pete@octopus.UUCP (Pete Holzmann) (08/05/88)
In article <791@riddle.UUCP> domo@riddle.UUCP (Dominic Dunlop) writes: >Byte, June 1988, carries a comprehensive review of thirteen high-speed >modems which can operate over the public telephone network. And, yes, the >true V.32 products (as opposed to cut-down V.32, the older V.29, or >proprietary) came out with the highest throughput in a random data test.... >[V.32 did nicely on clean lines, poorly on marginal ones] >By comparison, the much-discussed Telebit >Trailblazer Plus did 5568 and 5078 bps respectively in each test. If only they had used a more realistic marginal connection! Or had been willing to test at faster than 9600 baud. Their numbers are obviously bogus, since just about anybody can get more than 10000 bps out of a Telebit on a clean line, without compression. They got 5568 on a clean line? That's worse than the worst real-life Telebit connection I've ever heard of! >The >joker in the pack was the Telcor Accelerator 2496MA which, by using very >smart proprietary compression on top of old V.22 bis 2400 baud technology, >achieved 8256 and 8362 bps. This makes me *very* suspicious about their 'random' data. I wonder how well the Telcor would do on a compressed (ARC/etc) file? Anybody have experience? Anybody ever heard of a fast compression algorithm that gets 4x compression out of random binary data? Amazing! Pete -- OOO __| ___ Peter Holzmann, Octopus Enterprises OOOOOOO___/ _______ USPS: 19611 La Mar Court, Cupertino, CA 95014 OOOOO \___/ UUCP: {hpda,pyramid}!octopus!pete ___| \_____ Phone: 408/996-7746