Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.2 11/08/85; site unccvax.unccvax.UUCP Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxt!houxm!whuxl!whuxlm!akgua!mcnc!unccvax!dsi From: dsi@unccvax.UUCP Newsgroups: net.analog,net.misc,net.tv Subject: Re: TV & VCR standards Message-ID: <423@unccvax.unccvax.UUCP> Date: Fri, 17-Jan-86 08:06:50 EST Article-I.D.: unccvax.423 Posted: Fri Jan 17 08:06:50 1986 Date-Received: Sun, 19-Jan-86 05:13:08 EST References: <137@daisy.UUCP> <512@nicmad.UUCP> <395@packard.UUCP> <515@nicmad.UUCP> Organization: UNC-Charlotte Lines: 59 Xref: watmath net.analog:628 net.misc:9138 net.tv:3883 > In article <395@packard.UUCP> aat@packard.UUCP (AA Triolo) writes: > > He told me it could be copied digitally for about > >$175 for a two hour tape. He said the quality of the tape > >would suffer however. I should expect colors to wash-out > >and at times could lose verticle hold. > > Boy, is that guy full of shit. A digitally converted copy will actually > > Mr. Video > > Whoooooooaaaaaa, there, Mr. Consumer Video... You just might recall that the EBU broadcast primaries are different from the NTSC broadcast primaries. The two colour gamuts (particularly with respect to gamma precompensation) are not directly compatible, although it is possible to get a pleasing aesthetic result. Colour correction is an extremely expensive proposition which * no one * does for $175/2hr tape. As for "jitters", there are good and bad line + frame interpolation schemes. The first frame store PAL-B to NTSC-M converters suffer from severe motion artifacts, such as the "jerky" tennis ball we've all seen during Wimbledon matches of days gone by. Early "Muppet Show" tapes are also subject to severe temporal jitter. The converter at my spouse's former employer was horrendous in this respect. Yes, I agree that any digital frame store converter (which is not even slightly the same thing as a TBC) should reinsert the sync and burst in precise SC-H relationship for NTSC. However, the ability of the digital frame store converter to follow the original input signal is compromised by a consumer VTR recording. These are typically subject to gyroscopic errors, long time constants in the servo loop controlling headwheel speed and phase (done presumably to minimise "breakup" on cheap TV sets) and very high amounts of subcarrier frequency modulation from line to line due to stiction around the headwheel, etc....The input frame store has to recover a sampling clock from all this crud, and it also has to determine * which field is which * from trashy input synchronising information. Thus, a marginal PAL-B home videotape may actually *appear to roll* because the input converter has lost genlock. The image would still * be fully framed * at the output. I don't think the guy was full of BS at all. When you start with a trashy, timebase/frequency modulated signal that comes from consumer junk, the possibility of successful conversions are fairly low. I might add that professional equipment assumes (in many cases) a priori that there isn't a lot of IFPM on the timebase or colour subcarrier signals, so that the correction loop bandwidth is small where the write clock and timing are recovered. Thus, most professional video equipment has a very hard time genlocking to trash like VHS and Beta. One can only hope that a uniform digital videotape standard will finally filter down to consumers so that they can see what they've been missing. I've given up on the public demanding 4.2 mHz luminance bandwidth for now (but, yessirreee, they'll buy HDTV.) Mr. (Professional) Video David Anthony Director of Engineering DataSpan, Inc.