Path: utzoo!attcan!uunet!husc6!mailrus!ames!killer!elg
From: elg@killer.Dallas.TX.US (Eric Green)
Newsgroups: comp.sys.cbm
Subject: Re: Zmodem implementations
Message-ID: <7509@killer.Dallas.TX.US>
Date: 12 Mar 89 04:34:30 GMT
References: <1786.2416BB83@isishq.FIDONET.ORG>
Organization: The Unix(R) Connection, Dallas, Texas
Lines: 40

in article <1786.2416BB83@isishq.FIDONET.ORG>, izot@f171.n221.z1.FIDONET.ORG (Geoffrey Welsh) says:
>  > From: Hank@cup.portal.com (Hank W Oxford)
>  > 1)  As to 3/4 implementations, most PC BBS's are using Forsberg's own
>  >  DSZ.
>    Nonsense. PCBoard v14, Opus, QuickBBS, and plenty of others provide
> internal ZMODEM support. Terminal programs like Telix also provide
> internal 

Most of which programs are written in "C". Forsberg has released the
"C" source code to Zmodem protocol (see Unix "sz" and "rz"), which is
easily put into any program written in "C" (after inevitable
modifications to deal with OS differences... and assuming you have a
real computer to compile it on ;-).

>    I think Joe was talking about a C64 and an IEEE-488 drive. That setup 
> doesn't support DMA, either, but has no problems running the RS-232 port & 
> drives at the same time.

I wouldn't be so saguine about the perfection of simultaneous IEEE-488
and RS-232. I have tried out both the C-Link II and IEEE Flash and
found that if you do not disable the RS232 NMIs before doing disk I/O,
you will occasionally get a garbled character from the disk.
Especially bad was the case where incoming RS232 would corrupt writes
to the disk -- I lost several internal BBS file directories to that
problem, before I finally wrote a little wedgie to catch all
chkin/chkout and turn off incoming RS232.  I have not dug through
their modified Kernels to find out the exact reason (plus, in the case
of the Link it's a bit difficult -- they have 16K of EPROM, and
bank-switch the damned thing), but I suspect it's a matter of timing.
It's interesting that drive timing had a lot of significance in the
matter too -- it would happen with some SFD's, and not with others.

I suspect that the situation would be a lot better with a good IEEE
interface (e.g. the Buscard II or the original SKyles Quicksilver),
but still wouldn't trust it. 

--
|    // Eric Lee Green              P.O. Box 92191, Lafayette, LA 70509     |
|   //  ..!{ames,decwrl,mit-eddie,osu-cis}!killer!elg     (318)989-9849     |
| \X/         amoy mi amiga, pero no me gusta bazofia-DOS.