MITS 2SIO serial chip?

From: Richard Erlacher <edick_at_idcomm.com>
Date: Thu Dec 20 02:44:22 2001

You might find it amusing to know that, back in '83-'84, I designed a terminal
in two versions, one with a 68000 (overkill, and not really worth the extra
investment) and one with a 65C02. For the functions required in the terminal,
the 65C02 was MUCH faster.

The most frequently executed loop was
load left-justified ASCII byte with lsb hard-wired to '0' into index X,
jump, indirect, indexed by X through a table pointing to the exact routine
required for that character.

This loop required no stack or other memory. It simply fetched from I/O and
took appropriate action, in response to an interrupt. The rest of the time, it
scratched itself while waiting for the next interrupt. The entire process took
less time that the first instruction fetch of the 68K, not to mention its longer
interrupt response.

That's not to be taken as a general comparison, but it surely shows there were
things a 32-bit CPU couldn't do faster than an 8-bitter.

Dick

----- Original Message -----
From: "Ethan Dicks" <erd_6502_at_yahoo.com>
To: <classiccmp_at_classiccmp.org>
Sent: Wednesday, December 19, 2001 11:11 PM
Subject: Re: MITS 2SIO serial chip?


>
> --- Greg Ewing <greg_at_cosc.canterbury.ac.nz> wrote:
> > Ben Franchuk <bfranchuk_at_jetnet.ab.ca>:
> >
> > > what is the faster CPU -- A 6502 or Z80 style processor
> >
> > Back when I used to spend long blissful evenings hand-assembling Z80
> > programs [1] I got the impression that Z80 code was more compact than
> > 6502 code, being able to manipulate 16-bit values with single
> > instructions in many cases.
>
> If you had enough zero-page locations left over for your app after
> the OS was done with them, typically, 6502 code could be quite
> compact. Embedded stuff (like C= disk drives) was a little easier
> to write in that regard because you had the entire page to do as
> you wished. Infocom did just about the same with their Z-machines
> for the C-64 (which is why you had to power-down when you were done
> playing - the zero page was full of Z-machine values and the interpreter
> did not save them anywhere to restore them before exiting).
>
> I've always felt (personal impression) that a 4Mhz Z-80 could outperform
> a 1Mhz 6502, but I never got the chance to do any comparisons when faster
> chips (of both kinds) came out. I'd already moved on to 68000s. Because
> I like the 6502, I would hope that a 4Mhz 6502 could do a little better
> than a 4Mhz Z-80, but I'm not certain. The 6502 does have a one-cycle
> pipeline internally, and it does interleave CPU and memory operations
> efficiently (phi2 clock and what not), but I don't know the Z-80 to
> the same depth and can't make a fair comparison myself.
>
> -ethan
>
>
>
> __________________________________________________
> Do You Yahoo!?
> Check out Yahoo! Shopping and Yahoo! Auctions for all of
> your unique holiday gifts! Buy at http://shopping.yahoo.com
> or bid at http://auctions.yahoo.com
>
>
Received on Thu Dec 20 2001 - 02:44:22 GMT

This archive was generated by hypermail 2.3.0 : Fri Oct 10 2014 - 23:33:40 BST