From: Andrew Swallow on
krw wrote:
> In article <DZSdnaHeS49TzpTbnZ2dnUVZ8tXinZ2d(a)bt.com>,
> am.swallow(a)btopenworld.com says...
>> krw wrote:
>>> In article <fqWdnV-JLsRJ_ZXbRVnyiAA(a)bt.com>,
>>> am.swallow(a)btopenworld.com says...
>>>> Morten Reistad wrote:
>>>> [snip]
>>>>
>>>>> DEC _did_ come back with the alpha, just as soon as they had managed
>>>>> to deVAXify their brains. Except, by then the trust in the company had
>>>>> evaporated.
>>>> The only sensible use for the Alpha was to run microcode as a VAX.
>>>> When chip manufacturing technology allowed CISC CPUs on a single chip
>>>> the cost advantages of RISC were over.
>>> I think you'll find there are a few people who will disagree with
>>> you.
>>>
>> Probably but were they customers of DEC?
>
> Every Alpha ran VAX microcode? Dunno, never seen a real live Alpha.
>

The Alpha was the replacement for the VAX, so a lot of the software
running on the Alpha was VAX/VMS software. The software was either
recompiled or run using a software emulators. So a 500 MHz Alpha
ran like a 50 MHz VAX with expensive ram.

> Never the less, what DEC's customers did or didn't do has little to
> do with the reasons RISC CPUs exist. You pretend that RISC wouldn't
> exist without tube logic. How many RISC microprocessors today vs.
> RISC. How does legacy software fit into that mix?
>
>> The ARM chip is still around in mobile phones but the Acorn desk top
>> is not.
>
> And this has to do with DEC, RISC, or Alpha exactly how? The 8051 is
> still around too, but...
>
The ARM is the worlds best selling RISC machine and a rival to the
Alpha.

Andrew Swallow
From: krw on
In article <Stydndqde4bOxJTbRVnygAA(a)bt.com>,
am.swallow(a)btopenworld.com says...
> David Kanter wrote:
> [snip]
>
> >
> > A much more reasonable, and possibly true, assertion would be that the
> > advantage of RISC architectures decreased over time. However, even as
> > late as the Pentium 1, there was a huge advantage for RISC
> > architectures. With the Pentium Pro that became less clear, although
> > RISCs still ruled the roost for FP heavy applications.
>
> We can certainly have a nice debate as to whether anything containing
> floating point hardware is RISC.

What does the FPU matter to the price of oats in China?

--
Keith
From: Anne & Lynn Wheeler on

re:
http://www.garlic.com/~lynn/2007f.html#79 The Perfect Computer - 36 bits?

while '60s era 2702 & 2703 got supplanted by 3705s (and various
clones) ... the '60s era 2701 (with support for T1) held on well into
the 80s. Initially for gov. market, they did do Zirpel card for
Series/1 to handle T1 in the mid-80s (since the 2701 T1s were getting
quite long in the tooth). There was quite a large number of Series/1
deployed for numerous communication tasks ... and quite a few
applications were starting to bump up against Series/1 16bit address
infrastructure.

the clones, 3705, base Series/1 were handling up to 56kbit lines
relatively straight forward ... but there wasn't a lot of support for
T1 and higher speeds.

For HSDT (high-speed data transport) project starting in the early-80s
were using HYPERchannel for T1 (and higher speeds) ... but then got
into some of custom built stuff
http://www.garlic.com/~lynn/subnetwork.html#hsdt

for high-speed internal network backbone ... misc. past posts on
internal network
http://www.garlic.com/~lynn/subnetwork.htm#internalnet

one of the drivers starting to look at doing some custom made stuff
was encryption ... all lines that left corporate premise required
encryption ... and there wasn't a whole lot of choice at T1 and above
.... price, features, etc. recent post mentioning slightly orthogonal
http://www.garlic.com/~lynn/2007f.html#73 Is computer history taught now?

i guess we also somewhat corrupted the stuff leading up to NSFNET RFP
.... misc. old email from the period on the subject
http://www.garlic.com/~lynn/lhwemail.html#nsfnet

then we got prevented from bidding for NSFNET backbone (even tho NSF
audit of what we had running said it was at least five yrs ahead of
all bid submissions). although the backbone bid called for "T1"
.... the winning bid actually implemented 440kbit links ... with telco
multiplexor mapping multiple links into T1 trunks .... which i suppose
was what satisified the letter of the RFP (however, if you were to use
trunks as criteria ... you could probably have found that even a lot
of 56kbit links somewhere in the end-to-end process passed thru a T5
or possibly even higher speed trunk).

for other drift ... one of the telcos had implemented peer-to-peer
networking using Series/1 (somewhat like IMPs from the arpanet era)
.... but they also had done a (SNA) PU4/NCP emulation (as well as
SSCP/PU5 emulation ... telling the mainframe SSCP that the devices
were cross-domain "owned" by some other processor) ... carrying RUs
thru real peer-to-peer infrastructure. I tried to get it out as a
product with a port to RIOS (801/risc chips used in rs/6000) to
alleviate the 16bit limitations of the Series/1 ... some old posts
http://www.garlic.com/~lynn/99.html#66 System/1 ?
http://www.garlic.com/~lynn/99.html#67 System/1 ?
http://www.garlic.com/~lynn/99.html#70 Series/1 as NCP (was: System/1 ?)
From: Nick Maclaren on

In article <DZSdnaHeS49TzpTbnZ2dnUVZ8tXinZ2d(a)bt.com>,
Andrew Swallow <am.swallow(a)btopenworld.com> writes:
|> krw wrote:
|> > In article <fqWdnV-JLsRJ_ZXbRVnyiAA(a)bt.com>,
|> > am.swallow(a)btopenworld.com says...
|> >> Morten Reistad wrote:
|> >>
|> >> The only sensible use for the Alpha was to run microcode as a VAX.
|> >> When chip manufacturing technology allowed CISC CPUs on a single chip
|> >> the cost advantages of RISC were over.
|> >
|> > I think you'll find there are a few people who will disagree with
|> > you.
|> >
|> Probably but were they customers of DEC?

Yes.


Regards,
Nick Maclaren.
From: Stan Barr on
On Tue, 27 Mar 2007 18:26:39 +0100, Andrew Swallow <am.swallow(a)btopenworld.com>
wrote:
>
>The ARM chip is still around in mobile phones but the Acorn desk top
>is not.

There are still ARM (actually Strong-ARM) computers around, but they usually
run Linux these days.

--
Cheers,
Stan Barr stanb .at. dial .dot. pipex .dot. com
(Remove any digits from the addresses when mailing me.)

The future was never like this!