ferwerda@batcomputer.UUCP (05/30/87)
Keywords:NTSC, RGB, Video We are currently trying to put together a 3/4" video animation/editing system. One of the first big stumbling blocks in doing such a thing is figuring out a way of converting our 1280 X 1024 X 24 bit RGB images into some NTSC compatible format. Our local "professional" video supplier has never dealt with this problem. Is there anyone out there who has? Does anyone know the names/prices of NTSC converters that can do this? The specs of our monitors follow. Model: Conrac 7311 Sony GDM-1901-12 Addressability: 1280x1024x24 1280x1024x24 Vertical Scan: 60 Hz noninterlaced 60 Hz noninterlaced Horiz. Scan: 65Khz +/- 2.5 Khz 63.4 Khz Input: RGB RGB Sync: Internal (on Green) Internal (on Green)
mike@wsucshp.UUCP (06/02/87)
Give the folks at Lyon Lamb a call. They make encoders and VTR controllers. (818) 843-4831 ---- Mike ( Those who can do, do, those who can't simulate! ) - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Michael K. Kibler CSNET: mike%wsu@RELAY.CS.NET Computer Science Dept. UUCP: ..!ucbvax!ucdavis!egg-id!ui3!wsucshp!mike Washington State Univ. BITNET: kibler@wsuvm1.BITNET Pullman, WA. 99164-1210 PHONE: 509-335-2723 or 509-335-6636
thomas@spline.UUCP (06/02/87)
In article <1208@batcomputer.tn.cornell.edu> jaf@squid.tn.cornell.edu.UUCP (James Ferwerda) writes: > >a way of converting our 1280 X 1024 X 24 bit RGB images into some NTSC >compatible format. Cain't be done. NTSC only applies to "525" line systems. You need a frame buffer that has approximately 480 lines in the vertical direction, and a 30Hz interlaced scan (i.e. 60Hz field rate). It should produce "RS-170" compatible video signals. Then you can get a fairly simple circuit to do the conversion from RGB to NTSC. I don't know of suppliers off the bat, but I imagine that if you come back to your video supplier with this problem, he will be able to handle it. There are other issues: NTSC doesn't have the color bandwidth that an RGB display can put out, so you need to be careful about color changes in the image. One reference on this is "A Low Cost, Video Based, Animated Movie System for the Display of Time Dependent Modeling Results", by William E. Johnston, Dennis E. Hall, Fritz Renema, and David Robertson, at Lawrence Berkeley Labs. It was published in the proceedings of the Third Computer Graphics Workshop by USENIX Association (P.O. Box 7, El Cerrito, CA 94530 USA). It contains lots of useful information about NTSC. =Spencer ({ihnp4,decvax}!utah-cs!thomas, thomas@cs.utah.edu)
hutch@sdcsvax.UUCP (06/03/87)
Thus sayeth Spencer Thomas in article <2049@utah-gr.UUCP> (thomas@spline): >So spake (James Ferwerda) verily <1208@batcomputer.tn.cornell.edu> (jaf@squid.tn.cornell.edu): >>a way of converting our 1280 X 1024 X 24 bit RGB images into some NTSC >>compatible format. >Cain't be done. NTSC only applies to "525" line systems. You need a >frame buffer that has approximately 480 lines in the vertical >direction, and a 30Hz interlaced scan (i.e. 60Hz field rate). Well, yes & no. Lyon Lamb supposedly sells a nice little widget that will convert gold (high res RGB) into lead (NTSC). Basicly it is a frame buffer and some signal conversion gear. The 1280 is just pixel rate conversion, the 1024 is black magic to me (deep analog magic, or dropping lines?). Presuming that you know you will be losing colors and pixels in the conversion, I can't see why it is impossible. I am not berating Spencer, perhaps I have missed his meaning. Important note, you get about 424 color changes in the horizontal direction. There is also a limitation on how close together you can put color changes. It is all nasty analog biz, look at the equations/definitions and calculate how far you can push it before the chroma starts sneaking into the luminance. The number of changes is a function of how much the signal knows and *guesses* about the contents of a scanline. Yup, NTSC guesses. More nasty biz, when it is sliding through a sync pulse it is also continuing with the last thing it knew to display (color & luminance-wise). Well, along with Spencer's suggested reading, you might also look at the NTSC spec from EIA, the SMPTE journals are another suggested source. I'm no master, just saying a piece in the name of the revolution. - -- Jim Hutchison UUCP: {dcdwest,ucbvax}!sdcsvax!hutch ARPA: Hutch@sdcsvax.ucsd.edu Disklame'r: One greater than the greatest signature representable with 184 symbols.
piety@hplabsb.UUCP (06/04/87)
In article <2049@utah-gr.UUCP>, thomas%spline.uucp@utah-gr.UUCP (Spencer W. Thomas) writes: > In article <1208@batcomputer.tn.cornell.edu> jaf@squid.tn.cornell.edu.UUCP (James Ferwerda) writes: > > > >a way of converting our 1280 X 1024 X 24 bit RGB images into some NTSC > >compatible format. > > Cain't be done. NTSC only applies to "525" line systems. You need a ....easily, that is. You CAN use a scan converter (there are at least 4 suppliers) that takes in the 1280x1024x24 image and converts it to NTSC. They're not cheap, though... Bob
ccap001@ut-ngp.UUCP (06/05/87)
Lyon Lamb makes what they call the HRC (high resolution converter) it
accepts as input RS-343 signals and output RS-170 and NTSC composite
signals. I have never seen one in action so I can't give you any comments
about the quality or usability of the thing. I did speek with them about
it and they seemed very willing to discuss any thing you need to know about
it.
For what it's worth, There is a company out of Sunnyvale, California
called Faroudja Laboratories that is building an RGB converter for RS-170
type signals. It is definately the top of the line type box. If you are
interested in more info you can get in touch with them at:
Faroudja Laboratories Inc.
946 Benicia Ave.
Sunnyvale, California 94086
(408) 245-1492