[Info-vax] VAX 6000 on Young Sheldon

[Dave Froble]

On 2/7/2023 4:17 PM, Arne Vajhøj wrote:
> On 2/7/2023 10:04 AM, Johnny Billquist wrote:
>> On 2023-02-07 01:52, Arne Vajhøj wrote:
>>> On 2/6/2023 8:04 AM, Neil Rieck wrote:
>>>> VAX-6000 was the first time I encountered SMP (symmetric multi
>>>> processing) on VAX. We were running a 6430 which meant 3 CPUs
>>>>
>>>> Before that I worked for a year on a pair of VAX-8550 configured as a
>>>> VAXcluster
>>>>
>>>> Our VAX-6430 system felt faster than the pair of clustered VAX-8550
>>>
>>> Based on VUPS ratings then I would expect:
>>> - multi job throughput to be the same
>>> - single job execution slower
>>> - interactive responsiveness better
>>
>> Uh? The 64x0 single CPU VUPS rating is 7.0, while the 8550 is rated at 6.0
>> VUPS, so I would expect the 6430 to be faster for individual jobs, as well as
>> *way* better at handling load, since you actually have three CPUs compared to
>> just one on the 8550. Clustering two 8550 would not get close to a 3 cpu 64x0.
>
> Ooops.
>
> You are correct.
>
> I looked at the 63x0 instead of the 64x0 number.
>
> The 6430 is faster in every way than 2 x 8550.
>
> Mea culpa.

Reasonable.

I recall that the 8500s, 8600s, and such were still multiple board CPUs.  Then 
came the C-VAX and N-VAX micro-processors.  The micro-processors were much 
better, faster, and the later N-VAX CPUs were rather good, for a VAX.

My memory (yeah, right) is that the N-VAX was a RISC processor that emulated the 
VAX instruction set.  Or something like that.

-- 
David Froble                       Tel: 724-529-0450
Dave Froble Enterprises, Inc.      E-Mail: davef at tsoft-inc.com
DFE Ultralights, Inc.
170 Grimplin Road
Vanderbilt, PA  15486