According to Stephen Fuld <
sfuld@alumni.cmu.edu.invalid>:
On 6/20/2025 11:31 AM, Thomas Koenig wrote:
See the message I sent yesterday. They knew about dynamic relocation and virtual
memory, but considered it too risky to add to an already ambitious project. >>
Brooks himself wrote he considered not adding virtual memory to the /360
a mistake, so...
Where did he say it? I have Mythical Man Month, not seeing it there. IBM clearly
agreed in some sense that omitting VM was a mistake since they added it to S/370
pretty soon, but that was after they knew that S/360 was a success and that it would work at an affordable price. I'm sure the semi-accidental success of the /67 didn't hurt--TSS failed but CP/67 was great and created the way that most computers bigger than laptops work now.
IBM had good peripherals, they had a good upgrade path to very
powerful machines, and they were bit-compatible for user programs
(plus, they put in the microcode emulation of the 1401 so their
customers could transition smoothly - that was a genius move,
the /360 could probably would have been far less of a success
if that had not been possible). All of these were good reasons
to buy these machines.
And the larger machines had emulation of the 7080.
They provided emulation of all of their second generation machines, 1401/1460/1440 and 1620 at the low end up to 7070/7074 on the /50, and 7080 and 709X on the /65 and 709X on the /85. It was implemented in a combination of microcode and system software. The emulated computers all ran faster than the originals and I suspect that in many cases the rental of the new 360 was less than the originals because they were easier to maintain.
According to Pugh et al., they'd originally looked at software simulation but that was way too slow, like 40x slower than the real machine. But then a few engineers tried doing it in microcode and it worked great. In a few cases
they added microcode features to make emulation easier, like handling the 7070's decimal addressing.
If that hadn't worked the pressure to produce a successor to the 1401 would have been hard to resist and would have put the whole 360 program in jeopardy. --
Regards,
John Levine,
johnl@taugh.com, Primary Perpetrator of "The Internet for Dummies",
Please consider the environment before reading this e-mail.
https://jl.ly
--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)