[COFF] In memoriam: Gene Amdahl

[Dave Horsfall]

On Mon, 11 Nov 2019, Clem Cole wrote:

> According to my friend Russ Robulen (his coworker, lead on the 360/50, 
> worked on /91 and later lead for the IBM ASC),  Amdahl wanted the byte 
> to be 7-bits for S/360, but Fred Brook's overruled him.  Brooks was said 
> to have thrown Amdahl out his office and told him "not come back unless 
> it was a power of 2", as "he could not program it sanely otherwise."  
>  Amdahl semi-won the 24/32 bit war.   Brooks let he have a 24 bit basic 
> word, only if it stored it as 32 bits and ensured that all pointers were 
> stored in the same.  Russ says that Amdahl always thought both choices 
> were a terrible waste of hardware.   Gordon Bell later said, those two 
> choices were the most important in S/360's lasting impact.

Interesting; I'll try and summarise that for my calendar.  In the meantime 
I'm glad that Brooks' view prevailed, having worked with byte-less 12-bit 
(PDP-8) and 60-bit (CDC); I don't remember the word length of the 
Burroughs series.

-- Dave