Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!elroy.jpl.nasa.gov!swrinde!mips!pacbell.com!att!emory!gatech!taco!hobbes.catt.ncsu.edu!kdarling From: kdarling@hobbes.catt.ncsu.edu (Kevin Darling) Newsgroups: comp.sys.amiga.graphics Subject: Re: DCTV Message-ID: <1991Apr21.074135.22843@ncsu.edu> Date: 21 Apr 91 07:41:35 GMT References: <992@celia.UUCP> Sender: news@ncsu.edu (USENET News System) Organization: North Carolina State University Lines: 34 bobl@graphics.rent.com (Bob Lindabury - SysAdm) writes: > neil@celia.UUCP (Neil Richmond) writes: >> I am also concerned that RGB add on device that is coming out soon uses to >> composite signal and converts it to RGB. This sounds like a kludge. Does >> anyone have any more info on this? > > I don't have any specific information on the above problem but what I was > told by one of the reps at AmiExpo in NY a few weeks back was that the > current DCTV was just a cheap consumer version. They plan a complete > rack-mountable professional unit that will address all the concerns we > are having now with the product (ie: lack of RGB, resolution, etc.). > Also they are suppose to be coming out with a true 4:2:2 or 4:4:2 system. > Can someone tell us all what these numbers mean? I know they have something > to do with chroma and luminance channels That would be 4:2:2 YUV. It defines the ratio of sampling of Y (luminance), versus the U:V (color diffs R-Y, B-Y) data. Older systems use 4:1:1. For example, to sample 720 dots/line, you might sample Y at 13.5MHz, while U and V are each sampled at 6.7MHz (4:2:2). Thus the luminance can change for _every_ dot, but the color changes every two dots (altho you can interpolate). Using only 4:1:1 would halve that color change frequency. As there are YUV->RGB encoder chips, the answer to the original question is that they'd most likely tap into the stored YUV data, not the composite signal. PS: To save memory space, YUV can be encoded as Delta values (DYUV). In CD-I, for example, 4:2:2 DYUV is used. Each word defines two pixels: 4-bits delta for each pixel of Y, allowing each pixel to change luminance. The other dual 4-bits are used for the DU and DY for both pixels (interpolated linearly). Because the eye is more sensitive to luminance than color, a DYUV device is preferable to DRGB (Amiga HAM). best - kevin PPS: Pick up a Signetics Video Handbook... has the encode/decode chips.