Newsgroups: comp.sys.apple2.comm Path: news.weeg.uiowa.edu!news.uiowa.edu!hobbes.physics.uiowa.edu!math.ohio-state.edu!howland.reston.ans.net!pipex!uunet!comp.vuw.ac.nz!actrix.gen.nz!dempson From: dempson@swell.actrix.gen.nz (David Empson) Subject: Re: LineLink 14.4 Organization: Actrix Information Exchange Date: Thu, 13 Jan 1994 14:48:38 GMT Message-ID: References: <2grbig$lvq@usenet.INS.CWRU.Edu> Sender: dempson@actrix.gen.nz (David Empson) Lines: 84 In article sbdocker@acs.ucalgary.ca (Sean Dockery) writes: > Let's dispell a few rumours using first hand experience, shall we? > > > I have an original Apple IIc with a serial number beginning with D4... I also have an early-model IIc (D47001J), which belongs to my user group (signed personally by Steve Wozniak, and donated to the user group). I haven't encountered problems at any speed up to 9600 bps, when the IIc is connected to another computer (usually a IIgs) via a direct serial cable. I haven't tried using a modem faster than 2400 bps on it. I do know people who have had trouble, though, and have required the motherboard upgrade to fix it. It may depend on the exact device you have connected to the computer. > Any Apple IIc is fully capable of DSR/DTR handshaking with the proper > cable and modem port driver. One of the major stumbling blocks of > this is the improper wiring of the modem port's 6551 ACIA chip to the > hardware pins. It is possible to program around this (and it has been > done several times already). DSR/DTR handshaking? You mean dropping DTR to make the modem hang up? The only point of confusion is the IIc's output flow control line actually comes from the RTS signal on the 6551, not the DTR signal. There is a problem with CTS/RTS handshaking (limiting rate of data flow, used with high-speed modems, data compression, or if the computer to modem speed doesn't match the line speed). The 6551 has several design flaws. The RTS, DTR, CTS and DCD signals all interfere with transmitting and/or receiving data (only the DSR line has no effect on the transmitter or receiver, but the IIc uses the DSR signal for "external interrupt" on port 1, and "keyboard interrupt" on port 2). DTR: deactivating DTR will prevent any further characters from being transmitted or received. I think that a character in transmission will be cut short. RTS: deactivating RTS will prevent any further characters from being transmitted. I think that a character in transmission will be cut short. CTS: deactivating CTS (input) will cut off a character being transmitted and will prevent further transmission. DCD: deactivating DCD (input) will prevent the current and subsequent characters from being received. The IIc's flow control lines are connected to the DCD (input) and RTS (output) pins. The DCD input is the big problem: if the modem dropped its flow control, the computer would not be able to receive any more characters, including any character which was already on its way. The RTS problem can be avoided by delaying its turn-off until after the current transmit character has been completely sent. This is why a IIc cannot handle CTS/RTS handshaking properly. You are likely to lose receive characters. The 65C51 (CMOS version) has some design improvements, including CTS not chopping off characters. I suspect RTS doesn't chop characters either (but DTR probably does). On a IIe with a Super Serial Card, you have the option of using the CTS, DSR or DCD lines for input handshaking. DSR has no effect on transmitted or received data, so it could be used for RTS/CTS handshaking, in conjunction with RTS. CTS only affects the transmitter (which is the correct behaviour for CTS/RTS handshaking), but cuts off a character in transmission (unless you replace the 6551 with a 65C51). Ideally, you should have a 65C51 on a Super Serial Card, and be using the CTS and RTS pins for handshaking. This allows you to use DTR to drop the connection (or other functions like entering command mode), and the DSR signal can be used to detect connections (my BBS had the DCD line from the modem connected to DSR, since using the DCD input would have prevented characters from being received while the modem wasn't connected). -- David Empson dempson@swell.actrix.gen.nz Snail mail: P.O. Box 27-103, Wellington, New Zealand