4 Nov
2008
4 Nov
'08
1:50 p.m.
On 4 Nov 2008 at 5:51, Robert Borsuk wrote:
Don't know if you guys found this but I did find
one computer
reference.
http://www.old-computers.com/museum/computer.asp?c=196
and the data sheet on bit savers for the chips(thank you Al).
1983? I was trying to get the attention of the Fairchild sales
department around 1978. You'd have thought with DG not selling the
MicroNova CPU to outside customers and the 9440 being a Nova-like
chip that Fairchild could have cleaned up. There weren't many other
16-bit chips at the time (e.g. NS PACE, GI CP1600, TI TMS9900). But
Fairchild seemed not to be interested in selling the stuff.
My understanding is that the 9440 is an "almost" Nova in that it
lacks certain features needed to run Nova software. Can anyone
verify this?
Cheers,
Chuck
4 Nov
4 Nov
4:49 p.m.
Chuck Guzis wrote:
You'd have thought with DG not selling the
MicroNova CPU to outside customers
They tried to sell it, they just weren't good at it. They didn't want
small customers, though they did send me the data sheets and such. They
wanted big customers, but they didn't want big customers that would
produce general-purpose computers that would compete with low-end Nova
minis.
I suppose they just wanted embedded designs, but everyone was using the
8080, 6800, etc. for those. For most embedded systems there was no
obvious benefit to using a slow, expensive MicroNova instead.
DEC wasn't much better at selling their microprocessors either. They
got few outside design wins for the T11 and J11 processors, and the
pricing wasn't competitive. Same thing years later with the Alpha. The
marketing people bragged about how it outperformed the x86 and that the
whole world would move to Alpha, but the price/performance ratio was
substantially worse than that of the x86, so the only significant
external design wins were for specialized systems that needed the
highest possible uniprocessor performance.
5:29 p.m.
On 4 Nov 2008 at 13:49, Eric Smith wrote:
They tried to sell it, they just weren't good at
it. They didn't want
small customers, though they did send me the data sheets and such. They
wanted big customers, but they didn't want big customers that would
produce general-purpose computers that would compete with low-end Nova
minis.
Contrast that with the sales approach by Intel, who would sell to you
if they could hear the sound of change jingling in your pocket. I
could count on regular phone calls, literature and free lunches by
the local Intel sales guy--and we were definitely small potatoes. I
think Intel understood the idea of an architectural "lock-in" better
than DG or DEC, who apparently didn't think that principle applied to
microprocessors.
That Intel support meant a lot--you could get samples of just about
anything and expect an applications engineer to return your calls.
It wasn't that Intel had the best products out--it was that they made
you feel important and did their best not to let you fail. OTOH,
National, with some very good product offerings generally wasn't
trusted for much more than commodity TTL and linear, unless it was as
a second source for someone else's products that you were already
using.
Just goes to show that "if you build it, they will come" doesn't
always hold true.
Cheers,
Chuck
5:40 p.m.
Contrast that with the sales approach by Intel, who
would sell to you
if they could hear the sound of change jingling in your pocket. I
could count on regular phone calls, literature and free lunches by
the local Intel sales guy--and we were definitely small potatoes. I
think Intel understood the idea of an architectural "lock-in" better
than DG or DEC, who apparently didn't think that principle applied to
microprocessors.
I don't know about DG, but I think that DEC had more basic issues than lack of
understanding of "lock-in". Come to think of it, they did understand that well
enough -- for computers, not chips.
DEC had microprocessors all right, but they were really just a way to build computers.
(The T11 is perhaps an exception; I sometimes wonder how that came about.) So they made
plenty of use of them internally, but the notion of selling them to outsiders just
didn't compute.
Another consideration is that DEC didn't take sales seriously in any setting --
witness the fact that they had the only sales force in recorded history that wasn't
paid commission.
As for the lack of success of Alpha in competing with x86, the above are good reasons, and
another one would be the fact that by then it was too late for a new general purpose
architecture. Now if Alpha had been positioned as an embedded architecture, it might have
succeeded (taking from MIPS) -- except that it probably was too fast for most embedded
applications.
paul
5744
days inactive
5744
days old
4 comments
4 participants
participants (4)
-
bfranchuk＠jetnet.ab.ca -
cclist＠sydex.com -
eric＠brouhaha.com -
Paul_Koning＠Dell.com