11 Apr
1999
11 Apr
'99
1:27 p.m.
<Well, that's not very far from what I wrote, is it? I was just pointing
<out that although Allison seemed to imply that a 6 or 8MHz Z80 was much
<faster than a 4MHz(? I haven't got the original message any more) 6502, I
<believe that to be far from the case.
I was pointing out that is the processor was running fast enough even a
dog can look good. ;) Obviously using a 8mhz z80 as the standard your
comparison CPU had better be of similar generational speed or it may fail
the test. the inverse is with a 33mhz z185 I know I can blow the 65c02
out of the water unless someone has at least a 25-30mhz 6502!
Allison
12 Apr
12 Apr
11:06 a.m.
New subject: 6502 vs Z80, round 97
I was pointing out that is the processor was running
fast enough even a
dog can look good. ;) Obviously using a 8mhz z80 as the standard your
comparison CPU had better be of similar generational speed or it may fail
the test. the inverse is with a 33mhz z185 I know I can blow the 65c02
out of the water unless someone has at least a 25-30mhz 6502!
I think I remember reading that the 6502 was eventually included in gate
array logic libraries, so that GaaS parts with Ghz clocks likely exist
(perhaps only embedded in other designs though).
One of the things I have noticed is that great similarity exists between a
couple of instruction sets, the 6502 and the Sparc, and the IBM 360 and
68000. Kind of the classics of RISC and CISC, and I love them both.
Arguing about the 6502 vs Z80 makes about as much sense as arguing about
which cart is faster without discussing the horse or load. As I remember it
speed was a big "talking" point, but the real issue for most power users
was capacity, not speed.
5:48 p.m.
New subject: Round 98 (was Re: 6502 vs Z80, round 97)
Mike Ford wrote:
One of the things I have noticed is that great
similarity exists between a
couple of instruction sets, the 6502 and the Sparc, and the IBM 360 and
68000. Kind of the classics of RISC and CISC, and I love them both.
Um, what are you smoking, and where can I get some? :-)
Having programmed all four of those processors, I would not be willing
to concede much similarity between any two of the four.
7:53 p.m.
New subject: Round 98 (was Re: 6502 vs Z80, round 97)
Mike Ford wrote:
Maybe you forgot to enhale, there aren't more than a handfull of different
instructions between the pairs I mention. I haven't programmed the Sparc,
but I did spend about two weeks testing the feasibility of porting a 65C02
assembly language program to the Sparc, and it looked very very good. As
for the 360/68k, IBM made a varient of the 68k that ran 360 code native.
Wow this sure is sounding like a bunch of old coots isn't it? ;)
One of the things I have noticed is that great
similarity exists between a
couple of instruction sets, the 6502 and the Sparc, and the IBM 360 and
68000. Kind of the classics of RISC and CISC, and I love them both.
Um, what are you smoking, and where can I get some? :-)
Having programmed all four of those processors, I would not be willing
to concede much similarity between any two of the four.
13 Apr
13 Apr
10:04 a.m.
New subject: Round 98 (was Re: 6502 vs Z80, round 97)
Mike Ford wrote:
One of the things I have noticed is that great
similarity exists between a
couple of instruction sets, the 6502 and the Sparc, and the IBM 360 and
68000. Kind of the classics of RISC and CISC, and I love them both.
I replied:
Having programmed all four of those processors, I
would not be willing
to concede much similarity between any two of the four.
Mike replied:
Maybe you forgot to enhale, there aren't more than
a handfull of different
instructions between the pairs I mention.
Well, sure. How many different names for "add" can you come up with. But the
register sets are different, the addressing modes are different, the condition
codes are different, and the subroutine linkage methods are different. If
your criteria for processors being similar is that they have add and subtract
instructions, allmost all microprocessors are similar.
I haven't programmed the Sparc,
but I did spend about two weeks testing the feasibility of porting a 65C02
assembly language program to the Sparc, and it looked very very good.
I've been there and done that (binary translation from 6502 to Sparc) as part
of my video game simulation project. It did not map all that directly. The
6502 is little-endian and the Sparc is big-endian. And that's just the
beginning of the story.
As for the 360/68k, IBM made a varient of the 68k that
ran 360 code native.
I've got one.
The customized 68k has *completely* different microcode, and other changes to
the instruction decode logic. And it *still* didn't implement anywhere close
to the full 370 architecture. They had a second 68000 (a normal one) that
was used to emulate the instructions that they couldn't implement in microcode
on the hacked 68k. And a majorly hacked Intel 8087 to do the IBM radix-16
floating point.
The 360/370 don't even have a *stack*! How you can say with a straight face
that the 360 is similar to a 68k is beyond me.
16 Apr
16 Apr
12:31 p.m.
New subject: Round 98 (was Re: 6502 vs Z80, round 97)
>> One of the things I have noticed is that
great similarity exists between a
>> couple of instruction sets, the 6502 and the Sparc, and the IBM 360 and
>> 68000. Kind of the classics of RISC and CISC, and I love them both.
>Um, what are you smoking, and where can I get
some? :-)
>Having programmed all four of those processors, I
would not be willing
>to concede much similarity between any two of the four.
Maybe you forgot to enhale,
Could this be a common problem over at your side of the pond ?
there aren't more than a handfull of different
instructions between the pairs I mention. I haven't programmed the Sparc,
but I did spend about two weeks testing the feasibility of porting a 65C02
assembly language program to the Sparc, and it looked very very good.
I did a port of an 8080 code fragment to a /370 class CPU, and it
also looked quite good and straight foreward, althrough there is
no simirarity between the CPUs at all - so where is the point ?
I'm not 100% shure for the SPARC, but AFAIK it, there are no
similarities beyound anything common to most processors.
The same is especialy true for the 68K/360 pair - just to start with
one of the most important features of the 68K: Addressing modes -
the 360 got just ONE (base+offset[+index]) not to mention increment
or decrement instructions at all. And of course as a result no stack!
Hey, nothing against the /360/370 system I found that this singe
addressing mode is flexible enough to get any result needed as easy
as with an 68K processor with his dozens or so modi - but just not
compatible in any way.
As
for the 360/68k, IBM made a varient of the 68k that ran 360 code native.
Thats not an argument for either side, I don't know the IBM solution,
but in the mid 80s SIEMENS got a similar machine, the PC-2000 (a
NSC 16016 Multibus Unix system with a /370ish board), based on a
68020 with new microcode to emulate a X4 like CPU (a /370 compatible
CPU from SIEMENS with a lot of enhancements, like bit field operations
or a some stack like instructions (no, no push or pop, but rather
some kind of queuemanagement to enqueue and dequeue structures and
follow (almost) unlimited levels of indirection)). But this machine
was no longer any 68K - just the silicon has been the same.
So, where is your point of similarity ?
Gruss
H.
--
Stimm gegen SPAM: http://www.politik-digital.de/spam/de/
Vote against SPAM: http://www.politik-digital.de/spam/en/
Votez contre le SPAM: http://www.politik-digital.de/spam/fr/
Ich denke, also bin ich, also gut
HRK
9211
days inactive
9216
days old
5 comments
4 participants
participants (4)
-
allisonp＠world.std.com -
eric＠brouhaha.com -
Hans.Franke＠mch20.sbs.de -
mikeford＠netwiz.net