Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!caip!nike!ucbcad!ucbvax!sdcsvax!sdcc6!ix742
From: ix742@sdcc6.UUCP (James Hayes)
Newsgroups: net.micro.apple
Subject: Re: "dcom trouble, request for answer from jim hayes"
Message-ID: <2706@sdcc6.UUCP>
Date: Sun, 29-Jun-86 00:29:18 EDT
Article-I.D.: sdcc6.2706
Posted: Sun Jun 29 00:29:18 1986
Date-Received: Mon, 30-Jun-86 04:45:51 EDT
References: <45@excalibur.UUCP>
Reply-To: ix742@sdcc6.UUCP (James Hayes)
Distribution: net
Organization: UC San Diego EE/CS Department.
Lines: 81
Summary: The author speaks...

188622462@excalibur.UUCP (Greene T Richard) writes in 
article <45@excalibur.UUCP>:
>
>1) There is an error message quickly flashed on the top line just before 
>   the screen is cleared.  On using a ctrl-S at the right spot, I saw the
>   message was:
>      *Serial Init Error*
>   The screen would then go blank, and no key except ctrl-RESET would get any
>   response, including the open-apple combinations.

A few people have this problem.  I suspect the SSC card has
interupts enabled.   I've tested each configuration on all the
hardware drivers I've written.  They worked for me...

The "*Serial Init Error*" is part of a kludge to detect a problem in
itializing the serial card.  It appears EVERY time the program
boots.  If the serial card init works, the message is cleared, if
the serial card hangs during the init, the message stays.  (I know
it's a bad kluge, but it works pretty well.)

>
>2) The screen clears, then the program crashes me into binary, giving me this
>   memory dump:
>      5955- A=35 X=35 Y=F0 P=34 S=F1
>   I think that the Serial Init Error may also appear before this, but I have
>   been unsuccessful in catching it, if it exists.
>

This might be a genuine bug.  Most of the code is very tight and
does not bother saving registers,  the routines are designed to run
fast.  The trade off here is when you start to mix and match
video/serial drivers.  Some may interact differently (and crash)
with each other...

Here are the combinations that I have PERSONALLY TESTED:

Apple //e + 80 columns and : Super serial card
			   : Mountain Hardware CPS card

Apple //c + Apple modems

Apple ][+, VIDEX 80 columns and : Super serial card
				: CCS 7710 serail card

Apple ][, Omnivision 80 and ALS dispatcher

>One more piece of hardware info,  I do have some problems with my slots
>and card connecting properly.  With my other program (ASCII Express), I have
>had to re-boot several times, each time wiggling the card in its slot to 
>get a successful connect.

This could cause a problem.

-----
General note:

Some of the drivers were two configuration specific,  I fixed them and
released a patch file called PATCH.EXEC that also contained the VIDEX
driver.  The new updated drivers in PATCH are very hardware oriented
and assume very little about your configuration.  Some people have perfect
results with the old drivers and some have perfect results with the new
drivers...  These bugs are very hard to track down.

My mailbox seems to indicate much success with the program (barring
"cosmetic" things like schpelling errors).  So I really have nothing
more to offer than phrases like:
"Check your dipswitches" -or- "Get a fresh copy." -or- "I'm working
on it."

My thanks the to net people that have made my UNIX mailbox file to
it's current 201K.  I have discovered that "net" people are
not like the rest of the world... The word "human" comes to
mind.  I really appreciate the comments/bug reports.

-Jim Hayes, UCSD

ARPA: hayes@UCSD.ARPA
UUCP: {ihnp4, ucbvax, decvax, etc}!sdcsvax!hayes

P.S.  Now that school is out, I'll have more time to add new
features, check this group out in September...