, I've come to the conclusion that the best you can do is use a full 1/60 second shutter speed for the maximum natural motion blur and leave it at that. Supersampling and motion blur don't really add much, even on fast pans and fast motion.

"they faked a couple of the images in their demo, because I own 2 of the libraries they used for demos, and they played with the footage first. The kids in the last demo are also from the Artbeats library, Kids of Summer, and that footage was shot with 35mm film on tripod."

Hmmm, why do you think they faked the images? I think they used the RSMB Pro features "Tracking Point Guidance" and "Spline Guidance".

Where did you see the Kids of Summer, I didn't see this in the Gallery?

If the kids move in front of a locked camera, you have something to add motion blur to if desired, or incredibly enough with this product, even to remove motion blur from!