> I've always wanted to ask this question. How is a microprocessor made?
> Can someone here on the list answer this question?
Take a hand full of sand, compress it, supply voltage
and try to boot - if it doesn't work qout, change the
sand - try until a result is reached.
Gruss
H.
--
Ich denke, also bin ich, also gut
HRK
>> The original machine, no contest!. An emulator can't possibly include the
>> wonderful hot-component smells, hardware problems, being able to stick a
>> logic probe anywhere you like, the joys of fixing the darn thing, etc.
> SO how long is it before some really enthusiastic hacker creates a 3D
> emulator using a graphics engine like that in DOOM, that allows you to pop
> open the cover and go INSIDE the machine, home in on specific ICs,
> resistors, capacitors, etc. and stick a probe or voltmeter up to the
> leads? [...] Later on, when they invent the Scent
> Printer (a device which emits the smells of the computer you are working
> on) you'll be able to know when you've shorted something because all of a
> sudden a foul, toxic cloud will be emitted.
> I don't think this is so much of a fantasy. Its a lot of work, but it
> could be done (ok, save for the Scent Printer part).
Salam, just forget it - i will never klick on an
Open-The-Cover button (or point on wit a clumbsy
glove) when I can feel the snap of an Apple ][+
case by my own.
Gruss
H.
--
Ich denke, also bin ich, also gut
HRK
< I've always wanted to ask this question. How is a microprocessor made?
< Can someone here on the list answer this question?
Very carefully. Actually that is not a joke as it is a very precise
process.
The process is generally the same as making any integrated circuit save
for microprocessors are more complex.
The basic idea is that a piece of polished pure crystaline silicon has
areas photographically marked and either etched away or doped (infused)
with other chemicals (like boron, arsenic, others) to alter the local
properties of the silicon to form transistors where and as needed.
Other steps put metalization like printed circuit board traces to
interconnect active areas on the chip. Once complete they are tested
seperated and mounted in holders.
All of this is done on a scale of size that is in the range of 1/100000th
of an inch and smaller. So the "very carefully" comment is valid and
speaks to the level of accuracy and cleanliness needed.
That is a pretty trivial description but a detailed one could fill volumes
quickly.
Allison
>>Irrelevant to the discussion for two reasons:
>>1) No one has made mainframes out of discrete transistors since the late 60s.
>>2) Almost all modern mainframes are based on monolithic microprocessors.
> My point was that only recently has it become so that there is nothing
> more than an Alpha or an PII. There are many parallel implementations,
> but nothing like the difference between an Altair and an S/360. Yes, I
> know there are many computers way more powerful than CompUSA desktops,
> but the point is that they're just parallel implementations and there
> isn't so much of a class distinction.
Do you want to tell us that the mainframe world
didn't evolve ? Boy, they also have new and quite
speedy new processors - for example, the new SIEMENS
S150 delivers around 93 RPF on a single CPU and
a sustained rate of 760 RPF on a 12 CPU configuration.
RPF (Relative Performance Factor) is a SIEMENS unit
to classify the relative speed of their processors.
The baseunit is a processor from the beginnign of the
80's. Even in the main frame (/390 compatible) world
it's hard to compare processors just by speed, since
the manufacturers use different architectures below.
So every one has his own measurement. IBM uses their
MIPS (misleading informations on processor speed) as
a guideline - but even these are just a definde rate:
they just fixed the 9672-R15 as base with 63 MIPS.
A rough conversation between IBM MIPS and SIEMENS RPF
is 1.3 to 1.5 - so die S150 deliveres between 140 and
1000 IBM MIPS sustained load at average mix.
With the small SR2000 systems which are based on
MIPS (now thats the processor manufacturer :) R10000
CPUs there is a guideline for transformation: one
R10000 deliveres around 12 RPF or 15 MIPS at user level.
(if you know take the MIPS R10000 MIPS rating you'll
get also a fairly nice RISC to (IBM)CISC ratio, since
the firmware interpreter used in the SR2000 is very
performant)
Now if you take the SpecInt value of the R10000 and
of the Alpha or Pentium chips (sorry, I just not
have them ready), one could base a comparisation on
these values. If a Pentium deliveres about 10 times
the performance of a R10000, they might reach te region
of an actual mainframe (not any 60's transistor claim).
But in fact these numbers are again way beside the
actual usability - Mainframe performance is measured
by thruput and things like transactions per second
within a defind response time rather than raw CPU.
Just take a look at http://www9.s390.ibm.com/lspr/
to get an idea how specific measurment in the mainframe
world is don - not just instructions or megabytes
per second.
Gruss
H.
--
Ich denke, also bin ich, also gut
HRK
Every so often, people here mention how they wish they had TOPS-20
on their modern computers, how much superior it was to UNIX, etc.
COuld someone please explain the specific features (apparent to a
user) that are missing in UNIX? If I understand correctly, TOPS takes
a less minimalist attitude than UNIX, but since I've never even seen
it run, I wouldn't know. If it's not too much trouble, could someone
give some examples of interaction that demonstrates the difference?
______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com
>The whole reason Cray's are wired up as circles rather than linearly like a
>mainframe is to reduce wire lengths (and with them speed of light delays).
Although I never looked at the prints for one, all of the Cray 1's that I
had a chance to examine were not completely circular--there was always a
wedge missing.
I expect this was almost essential; all of the wiring between circuit
cards was on the inside of the circle, and the customer engineers had to be
able to get to that wiring.
I suspect that the reason Seymour Cray built the machine in the shape he
did was that the circuit cards, plus the machined columns that supported
and cooled them, were wider than the card connectors. By arranging the
card columns in a semi-circle with the connectors on the inside, he could
minimize wiring length.
----
John Dykstra jdykstra(a)nortel.com
Principal Software Architect voice: +1 651 415-1604
Nortel (Northern Telecom) fax: +1 612 932-8549
What amazes me is that no one has ever marketed a mainframe-like
machine out of modern processors. Someone mentioned they had made
one out of a 286 and some z80s, but why did no single company ever
sell any? I would think such machines could be very useful. OTOH, it
makes more money to sell 1000 machines than 3 machines and 1000 dumb
terminals...sometimes I wish capitalism was never invented...
>The raw CPU performance of thet 386 exceeds that of many (but not all)
of the
>discrete-transistor mainframes (and even IBM's SLT-based mainframes).
>
>However, system-level performance is another matter. How many 386
machines
>had intelligent I/O channels, *AND* took full advantage of them?
>
______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com
< From: William Donzelli <william(a)ans.net>
<
< Sure, a Dell Pentium machine might be plenty fast by the book, but simpl
< put, it could not even come close to sustaining that speed (140 Mflops f
< the Cray-1). A superfast CPU will always loose to slow I/O.
<
< Its all in the architecture...
yes, steaming along at 140MFLOPS only to slam into the IO bus that if ISA
*may* be able to sustain 16mb/s. PCI may be faster, I'd suspect in the
30+mb/s range (unknown to me). Anyhow being able to gind lots of numbers
is limited by memory bandwidth and maybe disk IO bandwidth and
PCs are not notably fast in that depatment. Cray and those other
machines were possibly a bit slower but they didn't have that I/O
bottlenecks.
Allison
Not to mention power problems! But I have the feeling that's a
bit small! Nothing on the scale of the Pentium II has ever been
achieved with transistors, never mind relays, and I'm sure they
tried. For example, it would be possible to place transistors at
1 per cubic inch, and have the cool adequately, but it's clearly
never been attempted, because if it were, there would be boxes
to plug into a PC XT to get Pentium II speeds.
>Assuming each relay occupied a cubic inch, one million relays would
>occupy a cube 100 inches on each side, or just a bit over 8 feet on
each side.
>Not ridiculously large.
>
>Of course, real relays much smaller than a cubic inch are readily
available,
>and a real computer of that density would have incredible cooling
>problems.
>
>Tim.
>
______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com
