Forgot your password?
cyhist Apr. 8 1998 e
Document Actions
========================================================================= Date: Wed, 8 Apr 1998 14:42:52 -0700
Reply-To: "CYHIST Community Memory: Discussion list on the History of
Cyberspace" <CYHIST@MAELSTROM.STJOHNS.EDU> Sender: "CYHIST Community Memory: Discussion list on the History of
Cyberspace" <CYHIST@MAELSTROM.STJOHNS.EDU> From: Stan Mazor <stan.mazor@BEASYS.COM>
Subject: 8085 history--design teams
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
______________________________________________________________________
Community Memory: Discussion List on the History of Cyberspace ______________________________________________________________________
8085 instruction set architecture and design teams 1974 8080 background
I worked on 8080 chip specification at Intel and specified all 256 op-codes in the instruction set. 13th from the bottom of my list was a complicated 1-byte instruction XTHL which executes 4 memory data cycles (2 memory reads and writes) to exchange the contents of the on-chip HL registers with the memory TOS. Shima designed the 8080 chip under Faggin, but when I explained this instruction to him, I recall he said "No More", and the subsequent 12 instructions were eliminated by him; so the 8080 had 12 unused operation codes.
Z-80 background
Faggin and Shima left Intel to found Zilog and their Z-80 chip was very competitive. I had proposed a feature based upon 8008 feedback which didn't make it into the 8080 (but was in the Z-80), which was to "overload" the Parity flag to be overflow for add/subtract. Additionally, the Z-80 had memory refresh and other novel improvements.
1976 8085 CPU chip (2 new instructions) and committee design Intel's response was the 8085 CPU chip which had some peripheral logic functions integrated to reduce the system cost and added 12 new instructions to the instruction set. A small group from our software group worked about 8 months on defining the "perfect" set to augment the 8080; I recall at least 2 of the team had PHd's in computer science. All of the new instructions were implemented on the chip, but as the chip got close to announcement, Intel started
work on the upwardly compatible 8086; any new instructions would be a burden!! The compromise was to announce only 2 (RIM/SIM?) of the 12; the other 10 instructions were not announced or in the specification. One obscure of these enabled an interrupt on overflow; another was a branch on greaterthan (new flag).
AMD copied the 8085 and I wondered if they thought our plan was to surprise them after AMD announced their product, to wit: "Intel has 10 more instructions". However, I understood that AMD studied the chip layout and "discovered" these 10, and at Intel I was worried that AMD would announce "New and improved 8085".
(I would like to here from an AMD person first hand about this ). I think that the remaining 10 instructions were never announced by Intel or AMD; a remarkable committee achievement.
8085 in the literature
In Dr. Dobbs journal, a German programmer, wrote an article in which he describes some useful functions of the "unused" 8085 op codes. My recollections are that he got them all except didn't quite grasp the "greater" than flag.
thx. stan
------------------------------------------------- Stanley Mazor 408-542-4120 phone
Training Department 408-542-4110 fax
BEA Systems smazor@beasys.com
385 Moffett Park Dr. Enterprise Middleware
Sunnyvale, Ca. 94089-1208
______________________________________________________________________
Reply-To: "CYHIST Community Memory: Discussion list on the History of
Cyberspace" <CYHIST@MAELSTROM.STJOHNS.EDU> Sender: "CYHIST Community Memory: Discussion list on the History of
Cyberspace" <CYHIST@MAELSTROM.STJOHNS.EDU> From: Stan Mazor <stan.mazor@BEASYS.COM>
Subject: 8085 history--design teams
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
______________________________________________________________________
Community Memory: Discussion List on the History of Cyberspace ______________________________________________________________________
8085 instruction set architecture and design teams 1974 8080 background
I worked on 8080 chip specification at Intel and specified all 256 op-codes in the instruction set. 13th from the bottom of my list was a complicated 1-byte instruction XTHL which executes 4 memory data cycles (2 memory reads and writes) to exchange the contents of the on-chip HL registers with the memory TOS. Shima designed the 8080 chip under Faggin, but when I explained this instruction to him, I recall he said "No More", and the subsequent 12 instructions were eliminated by him; so the 8080 had 12 unused operation codes.
Z-80 background
Faggin and Shima left Intel to found Zilog and their Z-80 chip was very competitive. I had proposed a feature based upon 8008 feedback which didn't make it into the 8080 (but was in the Z-80), which was to "overload" the Parity flag to be overflow for add/subtract. Additionally, the Z-80 had memory refresh and other novel improvements.
1976 8085 CPU chip (2 new instructions) and committee design Intel's response was the 8085 CPU chip which had some peripheral logic functions integrated to reduce the system cost and added 12 new instructions to the instruction set. A small group from our software group worked about 8 months on defining the "perfect" set to augment the 8080; I recall at least 2 of the team had PHd's in computer science. All of the new instructions were implemented on the chip, but as the chip got close to announcement, Intel started
work on the upwardly compatible 8086; any new instructions would be a burden!! The compromise was to announce only 2 (RIM/SIM?) of the 12; the other 10 instructions were not announced or in the specification. One obscure of these enabled an interrupt on overflow; another was a branch on greaterthan (new flag).
AMD copied the 8085 and I wondered if they thought our plan was to surprise them after AMD announced their product, to wit: "Intel has 10 more instructions". However, I understood that AMD studied the chip layout and "discovered" these 10, and at Intel I was worried that AMD would announce "New and improved 8085".
(I would like to here from an AMD person first hand about this ). I think that the remaining 10 instructions were never announced by Intel or AMD; a remarkable committee achievement.
8085 in the literature
In Dr. Dobbs journal, a German programmer, wrote an article in which he describes some useful functions of the "unused" 8085 op codes. My recollections are that he got them all except didn't quite grasp the "greater" than flag.
thx. stan
------------------------------------------------- Stanley Mazor 408-542-4120 phone
Training Department 408-542-4110 fax
BEA Systems smazor@beasys.com
385 Moffett Park Dr. Enterprise Middleware
Sunnyvale, Ca. 94089-1208
______________________________________________________________________
