Path: utzoo!utgpu!watmath!isishq!f171.n221.z1.FIDONET.ORG!izot From: izot@f171.n221.z1.FIDONET.ORG (Geoffrey Welsh) Newsgroups: comp.sys.cbm Subject: Re: Parallel transfer from IBM to C64 Message-ID: <1447.23E5B7FA@isishq.FIDONET.ORG> Date: 29 Jan 89 06:44:57 GMT Sender: ufgate@isishq.FIDONET.ORG (newsout1.25) Organization: FidoNet node 1:221/171 - Izot's Swamp, Kitchener ON Lines: 41 > From: jgreco@csd4.milw.wisc.edu (Joe Greco) > Message-ID: <601@csd4.milw.wisc.edu> > ]Chris Smeets' drivers are copied almost byte for byte from my FASTERM. > > Plagiarism? He asked so I gave him the code. > ] PaperClip III/64 should be able to XMODEM at 3600 or perhaps even 4800 > bps. > ]Steve's code differs slightly from mine, but I haven't spent the time > looking > ]at it closely. > > Then it might be my 64's. Although neither of the two that are hooked > to 2400's work at speeds higher than that, be it PCIII or FASTERM. Steve's 9600 bps on PC III/128 is 16% off - and I know why. Apparently the 1670 Modem/1200 works fine with the C64's slightly off 1200 bps, but not with a strictly correct 1200 bps! Thus PC II (note: "II", not "III") was designed to provide that error, and it grew at higher baud rates... PC III inherited that timing error (which only got worse on the 1 MHz C64). FASTERM was never tested against a high-speed modem, but DesTerm 128 was verified against a USRobotics HST. We found that, because the modem had to auto sync to a wide range of baud rates, it would not accept even slightly off timing. As a result, the code in DesTerm 128 is almost identical to that in FASTERM, but the constants were arrived at by means of an oscilloscope and confirmed by trial on the HST. Thus, you may not be able to use FASTERM with your 2400 MNP modem at 4800 bps because its autosync timers are too sensitive for it... -- Geoffrey Welsh - via FidoNet node 1:221/162 UUCP: ...!watmath!isishq!171!izot Internet: izot@f171.n221.z1.FIDONET.ORG