25 Dec
2010
25 Dec
'10
2:30 a.m.
Does anyone have information on undocumented 8080 instructions? There's
plenty of information out there about undocumented Z-80 and 8085
instructions, but I haven't found anything on the 8080.
Some of the non-Intel 8080 and 8080-compatible parts were independent
designs, rather than licensed second-sources using Intel-provided masks,
so those might have different behavior for the undocumented opcodes.
Thanks,
Eric
25 Dec
25 Dec
1:07 p.m.
On 25 Dec 2010 at 0:30, Eric Smith wrote:
Does anyone have information on undocumented 8080
instructions?
There's plenty of information out there about undocumented Z-80 and
8085 instructions, but I haven't found anything on the 8080.
Some of the non-Intel 8080 and 8080-compatible parts were independent
designs, rather than licensed second-sources using Intel-provided
masks, so those might have different behavior for the undocumented
opcodes.
I recall investigating this after Intel the undocumented 8085
instructions. To the best of my recollection, the undocumented 8080
opcodes are, by and large, no-ops or redundancies (i.e. they do the
same as a close-valued opcode probably from incomplete PLA decoding).
My test was pretty simple--plug the opcode, followed by two 00 bytes
(in case the opcode might be interpreted as a 3-byte instruction,
execute, check the flags and registers and location 0000 to see if
anything has changed, lather, rinse, repeat. No startling new
"triple precision divide" instructions.
I'm certain that if I had discovered some fabulous new instruction, I
would have certainly documented it.
--Chuck
2:58 p.m.
Chuck Guzis wrote:
To the best of my recollection, the undocumented 8080
opcodes are, by and large, no-ops or redundancies (i.e. they do the
same as a close-valued opcode probably from incomplete PLA decoding).
[...]
No startling new "triple precision divide"
instructions.
I wasn't expecting any. The reason I'm asking is to be able to
replicate the behavior in a simulator, in case any software happened to
use them (intentionally or otherwise).
Eric
3:20 p.m.
On 25 Dec 2010 at 12:58, Eric Smith wrote:
I wasn't expecting any. The reason I'm asking
is to be able to
replicate the behavior in a simulator, in case any software happened
to use them (intentionally or otherwise).
My own 8080 systems have been so long out of use that they probably
aren't functional. But if anyone has one kicking around, I can
probably dig up the code I used to test things.
--Chuck
8:18 p.m.
From: eric at brouhaha.com
Chuck Guzis wrote:
Hi
It seemed like I recalled the NEC 8080's had some extras but it
must have been brain rot on my part. I have one of their 8080's
in my Poly88. It is quite easy to single step with this computer
and display all the registers, memory pointed to by double
registers and flags. I found nothing unusual happening on any
of the other unused instructions. I have a newer AMD8080 I guess
I could try some day.
The Poly88 had one of the best software monitors I've ever
seen on an early uP system.
Dwight
To the best of my recollection, the undocumented
8080
opcodes are, by and large, no-ops or redundancies (i.e. they do the
same as a close-valued opcode probably from incomplete PLA decoding).
[...]
No startling new "triple precision
divide" instructions.
I wasn't expecting any. The reason I'm asking is to be able to
replicate the behavior in a simulator, in case any software happened to
use them (intentionally or otherwise).
Eric
10:07 p.m.
On 25 Dec 2010 at 17:18, dwight elvey wrote:
It seemed like I recalled the NEC 8080's had some
extras but it
must have been brain rot on my part. I have one of their 8080's
in my Poly88. It is quite easy to single step with this computer
and display all the registers, memory pointed to by double
registers and flags. I found nothing unusual happening on any
of the other unused instructions. I have a newer AMD8080 I guess
I could try some day.
I seem to recall that the problem with the NEC 8080 was somewhat
different treatment of the status flags. Apparently, NEC thought to
"improve" on the 8080's treatment of the flags. Good idea, but...
IMSAI let some of its early CPU boards out with the NEC CPU. Later,
I think they used up their stock of the NEC chips by using them in
their floppy controller.
Later revisions of the NEC 8080 (e.g. 8080AF) conformed with Intel's
interpretation.
I don't recall any added instructions, however.
--Chuck
26 Dec
26 Dec
2:43 p.m.
In article <4D164D02.6030004 at brouhaha.com>,
Eric Smith <eric at brouhaha.com> writes:
Chuck Guzis wrote:
Didn't the visual 6502 guys also have a project to do the 8080?
http://www.visual6502.org/
--
"The Direct3D Graphics Pipeline" -- DirectX 9 draft available for download
<http://legalizeadulthood.wordpress.com/the-direct3d-graphics-pipeline/>
Legalize Adulthood! <http://legalizeadulthood.wordpress.com>
To the best of my recollection, the undocumented
8080
opcodes are, by and large, no-ops or redundancies (i.e. they do the
same as a close-valued opcode probably from incomplete PLA decoding).
[...]
No startling new "triple precision
divide" instructions.
I wasn't expecting any. The reason I'm asking is to be able to
replicate the behavior in a simulator, in case any software happened to
use them (intentionally or otherwise).
4962
days inactive
4963
days old
6 comments
4 participants
participants (4)
-
cclist＠sydex.com -
dkelvey＠hotmail.com -
eric＠brouhaha.com -
legalize＠xmission.com