Back in 1981 I was working at the old RCA Sarnoff Labs in Princeton NJ, bsck when RCA was a real company that owned itself. We were kicking around the possibility of teletext reception in the USA. The teletext "carrier" frequency was to be 2.86MHz, or more precisely 4/5 the chroma subcarrier frequency. We did field tests of teletext reception, and pretty much found that everyone with a rooftop antenna could expect to get it, but forget it with rabbit ears. One task I had to do was create a psudeo-random teletext pattern generator to feed a French made Bit Error Rate teletext meter. One big problem was that all the documentation was written in French. I eventually figured out how to do it, by visiting another lab at WCBS-TV where they had a generator, and I looked with a scope at the preambles and so on. We already had a homebrew teletext generator, so all I had to do was add an extra circuit board with parallel to serial chips (for the preamble bit pattern) and a bunch of logic to create the psudeo-random pattern. A bunch of shifts and XOR gates, and lots of glue counters to tell it when to start and stop. Today you'd program an FPGA chip.
I once suggested that a teletext equipped TV set could make a great computer monitor (this was back in the old 8 bit days of Commodore 64 computers and Atari 2600 game machines, that used TV sets on channel 3 and you had fuzzy characters on the screen). Just have the computer "fake" a TV channel with teletext signals encoded on it. I don't think anyone ever actually did that.
|