STEVENS@SENECA.BITNET (11/29/88)
i just received a copy of Kermit. and after four days i got it working! sort of anyway! I have an Apple //e an extended 80 col. card, a ccs 7710 com card. the only thi ng cloned is my drive controller. Everything works well until the data hits the lower left hand corner and begins to scroll! then the first 3 or 4 characters get fuddled! EExmple! (if this is the bottom line) i also have Ascii Express Professional and it does the same thing! all other fun ctions seem to work fine! In local mode I don't have this problem! i have changed the 80 col. card, the com card, and all the Ram! the only thing e lse is the controller card and the mother board! self test says the Kernal is OK ! so there we have it! I only have one question. ... WHY!!?? If any one can help i sure would appreciate it! Thanks Murray Stevens Stevens@Seneca (any spelling mistakes are a result of a poor upbringing) *thanks Mom*
toth@tellab5.tellabs.CHI.IL.US (toth@tellab2 (Tellabs, Inc., - Lisle, Il.)) (12/02/88)
I had the same type of problem with an AST Multi/IO card. It continuously overflowed my Imagewriter 1k. print byffer during output of any reasonably long basic program. I would enter the following: >load 'any basic program' >pr #1 >list <this line is directed to the printer along with all output> after about 2/3 of a page, the output got munged because the device driver would not recognize the CTS signal and just kept on dumping. similarly on the device in slot 2, it never toggles its CTS to stop input from the modem when it goes into the long delay to advance the screen when a CR is done at the bottom of the screen. This is necessary since it doesn't handle interrupts to receive input during times where the processor is busy doing something else. Asside from the print problems, the clock has the Year go back to 1971 or something like that when entering 1988 or being set to 1988. :-( If anybody knows what to do to get the AST card to work, I would also like to hear about it. AST manuals give no information about register addresses to configure the device to use the control signals. I have given up since 2 letters to AST went unanswered. ( another :-( ) -- ------------------------------------------------+--------------------- Things can't really be as bad as they seem... | Joseph G. Toth Jr. Or CAN they!!!! | ihnp4!tellab5!toth
gwyn@smoke.BRL.MIL (Doug Gwyn ) (12/02/88)
In article <8811282050.aa14476@SMOKE.BRL.MIL> STEVENS@SENECA.BITNET writes: > I have an Apple //e an extended 80 col. card, a ccs 7710 com card. > Everything works well until the data hits the lower left hand corner and begins > to scroll! then the first 3 or 4 characters get fuddled! >i also have Ascii Express Professional and it does the same thing! My guess is that your Apple //e has not been "enhanced" (ROM upgrade) and that consequently the old 80-column firmware is taking too long during scrolling so that incoming data is lost. The //e enhancement was free at authorized Apple dealers as I recall.
SEWALL@UCONNVM.BITNET (Murph Sewall) (12/03/88)
>i just received a copy of Kermit. and after four days i got it working! >sort of anyway! > I have an Apple //e an extended 80 col. card, a ccs 7710 com card. the only >thing cloned is my drive controller. > Everything works well until the data hits the lower left hand corner and >begins > to scroll! then the first 3 or 4 characters get fuddled! >EExmple! (if this is the bottom line) Looks like the old "serial card doesn't support interrupts" phenomenon. Try SET FLOW XON and see if that helps any. I'd bet the CCS 7710 driver isn't one Ted's had much call to look at recently. You could ask: Ted Medin <medin-t@shark.nosc.mil> [The Far Side shall return (I hope)] Murph Sewall Sewall@UCONNVM.BITNET Business School sewall%uconnvm.bitnet@mitvma.mit.edu [INTERNET] U of Connecticut {rutgers psuvax1 ucbvax & in Europe - mcvax} !UCONNVM.BITNET!SEWALL [UUCP] -+- My employer isn't responsible for my mistakes AND vice-versa! (subject to change without notice; void where prohibited) "Close enough for government work" - source unknown (naturally ;-)
joseph@elbereth.rutgers.edu (Seymour Joseph) (12/03/88)
Sorry, but you remember wrong. The //GS rom upgrade was free, the Apple //e Enhancement package usually costs $70. The enhancement contains two firmware roms, a new character generator with the Mousetext characters in it and a 65C02 processor. Unless you are handy with chip insertion, it is a dealer installed upgrade and they can charge you for labor. Seymour
gwyn@smoke.BRL.MIL (Doug Gwyn ) (12/04/88)
In article <8812021714.aa20222@SMOKE.BRL.MIL> SEWALL@UCONNVM.BITNET (Murph Sewall) writes: >Looks like the old "serial card doesn't support interrupts" phenomenon. The CCS 7710 does support interrupts, although I have no idea whether Kermit uses them. However, it will do no good under the non-enhanced //e 80-column firmware, which disables interrupts for extended periods.
tucker@unocss.UUCP (Gregory A. Tucker) (12/05/88)
From article <8811282050.aa14476@SMOKE.BRL.MIL>, by STEVENS@SENECA.BITNET: > i just received a copy of Kermit. and after four days i got it working! > sort of anyway! > I have an Apple //e an extended 80 col. card, a ccs 7710 com card. the only thi > ng cloned is my drive controller. > Everything works well until the data hits the lower left hand corner and begins > to scroll! then the first 3 or 4 characters get fuddled! > EExmple! (if this is the bottom line) > > i also have Ascii Express Professional and it does the same thing! all other fun > ctions seem to work fine! In local mode I don't have this problem! > > i have changed the 80 col. card, the com card, and all the Ram! the only thing e > lse is the controller card and the mother board! self test says the Kernal is OK > ! > so there we have it! I only have one question. ... WHY!!?? > If any one can help i sure would appreciate it! > > Thanks Murray Stevens > Stevens@Seneca > (any spelling mistakes are a result of a poor upbringing) *thanks Mom* I don't think there is one Apple user that hasn't had to fuddle with this problem at one time or another in their Apple hacking career. Let me explain the problem and the solutions. The whole problem is simply this: the Apple isn't fast enough to scroll the 80 column screen before the next few bytes of information are received by the modem. The normal way of getting around this problem is to turn on interrupts, which will automatically stop the screen scrolling process and have the software put the x number of bytes in a temporary holding place until the screen scroll can finish. However, you can still have problem even if your interrupts are set when software, such as Kermit, uses the ROMs to do the 80 column scrolling and your ROMs are not the enhanced ones. The original ROMs were so bug ridden that the scroll routines masked out interrupts during almost the entire scroll process, resulting in the dropped characters you now have. Here are my suggestions on repairing the problems: 1- Check to make sure interrupts are set. However, this won't do any good with Kermit, assuming you have the unenhanced ROMs. I don't what AE Pro did to deal with the problem. Perhaps your manual would say. 2- With Kermit, put two commands in your Kermit.init file set flow xon set flow delay 100 This will take care of the problem, but it will also make things slow going. This sequence simply tells the host to stop sending data until Kermit can scroll the screen. You may need to adjust the 100 after delay up or down. (100 works on my system.) 3- Get a copy of ATP, this only 100% compatible VT-100 emulator. (That I am aware of.) It is share ware, but it is good quality nonetheless. The programmer dealt with the interrupt problem by completely rewriting the scroll routines such that they work with no problems up to 9600 baud. (I have not tested this.) 4- Enhnace your ROMs, if they are unenhanced. This will cost you $70, though, and the ROMs aren't that much better that it is necessarily worth the price. Fixing bugs is something I expect a company to do for free. -- ------------------------------+------------------------------------------- Gregory A. Tucker- Sophmore | Internet: tucker%zeus.dnet@fergvax.unl.edu Electrical Engineering | Bitnet: TUCKER@UNOMA1 Univ. of Nebraska at Lincoln | UUCP: uunet!btni!unocss!tucker
toth@tellab5.tellabs.CHI.IL.US (J. G. Toth Jr.) (12/06/88)
In order to eliminate confusion, my system was purchased as an Enhanced //e
and included the Apple 80 column card an an Apple Super Serial card to drive
my ImageWriter. I purchased the AST Multi/IO card around two years ago.
I use Kermit for all of my operations when using a modem.
A friend let me try his Apple Access II and Ascii Express Pro when I bought
my AST card.
Kermit not only allows interrupts, but REQUIRES them. The Apple Super Serial
card works perfectly when the strap is set to enable the interrupts, and the
80 column card does not get in the way at all..
My friends programs work with the Super Serial Card even with the strap set
to enable interrupts since they override the processor interrupt. I don't
know how they do it, but they have no problems when the page scrolls.
Since Kermit worked, I didnt't bother purchasing either of the others,
especially since we couldn't get vt100 emulation working on the Ascii
Express Pro ( but that is beside the point ), and use Kermit with my
Apple Super Serial Card.
Not one of the three programs work with AST card using the 6810 chip.
Kermit - since I don't have information on how to enable interrupts,
or the RS-232 control signals for data flow control.
the other two - due to timing since the cards ignore the use of the
RS-232 control signals for data flow control.
The big problem again comes to the lack of adequate documentation and support
for products.
I had to search for information on the 6810 chip used on the AST card hoping
to find out more about their register formats for word length / parity
control and interrupt operation control, since there is only a short section
in AST's manual about configuring word length/parity.
What I find is, it appears that AST uses a memory location for programs to
write the word length / parity configuration, and then the ROM firmware
reads this an writes to the 6810 registers. This means that the average user
has no way of configuring some of the other NICE features of the 6810 chip.
Even people who know how to use all the features of a 6810 won't know how
to use it when on one of these cards since there is no correlation between
the chip's registers and the memory addresses used to set the information
in the register ( why can't the registers be directly mapped to an address? ).
Their manual gives no information on how to enable the interrupts.
And the worst part is that it generally takes so long to find out that there
are serious flaws in the design of a device that your card is out of warrenty
and/or are unable to return the card for a refund.
Since I have yet to hear from my letters to AST, I hope that somebody is out
there listening so that manufacturers change their policies. I envision
being told that the fixes are in a PROM upgrade that will cost me almost
as much as the card did when new, and/or that I need to purchase a detailed
programmers manual to get the fixes/information that should have been part
of the package to begin with...
--
------------------------------------------------+---------------------
Things can't really be as bad as they seem... | Joseph G. Toth Jr.
Or CAN they!!!! | ihnp4!tellab5!tothSEWALL@UCONNVM.BITNET (Murph Sewall) (12/13/88)
>during scrolling so that incoming data is lost. The //e enhancement >was free at authorized Apple dealers as I recall. THAT enhancement not only wasn't (isn't) free, the last time I checked it was on the order of $80. Other mail I've gotten says that the Internet, and MILnet in particular, has been waging mighty battle with attempts to gain "unauthorized access" to computing resources (more 'virus' writers testing their "skills" ??). That may explain why posts to info-apple from 2 Dec are just now beginning to arrive in the mailboxes of BITNet subscribers (only real short ones too, longer messages may not arrive for another week :-( [The Far Side shall return (I hope)] Murph Sewall Sewall@UCONNVM.BITNET Business School sewall%uconnvm.bitnet@mitvma.mit.edu [INTERNET] U of Connecticut {rutgers psuvax1 ucbvax & in Europe - mcvax} !UCONNVM.BITNET!SEWALL [UUCP] -+- My employer isn't responsible for my mistakes AND vice-versa! (subject to change without notice; void where prohibited) "Close enough for government work" - source unknown (naturally ;-)