Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!sdd.hp.com!hplabs!hpda!hpcuhc!hpcupt3!ken
From: ken@hpcupt3.cup.hp.com (Kenneth M. Sumrall)
Newsgroups: comp.sys.atari.8bit
Subject: Re: 65C02?
Message-ID: <48510001@hpcupt3.cup.hp.com>
Date: 16 Mar 91 22:20:59 GMT
References: <1991Mar5.110713.24023@cs.city.ac.uk>
Organization: Hewlett Packard, Cupertino
Lines: 46

>	Ooops, sorry, I guess I'm wrong. The 65C02 must have TTL line
>drivers installed. I dunno if it would make the computer faster however,
>since it is still clocked at the same clock speed, and the peripheral
>interfaces are still the same type (I think the XE series has upgraded
>PIAs). But, I would like to hear from those that modifed it. The key
>
The 65C02 performed some instructions faster than the 6502.  However, the
the flags are set correctly on decimal math mode on the 65C02 at the cost
of one addtional cycle.  The speedup would not really be noticable IMHO.

Also, the 6502 was used on the 400 and 800, but the XL and XE machines used
the a chip called 6502C or Sally.  In this chip, a few pins were changed
6502C		6502
Pin  function	pin  function
36   READ/WRITE 36   N.C.
35   /HALT      35   N.C.
34   N.C.       34   READ/WRITE

the HALT pin on the 6502C stopped the CPU clock and tri-stated the address
bus inside the CPU, instead of it being done externally like the 400 and
800 did.  Also, because of the way the CPU clock is halted, (the clock is
stretched during phase 0 of the clock at logic 0 level) the 65802 isn't
a drop in replacement fo the 6502 even on the 400 and 800 because the
65802 doesn't allow the CPU clock to be stretched low for more than
10 uS.  ANTIC stretches the clock for more than that when doing video
memory accesses.
BTW, I don't know how the DataQue 65816 board works, so don't ask.

>advantage to CMOS technology is that it requires very little power, but its
>drawbacks are speed and also driving devices. CMOS requires little power
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Not anymore.  This was true back when the RCA 4000 series CMOS chips
were made, but all the latest CPU's are done in CMOS at high
clock speeds.  The 68030 clocks at 50Mhz now, the 68040 clocks at 25Mhz
with 33 and 50 planned (I think), and the 80486 runs at 33 Mhz, with
higher speeds planned (I think).  Intel did talk about a one of a kind
100 Mhz 80486 that they got running in the lab.  And here at HP, we
have RISC processors running at 50Mhz.  As for fan out, depending
on the process, a CMOS output can drive several TTL loads, and LOTS
of CMOS loads.

Just my $0.02.

Ken Sumrall
ken%hpda@hplabs.hp.com
...!hplabs!hpda!ken