[Info-vax] Vaxes shutting off this week

[Tom Linden]

On Mon, 02 Mar 2009 00:04:57 -0800, Michael Kraemer <M.Kraemer at gsi.de>  
wrote:

> JF Mezei schrieb:
>>  The could have decided to do what they did to the 8086. Have a front  
>> end
>> decoder that generates RISC instructions from the complex CISC  
>> instructions.
>
> This was much easier on the x86 because this chip was so primitive
> that it was almost RISC (Load/Store etc) already.
> And the effort would pay off since intel could hope to recoup
> the money from millions of sales to the desktop.
> On VAX with its overly complex instruction set and addressing modes
> this would have been a nightmare.
> Look at the other extreme CISC archictecture of that time, the 68K.
> Although it sold in much higher numbers than VAX, Motorola
> decided not to "risc" the 68K. They proceeded to the 88K and PPC instead.
> The 68060 was the last hooray, it maxed out at 66MHz, not far from the
> VAX 4000 chip.

Actually, it wasn't that bad.  I wrote an instruction decoder for the  
NS32000
the instruction set of which was modelled on the VAX and IIRC there were 19
formats and I think you had to get to the third byte on few of them to
uniquely determine the instruction. but that was 27 years ago so I may
have forgotten a few things.  As for x86 I would have characterized it not
so much as aload store architecture, but rather accumulator like the Prime
V-mode series.

>
>>  Unfortunatly, at the time the decision was made to dump VAX in favour  
>> of
>> Alpha, they did not have the advantage of hindsight and
>> didn'T know that Intel would succeed in getting the 8086 to break so
>> many barriers.
>
> VAX wasn't dumped but lived on until the end of the decade.
> The most important barrier x86 succeeded to break was simply
> ROI, something a small company like DEC could never achieve
> by pimping up their aging VAX chip.
>
>> And lets not forget that at the time, there was Sun breathing down
>> Digital's neck with its own risc chips with Sun/Apolllo systes getting
>> better price performance than VAX. (that was a marketing issue with
>> prices for DEC gear still priced too high)
>
> The workstation makers of the 1980s (Sun, Apollo, SGI, HP)
> all started on the 68K, giving them almost VAX performance
> but at much lower price. Just in time, i.e. towards the
> end of that decade, they were about to switch to their own
> successful RISC chips (Sparc,Mips,PA), whereas DEC
> was goofing around with various projects that never came
> to fruition. DEC simply missed the boat here.
>
>> Had DEC lowered prices of VAX sufficiently, it could have competed
>> against Sun,
>
> I'd estimate in 1990 a VAXstation would have to be as cheap
> as a classical X-terminal to be price/performance competitive.
> I'm not sure whether this would have covered the costs to produce it.
>



-- 
PL/I for OpenVMS
www.kednos.com