Side-by-side: DV & BetacamSP
Home
SW Engineering
Film & Video Production
Video Tidbits
>DV<

This article was originally published 18 October 1995 on rec.video.professional when the Sony DCR-VX1000 first became available. Some 1998-era comments have been added in [brackets]. Copyright (c) 1998 Adam J. Wilt

See also the comparison between DV and Hi8.


Tonight I hooked up the Sony DCR-VX1000 to Betacam SP at Hammar Communications, put a tape in each, and recorded ten minutes. The Y/C out (S-Video) from the VX1000 was sent through a Panasonic UTP-2 format transcoder, and emerged as YUV for input to a BVW-50 via the camera input multipin connector.

For playback, the Betacam came through the UTP-2 again to re-emerge as Y/C, and both it and the VX1000 were fed to a pair of PVM-1344Qs in Y/C mode for side-by-side comparison. Pete Hammar and I watched, exclaimed, and compared notes. However, the comments that follow are mine alone, and do not necessarily represent those of Pete Hammar, Sony, etc. Remember, your mileage may vary, and your perceptions may differ.

The short take: DV and Betacam SP were of comparable quality. Both showed highly transparent, very watchable pictures. Both showed annoying artifacts to the same degree, although different in kind; it was a toss-up as to which format rendered reality "better."

Neither DV nor Betacam showed noticeable noise beyond that introduced by the VX1000 camera head. Both were superb at rendering luminance detail, large areas of saturated solid color, and small color details. In most ways, pictures were visually identical. However, each format had its own failings.

DV does show "mosquito noise" (fine DCT "noise" at points of sharply contrasting details, especially around diagonal lines and corners) at a very low level (we had to work at picking images to generate noticeable mosquito noise), and I was able to see some "quilting" artifacts (a large-scale "fixed pattern" distortion due to differing quantization characteristics between different DCT blocks) when doing slow, gentle pans of ceiling-mounted fluorescent lights (when an edge of the light was tilted about 15 degrees from the horizontal plane of the image, quilting artifacts could be noticed by close observation of the boundary between the light and the surrounding acoustic tiles; a slow pan revealed fixed boundaries between blocks where the edge "aliased" across the scanlines slightly differently). Both these artifacts were not normally apparent, but showed up a) because we were looking for them, and b) we could see the difference with Betacam images in an A/B comparison.

Betacam showed a sort of dither or instability on fine horizontal detail (that is, images of finely spaced vertical lines), which I attribute to random residual timebase errors on a scanline-by-scanline basis. While we are all used to seeing this, the DVC images were by contrast free of any such "liveliness" of vertical edges; the DV images were perfectly quiescent, looking more like E-E camera images. Again, this "liveliness" was not normally apparent, but when we zoomed out slowly from a shirt with fine, closely-spaced vertical pinstripes, the DV image was clean and stable until the pinstripes blurred together, whereas the Betacam image became distractingly "wiggly" and disquietingly "alive."

In neither tape format were the recording artifacts present and visible to such a degree as to be distracting, or even noticeable upon casual viewing. We knew what to look for and where to look for it, so we saw it, but neither of us (as I recall - correct me, Pete, if I misrepresent your views) found either the DV digital artifacts nor the Betacam SP analog artifacts to be impediments to using either format; they appear to be of comparable transparency and realism. Pete said, upon viewing the images and seeing the DV cassette, that "DV is the DAT of video."

Now, if/when you can get a real professional camera head to feed a DV (or DVCPRO) recorder [which you can now do]... and when we have editing capability [available in DV, DVCAM, and DVCPRO]... and if the DVCPRO editors really will accept consumer mini-DV cassettes for source material, allowing use of the consumer camcorders for low-cost acquisition (for those of us who can live with the fiddly little controls, the five-speed-only zoom, and the funky color finder) [you can]... a new day is dawning; I'd say that DV is as significant a milestone as the introduction of Betacam or the first D1 machine (but that's just my opinion, mind you!).

Copyright (c) 1998 by Adam J. Wilt.
You are granted a nonexclusive right to duplicate, print, link to, frame, or otherwise repurpose this material,
as long as all authorship, ownership and copyright information is preserved, and a link to this site is displayed.


Home SW Engineering Film & Video Production Video Tidbits >DV<

Contact me via email

Last updated 30 March 1998.