pgc@csadfa.cs.adfa.oz.au (Phil Clark) (03/26/91)
I seem to remember reading somewhere, possibly in this newsgroup, that
Applesoft basic has problems keeping up with data via the serial ports of
a //c at greater than about 1200 baud.
Does anyone know if this is true?
I have a problem trying to input a data string into a basic program via
serial port 2 at 4800 baud. I realy did not want to resort to a machine
language sub-routine for a simple function unless necessary, but it seems
as if the input statement in basic cannot handle the data at this rate.
Has anyone else experienced this problem and if so, is there a simple
answer?
Incidentally, a PC or compatible or a dumb terminal has no problem handling
the data string and the baud rate of the input cannot be changed.
The data format is 8 data bits, no parity, 2 stop bits, 4800 baud.
Thanks,
Phil Clark [VK1PC] Department of Computer Science, Phone:
Australian Defence Force Academy, +61 6 268 8157
Email: Northcott Drive, Campbell, Fax:
pgc@csadfa.cs.adfa.oz.au Canberra, Australia, 2600. +61 6 268 8581a.guillaume@trl.oz.au (andrew guillaume) (03/27/91)
In article <1991Mar26.043509.24160@sserve.cc.adfa.oz.au>, pgc@csadfa.cs.adfa.oz.au (Phil Clark) writes: > I seem to remember reading somewhere, possibly in this newsgroup, that > Applesoft basic has problems keeping up with data via the serial ports of > a //c at greater than about 1200 baud. > Has anyone else experienced this problem and if so, is there a simple > answer? > The data format is 8 data bits, no parity, 2 stop bits, 4800 baud. G'day ! I have had similar problems, at 1200 baud. Data format : 8 bits, no parity, 1 stop bit, 1200 baud. I was using Visiterm (a comms program), hooked up directly to my unix box, to download text. At 1200 baud, I would loose a character ot two every time the Apple //c would send a stop character to write to disk. No problems at 300 baud. I was wondering whether this was a software problem or whether that was linked to the serial port. Theoretically, the //c port can go up to 19.2k baud. So says the tech manual. I would also appreciate if someone could shed light on this, and give an indication as to whether the problem disappears with an upgrade (I have to upgrade my //c from a "boat anchor" :-) to a unidisk in order to be able to buy a Chinook HD). Thanks in advance. Andrew Guillaume Internet : a.guillaume@trl.oz.au
jdeitch@umiami.ir.miami.edu (Jonathan Deitch) (03/27/91)
In article <7175@munnari.oz.au>, a.guillaume@trl.oz.au (andrew guillaume) writes: > > In article <1991Mar26.043509.24160@sserve.cc.adfa.oz.au>, > pgc@csadfa.cs.adfa.oz.au (Phil Clark) writes: >> I seem to remember reading somewhere, possibly in this newsgroup, that >> Applesoft basic has problems keeping up with data via the serial ports of >> a //c at greater than about 1200 baud. > >> Has anyone else experienced this problem and if so, is there a simple >> answer? > >> The data format is 8 data bits, no parity, 2 stop bits, 4800 baud. > > G'day ! > > I have had similar problems, at 1200 baud. Data format : 8 bits, no parity, > 1 stop bit, 1200 baud. I was using Visiterm (a comms program), hooked up > directly to my unix box, to download text. At 1200 baud, I would loose a > character ot two every time the Apple //c would send a stop character to > write to disk. No problems at 300 baud. I was wondering whether this was > a software problem or whether that was linked to the serial port. > Theoretically, the //c port can go up to 19.2k baud. So says the tech > manual. > > I would also appreciate if someone could shed light on this, and give > an indication as to whether the problem disappears with an upgrade (I have > to upgrade my //c from a "boat anchor" :-) to a unidisk in order to be able > to buy a Chinook HD). Thanks in advance. > > Andrew Guillaume > > Internet : a.guillaume@trl.oz.au On the "Boat Anchor" //c, Apple had some design problems with the serial ports. It seems that they tried to derive the timing signal for the serial ports from the 14.8 mhz master timing clock. What they got was very close but not exact to what the 6551 serial chips expect. End result : garbled text and serial communications problems. This is fixed with later versions. Here's how to tell which version you have : take your //c apart and see if there is a silver TTL clock generator in the row of chips next to the power supply. If so, you shouldn't have any problems. If not so (the third back is a 74LS161) than you should be able to get a free motherboard upgrade from your friendly Apple dealer. BTW : they sometimes will give you the Unidisk 3.5" motherboard as this upgrade for nothing extra. Good Luck to all ! - Jonathan ------------------------------------------------------------------------------ Internet : jdeitch@umiami.miami.edu | "Good musicians execute ------------------------------------------------- | their music but bad ones "I'm a Time Lord. I walk in eternity !" - Dr Who | murder it !!! "
edwatkeys@pro-sol.cts.com (Ed Watkeys) (03/27/91)
In-Reply-To: message from pgc@csadfa.cs.adfa.oz.au I'm not sure about this stuff, so if anyone corrects me, trust them... There are a few issues that we're dealing with at the same time. First of all, when APple first put out the IIc, there was a timing problem with the serial ports (they used the system clock and divided it down, to the correct frequency, but it wasn't totallt 'correct',, if I recall). What it caused was data errors at speeds of 1200 ps and higher. The solution, go to your Apple dealer and ask him to check the serial number, and replace the motherboard (for free, really) if it is old enough. I had to have one of my IIcs upgraded, and it had a sn of D4B0105. OK, now BASIC... at 2400bps, there could be a character coming every 4.17E-3 seconds, which means, basically, faster than BASI can handle. Note that GETLN is written in ML, and can handle stuff at a pretty fast rate. I think the problem could come when you get sent a string, a CR and then another string. It will take the first string okay (provided that you're waiting for it), get the CR, and go to the next BASIC statement. While BASIC is picking its nose, the next line of characters is screaming by. For BBSs, this is not a problem, because most of us aren't *that* fast as typers. The only solution (the simplest) is to write the ML routines. You could get into interrupts and all that, but between you and I, Apple never really told us how to use serial interrupts with ProDOS (they did in the IIc Tech Ref, but there are a few fatal typos, and it didn't deal with ProDOS...) One other possibility is to talk to MDAVIS at the site where I am, and ask him about ModemWorks. Tell him what you need, and what you don't, and the fact that you have a IIc, and he'll tell you if ModemWorks will do the job for you... Edwin Howell Watkeys III Internet: edwatkeys@pro-sol.cts.com ProLine: edwatkeys@pro-sol UUCP: crash!pro-sol!edwatkeys ARPA: crash!pro-sol!edwatkeys@nosc.mil BitNET: edwatkeys%pro-sol.cts.com@nosc.mil
pgc@csadfa.cs.adfa.oz.au (Phil Clark) (03/28/91)
From article <1991Mar26.191258.8367@umiami.ir.miami.edu>, by jdeitch@umiami.ir.miami.edu (Jonathan Deitch): > In article <7175@munnari.oz.au>, a.guillaume@trl.oz.au (andrew guillaume) writes: >> >> In article <1991Mar26.043509.24160@sserve.cc.adfa.oz.au>, >> pgc@csadfa.cs.adfa.oz.au (Phil Clark) writes: >>> I seem to remember reading somewhere, possibly in this newsgroup, that >>> Applesoft basic has problems keeping up with data via the serial ports of >>> a //c at greater than about 1200 baud. (Lots & lots of unnecessary deleted!) > On the "Boat Anchor" //c, Apple had some design problems with the serial ports (more lots deleted) >BTW : they sometimes will give you the Unidisk 3.5" > motherboard as this upgrade for nothing extra. Yes, they will do this for you free!! of charge and the unidisc upgrade is included. I had this done but the problem is still there even afterward. Any more ideas?? Phil Clark [VK1PC] Department of Computer Science, Phone: Australian Defence Force Academy, +61 6 268 8157 Email: Northcott Drive, Campbell, Fax: pgc@csadfa.cs.adfa.oz.au Canberra, Australia, 2600. +61 6 268 8581
veale@Colorado.EDU (Anthony Veale) (03/28/91)
pgc@csadfa.cs.adfa.oz.au (Phil Clark) writes: >> <7175@munnari.oz.au>, a.guillaume@trl.oz.au (andrew guillaume) writes: >>> pgc@csadfa.cs.adfa.oz.au (Phil Clark) writes: >>>> I seem to remember reading somewhere, possibly in this newsgroup, that >>>> Applesoft basic has problems keeping up with data via the serial ports of >>>> a //c at greater than about 1200 baud. > (Lots & lots of unnecessary deleted!) >> On the "Boat Anchor"//c, Apple had some design problems with the serial ports > (more lots deleted) >>BTW : they sometimes will give you the Unidisk 3.5" >> motherboard as this upgrade for nothing extra. >Yes, they will do this for you free!! of charge and the unidisc upgrade is >included. I had this done but the problem is still there even afterward. >Any more ideas?? >Phil Clark [VK1PC] Department of Computer Science, Phone: >pgc@csadfa.cs.adfa.oz.au Canberra, Australia, 2600. +61 6 268 8581 I also have the "Boat Anchor" flavor of //c. I bought it before the Unidisk was released and have never upgraded. Yet, here I type away happily at 2400 baud. Occasionally line noise adds a few characters to what I type and my 'i' key is apt to double strike. But I have had no downloading losses at all. Well, all right, sometimes I have had everything hang, but I am talking all or none. This is not dropping of characters-- this is one of the modems or the campus network getting stuck due to line noise sending special chars or some- thing. I have never had any opportunity to run faster than 2400 baud so it may indeed have problems, but 12 and 2400 baud work fine for me. Sorry, that I have no suggestions for a solution, but even negative information can help isolate the problem. -- Anthony Veale' veale@coma.colorado.edu You're failing to take into Grad. Student VEALE@JILA.BITNET account something fairly basic in Dept. Physics JILA::VEALE the relationship between men and UColorado Boulder robots. - Marvin
jdeitch@umiami.ir.miami.edu (Jonathan Deitch) (03/28/91)
In article <1991Mar28.013526.17057@sserve.cc.adfa.oz.au>, pgc@csadfa.cs.adfa.oz.au (Phil Clark) writes: > From article <1991Mar26.191258.8367@umiami.ir.miami.edu>, by jdeitch@umiami.ir.miami.edu (Jonathan Deitch): >> In article <7175@munnari.oz.au>, a.guillaume@trl.oz.au (andrew guillaume) writes: >>> >>> In article <1991Mar26.043509.24160@sserve.cc.adfa.oz.au>, >>> pgc@csadfa.cs.adfa.oz.au (Phil Clark) writes: >>>> I seem to remember reading somewhere, possibly in this newsgroup, that >>>> Applesoft basic has problems keeping up with data via the serial ports of >>>> a //c at greater than about 1200 baud. > > (Lots & lots of unnecessary deleted!) > > >> On the "Boat Anchor" //c, Apple had some design problems with the serial ports > > (more lots deleted) > >>BTW : they sometimes will give you the Unidisk 3.5" >> motherboard as this upgrade for nothing extra. > > > Yes, they will do this for you free!! of charge and the unidisc upgrade is > included. I had this done but the problem is still there even afterward. > > Any more ideas?? > > > Phil Clark [VK1PC] Department of Computer Science, Phone: > Australian Defence Force Academy, +61 6 268 8157 > Email: Northcott Drive, Campbell, Fax: > pgc@csadfa.cs.adfa.oz.au Canberra, Australia, 2600. +61 6 268 8581 I have no idea what could cause this problem to continue, unless it's your modem or cable or both at fault - or telephone line noise. I've got more unathorized stuff in my //c like MIDI ports and the like and have desoldered and moved serial chips and the like and haven't had a bit of trouble. I'm using the //c right now via modem to call in to Miami's VAX. I'm stumped as to a computer based cause of your problem. Try using a different modem or taking your //c to the computer center and using a direct connect line with a shielded cable. Other than that, I with you luck ! - Jonathan ------------------------------------------------------------------------------ Internet : jdeitch@umiami.miami.edu | "Good musicians execute ------------------------------------------------- | their music but bad ones "I'm a Time Lord. I walk in eternity !" - Dr Who | murder it !!! "
vw3@cunixf.cc.columbia.edu (Vernon Williams) (03/28/91)
In article <8214@crash.cts.com> edwatkeys@pro-sol.cts.com (Ed Watkeys) writes: >In-Reply-To: message from pgc@csadfa.cs.adfa.oz.au > >I'm not sure about this stuff, so if anyone corrects me, trust them... > >There are a few issues that we're dealing with at the same time. First of >all, when APple first put out the IIc, there was a timing problem with the >serial ports (they used the system clock and divided it down, to the >correct frequency, but it wasn't totallt 'correct',, if I recall). What it System clock??? The Apple IIc doesn't have a System Clock (unless you buy one from AE) I think the proplem was in a crystal. [much stuff deleted] >typers. The only solution (the simplest) is to write the ML routines. You >could get into interrupts and all that, but between you and I, Apple never >really told us how to use serial interrupts with ProDOS (they did in the >IIc Tech Ref, but there are a few fatal typos, and it didn't deal with >ProDOS...) One other possibility is to talk to MDAVIS at the site where I You might also try the ProDOS Technical Reference Manual and/or Beneath Apple ProDOS. I think Apple tells how to deal with interrupts, it's just that it's in the ProDOS manual. >am, and ask him about ModemWorks. Tell him what you need, and what you >don't, and the fact that you have a IIc, and he'll tell you if ModemWorks >will do the job for you... From what I've seen of ModemWorks...it probably will fit most users' needs. > >Edwin Howell Watkeys III [Sig deleted] +---------------------------+-------------------------------------------------+ |Vernon L. Williams | Electronic Mail: vw3@cunixf.cc.columbia.edu | |Thomas J. Watson Library | Campus Mail: 130 Uris Hall | |Columbia University | Phone Mail: 212-854-6798 |
greggb@pro-fred.cts.com (System Administrator) (03/29/91)
The upgrade from "Boat Anchor" //c to Uni-Disc compatible //c solves most
of the serial port problems. 19200 seems to garble a bit when scrolling text,
but for y or z modem transfers, it is a champ...
-----------------------------------------------------------------------------
ProLine : greggb@pro-fred InterNET : greggb@pro-fred.cts.com
UUCP : crash!pnet01!pro-fred!greggb
-----------------------------------------------------------------------------
Listening for wisdom, weeping bitter tears. Eyes like | REZ
diamonds hearts like steel. Murdering truth for passionless | 'n
gain. The gifted youth, kneeling at the altar of pain. | ROLL!