On Thu, Jan 16, 2014 at 12:08 PM, Sean Conner <spc at conman.org> wrote:
It was thus said that the Great Eivind Evensen once
stated:
On Tue, Jan 14, 2014 at 10:28:20AM +0000, Peter
Corlett wrote:
... t more reliable than the 14MHz 68010 I tried
back in 1990, and
which was sent back within the week as Not Fit For Purpose.)
Just curious, were the problems related to hardware or software relying
on cpu timing and the likes?
First off, the 68000 and the 68010 had a different stack layout for
exceptions, so the operating system needed patching.
In the case of the Amiga, dropping in a '010 was fine from the OS standpoint.
Also, at least one
instruction (MOVE SR,dest) was bumped from a user instruction to a
priviledged instruction, so further changes to the operating system were
required (to emulate that instruction in userspace).
The calculator app that came with AmigaDOS 1.0 and 1.1 used a DBcc
instruction and threw an error. There was a simple error trap handler
one could install. Starting with AmigaDOS 1.2, no component of the OS
flinched if you had a '010, but some games still used privileged
instructions.
I had an 8MHz 68010 in my A1000 for most of the time I used it. I got
a real-world speed boost of about 5%. I also originally had a non-DMA
disk interface, so the transfer loop was greatly enhanced by the
1-instruction cache, so I liked it, and the chip cost less than $20,
so it was the cheapest way to enhance an A1000 (or A500).
I never tried it in any other machine, so I can't speak to
compatibility issues, but because instruction restart really worked,
the 68010 _was_ used for early UNIX machines where a simple 68000
system wouldn't have worked (there were some elaborate schemes to
support demand-paged VM with a real 68000, including having two 68000s
and letting the second one take over to handle things when the first
page-faulted).
-spc (Who loved programming on the 68000 ... )
One of my favorite architectures ever. I feel privileged to have made
my living for several years writing code for it.
-ethan