[Info-vax] Three boot camp sessions on YouTube

[David Froble]

Johnny Billquist wrote:
> On 2016-10-12 13:10, Jan-Erik Soderholm wrote:
>> Den 2016-10-12 kl. 13:04, skrev Neil Rieck:
>>> For me, the big difference between Alpha and Itanium in those videos was
>>> the register count (32 vs. 128).  Having 128 registers may have seemed
>>> like a good idea when they were designing that chip, but we all know
>>> that an OS is going to require some/all those registers to be
>>> periodically saved then restored (well, not all on every interrupt). I
>>> wonder if Intel would have implemented 128...
>>
>> I'm not sure of the timing here, but wasn't the root of the IA64
>> arhitecture designed at HP before it was transfered to Intel?
> 
> That was one source, yes. Unless I remember wrong, it was a pooling of 
> ideas from HP and Intel.
> 
>     Johnny
> 

No.  It originally was a HP initiative.  The HP folks didn't think OoO could be 
gotten to work.  When it was proven to work, Alpha and Power, some HP people 
decided that perhaps they should look at OoO instead of their design.  HP 
management nixed that idea.  Then HP must have had problems, or, some management 
wasn't in favor of being a chip manufacturer, don't know, and pitched a joint 
effort with Intel.  There was also the transfer of the HP chip people to Intel.

As to why Intel decided to join in, there can be plenty of speculation.  At the 
time, x86 was considered a PC CPU.  Perhaps they wanted to have a "big time" 
CPU.  Later, at some time, Intel got the idea that if they could kill off all 
the competition, and get everything using the itanic, they would get a "nick" 
from every CPU sold, regardless of the mfg.  Intel really liked that idea. 
Thanks to AMD, that "corner the market" scheme failed.  But with few exceptions, 
that left us with x86.

Jan-Erik is correct, it originally was an HP design.