Re: Low-fat samizdat -Reply
- To: "Mike Simon" <mlsimon@xxxxxxxxxxxxxxxxx>, <MISC>, <theFox@xxxxxxxxxxxxxxxxxxx>
- Subject: Re: Low-fat samizdat -Reply
- From: "Myron Plichota" <mplichot@xxxxxxxxx>
- Date: Tue, 15 Jun 1999 20:28:31 -0700
>The 8530 is a nice COM chip. Easy to interface.
>Some interesting modes (I like FM because it is phase insensitive.)
>
I agree it is a nice chip, I got a lot of experience in the 80s designing
datacomm test equipment around it, however:
With ~15 man-years experience with the SCC at the company,
I still had to write some code that finally clarified some of the
still-outstanding care and feeding issues and behaviour of the
sync pin as an output.
Then I designed a couple of companion ASICs that allowed the
capabilities that had eluded us so far. If we had the CPU
performance available today, it all could have been s/w, and
exactly what we wanted right off the mark. As you know, the
effort involved in developing a robust SCC s/w driver suite is
not trivial.
You only need to implement s/w once on a given platform.
Hardware has to be procured, assembled and tested for
every unit built. BTW the design->engineering sample cycle
for the companion ASICs was 5 months, and that was fast!
Of course the investment in the ASICs inevitably reduced the
freedom of the engineers, once they were on the scene!
>And I have source code.
>Plus I currently give away the code.
>
Anyone planning to use the SCC should take advantage of your
generous offer and make sure they have the SCC user's manual :>)
>Under no circumstances use the 8251 UART. It is a dog.
>
I tangled with that puppy back in the 80s as well. As I recall, the
filthy secret was that you had to issue the reset command 3
times to get it to work properly. This did not appear in the
Intel databooks until shortly before they obsoleted it, and
only as a tiny footnote.
Myron Plichota