home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.dcom.modems
- Path: sparky!uunet!rde!ksmith!keith
- From: keith@ksmith.uucp (Keith Smith)
- Subject: Re: Serial and Parallel interface ??????
- Organization: Keith's Computer, Hope Mills, NC
- Date: Mon, 27 Jul 92 23:24:32 GMT
- Message-ID: <1992Jul27.232432.5603@ksmith.uucp>
- References: <1992Jul21.205557.29123@cc.ic.ac.uk> <1992Jul23.011759.14499@ksmith.uucp> <1992Jul24.184318.18883@cc.ic.ac.uk>
- Lines: 97
-
- In article <1992Jul24.184318.18883@cc.ic.ac.uk> cmaae47@cc.ic.ac.uk writes:
- >Intuitively I always knew that the 80xxx series of processors was a bad
- >product, now you give me another nail for the coffin. I can still remeber
-
- Hehe, The Z-80 Had a "flip-set" of registers for interrupt handling.
- The instruction to flip-flop them took like 3 cycles. My favorite CPU.
-
- >On the mint condition 1981 PC with an 8088 this may mean doing either a
- >(floppy-) disk or a serial port transfer, and in particular telling the modem
- ^^^^^^^^^^^^^^^^^
- >to shut up for some time. This does not help nominal throughput, it cannot do,
- ^^^^^^^^^^^^^^^^^^^^^^^^
-
- Uh ... Okay, firing down CTS, or sending XOFF a buffer full (which is
- even MORE instructions BTW) may stop your modem from sending chars to
- the CPU, but what is gonna prevent the REMOTE site from continuing.
- Assume a dumb modem at both ends?
-
- >as the CPU on the receiving end is the bottleneck.
- >
- >However, not obeying flow control will be more expensive, because files or
- >packets have to be retransmitted. We did exactly that here to get early
- >Kermits in binary form: read from the serial port, write to disk, if bytes
- >are lost ask for retransmission. Wasteful, but its get you going.
-
- Exactly, So Now your File Transfer Protocall is handling flow control
- by not ACK'ing packets until the writes have completed.
-
- >
- >But the point is that NO manufacturer I know of has sold machines that
- >dropped bytes in transfers to and from floppy disks, I someone did they
- >certainly did not sell many of those. Serial (or parallel) ports should
-
- Again, The dropping of characters cannot be guaranteed by the receiving
- hardware, even with flow control. In fact, YOU JUST SAID THAT U USED
- ONE that required packet re-transmission. That means the hardware
- dropped the ball during the floppy write. Period. It was either
- dropped by the SIO chip because the CPU didn't read it fast enough, or
- it was dropped by a buffer overrun in the modem. Otherwise there would
- have been no neccesity to re-transmit the packets...
-
- >have the same reliablility, and manufacturers should give people *exact*
- >ideas what their machines can and cannot do. There is only one way to
- >achieve this: keep pestering them about it.
-
- Again dropping chars has nothing to do with the hardware manufacturers
- per se. A properly designed File Transfer Protocall will not drop
- characters provided no real hardware limits are exceeded. This requires
- cooperation between BOTH machines.
-
- i.e. A bit of well written assembler to do Zmodem will allow the 8250 on
- the XT to run at 115K baud. Your thruput however will not be quite that
- good.
-
- >
- >Finally, a parallel port has got some big advantages. At low speeds, you can
- >control a LED bar or a robot, at high speeds there is the fact that each byte
- >is handed off and acknowledged. This is also done in (asynchronous) SCSI.
- >
-
- The problem with Parallel OR Serial is the lack of brains on both ends
- of the interface, meaning the main CPU has got to handle ALL the
- overhead. Drop a CPU on the other side (as the case with async SCSI or
- a parallel interface doing an ACK on each byte) and the WAY it is
- transmitted becomes sort of irrelevant as long as both are on the same
- sheet of music.
-
- >The problem with SCSI is that there is a fairly rigid protocol, and that
- >people who have SCSI buses on their computers usually have disks drives
- >hanging off them. They are therefore reluctant to put anything onto a SCSI
-
- The problem with SCSI is that it is a Peer to Peer protocall, and
- requires a computer at both (all) addresses to handle the protocall. In
- other words mo' money. A "standard" parallel printer (Centronics Style)
- protocall is a hell of a lot more rigid than SCSI. In fact, it doesn't
- support bi-directional data transfer as spec'd.
-
- >bus before it is known to behave correctly. Once it is understood what a
- >peripheral does, the software has been debugged and the throughput
- >requirements are known it can well go on to the SCSI bus if that is
- >appropriate. But it could go onto a parallel port much earlier than that.
-
- Well, If I'm debugging my new homebrew SCSI modem interface I wouldn't
- really care either way because a software problem in my parallel port
- modem driver could hang my machine just as quick, if not quicker than
- one in my SCSI modem driver. On the other hand I'd much prefer to
- purchase one that already works either way in which case I'll simply
- plug and play it.
-
- The ONLY "STANDARD" parallel interface on IBM stuff is the one defined
- in the timing diagrams in the back of most printer manuals which is ONE
- WAY, OUT!
-
- --
- Keith Smith uunet!ksmith!keith 5719 Archer Rd.
- Digital Designs BBS 1-919-423-4216 Hope Mills, NC 28348-2201
- Somewhere in the Styx of North Carolina ...
-