I've taken quite a bit of analytical sports video at 24P (shutter speed = 1/1000) with my Panasonic DVX100A and slowed it to 30% with velocity with Vegas 5. That's actually pretty smooth, certainly smooth enough for biomechanics analysis.

When I slow this type of video to 10% it is definitely jerky, like amateur 'stop motion'--it's still okay for biomechanics analysis but this video is always taken from a stationary camera on a tripod. I don't know, nor do I want to find out, what that might look like if the camera was panning or not stabilized on a tripod.

At 24P and shutter = 1/1000 most of the frames are sharp and clear (the algorithim does seem to make an interlaced frame after every 4 or 5 "progressive" frames...I never really learned the deep details behind this). But, the problem for me is, of course: This frame rate correponds to about 40 millseconds per frame. At many sports motion speeds, a lot happens in 40 milliseconds. That means that a lot of information is lost between frames: A 90 mph fastball travels 5.3 feet in 40 milliseconds.

Recently, I got hold of a Fastec Imaging SportsCam. This baby does 250 frames per sec, true progressive, and saves the data as uncompressed .avi files. The max resolution is 640 x 480 and I can hold about 5 seconds of video in the internal memory at 250 fps with 640 x 480. At this frame rate, each frame is recorded in 4 milliseconds and there is only 5.3 " of movement between frames for something the speed of a 90 mph heater.

Saving an entire 5 sec clip of data to permanent compact flash card storage is a bitch--that takes ~1 1/2 minutes, so if something new happens while you're saving your last clip...tough luck.

There are some nice features though...if you basically know what you're looking for it is easy to operate the SportsCam in "Record After" mode...i.e., the camera will continually stream video into its buffer until you see something happen that you know you want to save. Then you simply push the 'stop' button and the camera will save the last 5 sec of video to internal memory. There are some nice controls that also let you trim the video on the fly before saving it to a flash card. That's good stuff because the difference in 'save time' for, say, 2 sec of video is 60% less than a full 5 sec clip. This is important for another reason: At the 250 fps rate, 5 secs is 1250 frames of progressive 640 x 480 video and that requires about 300 MB of space on a CF card. In the field, it's much better to trim the video clips to their minimum useful length before saving them.

Slo-mo with any type of interlaced video is always plagued by the potential problems caused by the subject's speed: Each frame of interlaced video is composed of two fields recorded at slightly different times. If that difference is significant with respect to the speed of your subject, individual frames will show interlace artifacts.