9 Dec
2005
9 Dec
'05
2:49 a.m.
Subject: Re: Legacy apps in Windows/OS X was Re: Old MS-DOS & Win Software
From: woodelf <bfranchuk at jetnet.ab.ca>
Date: Thu, 08 Dec 2005 19:30:35 -0700
To: General Discussion: On-Topic and Off-Topic Posts <cctalk at
classiccmp.org>
What I do find odd that the 8086 came out first then a few years later
the 8088.
It still think the 8086/8086 could have had two instruction sets --
8080 emulation and
a better 16 bit design.
Not that I know of. I remember being able to get 8088s before the 86s
and that was back in 78 or early 79.
It was quite somethig to take an 8088, 8284(clock gen), 8205 (aka74138),
8755, 8155 and have a complete 16bitter in 5 or so chips. However, the
8089 IOP was a piece of cruft and one big pain to debug.
Allison
9 Dec
9 Dec
5:59 a.m.
On 12/8/2005 at 9:49 PM Allison wrote:
I think you're thinking of the NEC V20/ uPD70108 and the V30/uPD70116 CPUs.
Plug-compatible with the 8088 and 8086, but with a couple of instructions
added to switch modes into 8080 emulation. Overall, emulation was pretty
good, but should have been Z80 instead of 8080 and there were a couple of
small bugs in the emulation that weren't fixed.
Most vendors used the V20/V30, however, because they were low-power CMOS.
They also implemented some of the 80186 entensions, like ENTER and LEAVE
and PUSHA/POPA and multi-bit immediate count shifts like SHR AX,5. If
you've got an old 8088/8086 box and need to run some code compiled to use
the 186/286 extensions, the V20/V30 is a very neat drop-in solution.
Cheers,
Chuck
.
>
>Subject: Re: Legacy apps in Windows/OS X was Re: Old MS-DOS & Win
Software
From:
woodelf <bfranchuk at jetnet.ab.ca>
Date: Thu, 08 Dec 2005 19:30:35 -0700
To: General Discussion: On-Topic and Off-Topic Posts
<cctalk at
classiccmp.org>
What I do find odd that the 8086 came out first then a few years later
the 8088.
It still think the 8086/8086 could have had two instruction sets --
8080 emulation and
a better 16 bit design.
Not that I know of. I remember being able to get 8088s before the 86s
and that was back in 78 or early 79. 
6:28 p.m.
It was quite somethig to take an 8088, 8284(clock
gen), 8205 (aka74138),
8755, 8155 and have a complete 16bitter in 5 or so chips. However, the
IIRC there was a Circuit Cellar article in Byte for a 5-chip machine
using the 8088. The other chips were, IIRC, the clock generator, 8155
(RAM + I/O), 8355 (ROM + I/O) and an Intel RAM chip with multiplexed
address/data buses. The address decoding was done by just connecting the
chip enables to the high-order address lines...
8089 IOP was a piece of cruft and one big pain to
debug.
Well, I've read the datasbook, I've seen it in use, but have never
designed with it. What's the problem? It always struck me as a lot nicer
than the 8237 + page registers that IBM used in the PC
-tony
7:09 p.m.
On 12/8/2005 at 9:49 PM Allison wrote:
However, the 8089 IOP was a piece of cruft and one big pain to debug.
Expensive, short-sighted and brain-dead, but it did have a flat 20-bit
addressing space. The 186 with integrated periopherals and 20 bit DMA
pretty much put the kebosh on it.
Cheers,
Chuck
6972
days inactive
6972
days old
3 comments
3 participants
participants (3)
-
ajp166＠bellatlantic.net -
ard＠p850ug1.demon.co.uk -
cclist＠sydex.com