9 Mar
2007
9 Mar
'07
6:52 p.m.
I bought this from dvdplanet the other day (good deal: $15) and
watched it recently.
They seemed to skip the entire minicomputer/workstation phase and
jumped straight from mainframes to the personal computer.
What was interesting though is the collection of footage that they
used throughout. I spotted what looked like footage of an HP terminal
factory, manufacturing 2621's. On the shelves in the background were
2648 style cabinets. It was a brief clip, but I felt "cool" knowing
that I was one of the few people who would recognize the equipment!
--
"The Direct3D Graphics Pipeline" -- DirectX 9 draft available for download
<http://www.xmission.com/~legalize/book/download/index.html>
Legalize Adulthood! <http://blogs.xmission.com/legalize/>
10 Mar
10 Mar
10:03 a.m.
New subject: Modern Marvels: Computers ?? no graphics supers
Richard,
You are right, they skipped the whole branch of computer evolution where the
real scientific computing was done.
The computer as an IC and electronics today would not exist without the
bootstrap of mainframe based circuit design onto 6800/MIPS.
Without Sun, HP, Apollo the next gen CPU ICs could not be placed and routed.
Engineers needed to interact in realtime with the design at a graphics
workstation to get the job done, to create the IC masks. You could not lay
it out in rubylith tape any more. The problem was not the design or
architecture, it was interconnect complexity. The autorouters could not
connect the logic. (lee's algorithm, later Cooper and Chyan, gridless)
Lost, sadly was the machine between then and now, the Graphics
Supercomputer. In an effort to add computational speed to graphics and
scientific visualazation, two vendors went head to head on this problem,
Ardent and Stellar.
If you were around at the time, and saw one of these I would love to hear
from you. The performance was truly spectacular. I
had a chance to use one
for a couple of years, and it still comes pretty close to
current GPU tec in
graphics performance. With pipeline vector processor and compiler to unroll
loops it was WOW. Todays Ghz processors cannot beat a vector machine in
computation, Titan had a 16 Mhz 1K floating point vector ALU.
I am currently working with Ardent/Stellar/Stardent Dore' visualization
library from these machines. It compiles clean from the freebsd ports
collection, and the examples work in xfree86.
I have a few Sillicon Graphics Iris and Indys too.
Are there any graphics guys on the list?
Randy Dawson
From: Richard <legalize at xmission.com>
Reply-To: "General Discussion: On-Topic Posts Only" <cctech at
classiccmp.org>
To: cctalk at classiccmp.org
Subject: Modern Marvels: Computers
Date: Fri, 09 Mar 2007 19:52:39 -0700
I bought this from dvdplanet the other day (good deal: $15) and
watched it recently.
They seemed to skip the entire minicomputer/workstation phase and
jumped straight from mainframes to the personal computer.
What was interesting though is the collection of footage that they
used throughout. I spotted what looked like footage of an HP terminal
factory, manufacturing 2621's. On the shelves in the background were
2648 style cabinets. It was a brief clip, but I felt "cool" knowing
that I was one of the few people who would recognize the equipment!
--
"The Direct3D Graphics Pipeline" -- DirectX 9 draft available for download
<http://www.xmission.com/~legalize/book/download/index.html>
Legalize Adulthood! <http://blogs.xmission.com/legalize/>
_________________________________________________________________
With tax season right around the corner, make sure to follow these few
simple tips.
http://articles.moneycentral.msn.com/Taxes/PreparationTips/PreparationTips.…
10:10 a.m.
New subject: Modern Marvels: Computers ?? no graphics supers
Yes, I'm a graphics guy. I remember that Stardent; it had an
interesting "Z blt" whereby it would blt an image and an associated Z
buffer chunk that would be put through the Z buffer compare operation.
The result was that you could render a high quality phong sphere to
external memory and then Z blt it all over the place for high quality
molecular modelling. Pretty cool, although I don't know what other
applications could have used the feature beyond molecular modelling.
--
"The Direct3D Graphics Pipeline" -- DirectX 9 draft available for download
<http://www.xmission.com/~legalize/book/download/index.html>
Legalize Adulthood! <http://blogs.xmission.com/legalize/>
11:44 a.m.
New subject: Modern Marvels: Computers ?? no graphics supers
Randy Dawson wrote:
...
Lost, sadly was the machine between then and now, the
Graphics
Supercomputer. In an effort to add computational speed to graphics and
scientific visualazation, two vendors went head to head on this problem,
Ardent and Stellar.
If you were around at the time, and saw one of these I would love to
hear from you. The performance was truly spectacular. I had a chance
to use one for a couple of years, and it still comes pretty close to
current GPU tec in graphics performance. With pipeline vector processor
and compiler to unroll loops it was WOW. Todays Ghz processors cannot
beat a vector machine in computation, Titan had a 16 Mhz 1K floating
point vector ALU.
Randy, no doubt you know a lot more about the ardent/stellar/stardent
stuff than me. I was aware of it and I once got hold of the design spec
for the TOE processor (the 4x4 pixel "stamper"). However, I think you
aren't aware of how sophisticated todays GPUs are.
16 MHz * 1K flops = 16 Gflops. A single top end GPU is more like 500
GFLOPS (single prec only, though). Today's GPUs have myriad pixel
formats, including ARGB with an FP32 for each component. Pixel shaders
are highly programmable. A single GPU can have > 80 GB/sec of bandwidth
to DRAM (not cache).
The TOE processor was a fixed point affair with limited, fixed point
precision. There is no comparison. I wish I still had the spec to make
a more concrete comparison.
A google search turned up this quote:
With the Dore' rendering package [Borden89], each processor is capable
of rendering a maximum of 20,000 smoothly shaded small polygons/seconds.
Today's GPUs can render thousands of times more triangle per second,
antialiased, with multiple, high quality texture maps and arbitrary
blending.
Another google search
http://www.ece.cmu.edu/~ece548/handouts/17v_perf.pdf
says that the Titan 1 had a 125 ns clock period and two FPUs, for 16
MFLOP/s peak. Perhaps you recall 1K FPUs, but maybe it was a 1K vector
register length. The same pdf (written by Philip Koopman) says that
even with four processor, and with a large (1000x1000) array size, the
titan-1 peaked at 15.7 Mflops. It attributes this to the fact that the
aggregate bus bandwidth of the titan was 256 MB/sec. By rewriting the
linpack code to block the data appropriately, they got it up to 46 MFLOP/s.
So, overall, I think there is no comparison. The rose colored glasses
of time have fooled you.
...
Are there any graphics guys on the list?
Yes, from the hw end of things.
12:49 p.m.
New subject: Modern Marvels: Computers ?? no graphics supers
Jim Battle wrote:
Today's GPUs can render thousands of times more triangle per second,
antialiased, with multiple, high quality texture maps and arbitrary
blending.
mmm. another engineer and I made a triangle blt machine in 1983/1984
from raw ttl. it followed a display list. I think we
were in the
"100's" of triangles at the time.
wish we'd patented that :-) it made from some fun coin-op game prototypes but
never got out the door...
-brad
12:58 p.m.
New subject: Modern Marvels: Computers ?? no graphics supers
In article <200703102049.l2AKnYHJ004478 at mwave.heeltoe.com>,
Brad Parker <brad at heeltoe.com> writes:
mmm. another engineer and I made a triangle blt
machine in 1983/1984
from raw ttl. it followed a display list. I think we were in the
"100's" of triangles at the time.
wish we'd patented that :-) it made from some fun coin-op game prototypes bu
t
never got out the door...
You probably would have run into patents by Evans & Sutherland...
--
"The Direct3D Graphics Pipeline" -- DirectX 9 draft available for download
<http://www.xmission.com/~legalize/book/download/index.html>
Legalize Adulthood! <http://blogs.xmission.com/legalize/>
1:06 p.m.
New subject: Modern Marvels: Computers ?? no graphics supers
Jim,
I totally agree with you, todays GPUs are screamers. Did you know that the
core of NVIDIA's design team came from SGI?
Sad to see that company went wintel... I think they were almost de-listed
from the stock excchage at one point, as the stock
dipped below a dollar.
Jim, have you followed the recent threads in the news, to use the GPU
processor fabric as a general purpose engine, not just for pixels? If you
have IEEE access, search GPU.
Due to the PC gamers, we really do have a supercomputer on a card for 40
bucks or less. If anyone else is exploring this topic, please reply or
email me rdawson16 at hotmail dot com.
I think the strenght of the Stardent system was a combination of things from
Gordon More and Seymour Cray, their rules for supercomputer design. The
memory bandwidth was awsome for the time, and with 4 processors and a vector
unit accessing it independently, crunching code from a compiler that knew
how to vectorize and divide tasks.
I had a chance to represent another company (I was a graphics consultant to
NASA) Superset- they had a cool idea too, to process Fortran in near native
instruction of the machine. They used bit slice 2901s from AMD to create a
A (operation) B = C machine, and a compiler to generate this code. So the
machine was sort of a hardware interpreter. If you look at todays DSPs they
are very similar, they can read an operand, write an operand and perform a
computation in a single cycle.
I have Stardent (Kubota) Dore' running today and porting my old app to it.
I wrote the PC roller coaster simulator, COASTER. I want to re-spin it, now
that fast GPU hardware has finaly arrived and in every PC.
My rose colored glasses recall all of these machines booting to prompt and a
screen in seconds. They always ran for months without a crash, and could
read/write 500 mbyte images/data files in seconds too. The PC aint there
yet...
I write this from FreeBSD, I took all the 'Gates' out of my computer, and
guess what, it still works.
Thanks for your coments, Jim.
Randy
From: Jim Battle <frustum at pacbell.net>
Reply-To: "General Discussion: On-Topic and Off-Topic
Posts"<cctalk at classiccmp.org>
To: General Discussion: On-Topic and Off-Topic Posts
<cctalk at classiccmp.org>
Subject: Re: Modern Marvels: Computers ?? no graphics supers
Date: Sat, 10 Mar 2007 13:44:55 -0600
Randy Dawson wrote:
...
_________________________________________________________________
The average US Credit Score is 675. The cost to see yours: $0 by Experian.
http://www.freecreditreport.com/pm/default.aspx?sc=660600&bcd=EMAILFOOT…
Lost, sadly was the machine between then and now,
the Graphics
Supercomputer. In an effort to add computational speed to graphics and
scientific visualazation, two vendors went head to head on this problem,
Ardent and Stellar.
If you were around at the time, and saw one of these I would love to hear
from you. The performance was truly spectacular. I had a chance to use
one for a couple of years, and it still comes pretty close to current GPU
tec in graphics performance. With pipeline vector processor and compiler
to unroll loops it was WOW. Todays Ghz processors cannot beat a vector
machine in computation, Titan had a 16 Mhz 1K floating point vector ALU.
Randy, no doubt you know a lot more about the ardent/stellar/stardent stuff
than me. I was aware of it and I once got hold of the design spec for the
TOE processor (the 4x4 pixel "stamper"). However, I think you aren't aware
of how sophisticated todays GPUs are.
16 MHz * 1K flops = 16 Gflops. A single top end GPU is more like 500
GFLOPS (single prec only, though). Today's GPUs have myriad pixel formats,
including ARGB with an FP32 for each component. Pixel shaders are highly
programmable. A single GPU can have > 80 GB/sec of bandwidth to DRAM (not
cache).
The TOE processor was a fixed point affair with limited, fixed point
precision. There is no comparison. I wish I still had the spec to make a
more concrete comparison.
A google search turned up this quote:
With the Dore' rendering package [Borden89], each processor is capable of
rendering a maximum of 20,000 smoothly shaded small polygons/seconds.
Today's GPUs can render thousands of times more triangle per second,
antialiased, with multiple, high quality texture maps and arbitrary
blending.
Another google search
http://www.ece.cmu.edu/~ece548/handouts/17v_perf.pdf
says that the Titan 1 had a 125 ns clock period and two FPUs, for 16
MFLOP/s peak. Perhaps you recall 1K FPUs, but maybe it was a 1K vector
register length. The same pdf (written by Philip Koopman) says that even
with four processor, and with a large (1000x1000) array size, the titan-1
peaked at 15.7 Mflops. It attributes this to the fact that the aggregate
bus bandwidth of the titan was 256 MB/sec. By rewriting the linpack code
to block the data appropriately, they got it up to 46 MFLOP/s.
So, overall, I think there is no comparison. The rose colored glasses of
time have fooled you.
...
Are there any graphics guys on the list?
Yes, from the hw end of things.
2:10 p.m.
New subject: Modern Marvels: Computers ?? no graphics supers
In article <BAY120-F35B55A58C25AAA4E54FD1ABA7F0 at phx.gbl>,
"Randy Dawson" <rdawson16 at hotmail.com> writes:
I totally agree with you, todays GPUs are screamers.
Did you know that the
core of NVIDIA's design team came from SGI?
Not just SGI, but ATI as well. Basically the hardware designers at
lots of graphics outfits left the mainframe/minicomputers/workstation
companies to go work for PC graphics card companies. Its interesting
that the employees can adapt to the change in technology, but the
organizations and companies seem stuck.
This wouldn't be the first time that a company didn't survive a change
in the computing platform. DEC seemd to do OK when workstations hit the
scene, but seemed to totally miss it when PCs hit the scene. SGI had
a PC product very early on and found the results "disappointing" when
compared to workstations, but that misses the point entirely. At the
time it still would have run rings around other comparable PC offerings.
As a result they abandoned the PC platform as a company until it was too
late to be the dominant player. Hell, missing market opportunities is
the whole story of why E&S is now down to 60 employees and circling the
drain for what seems one last time.
Sad to see that company went wintel... I think they
were almost de-listed
from the stock excchage at one point, as the stock dipped below a dollar.
They *were* delisted. They were relisted later. After 9/11 they got
a financial shot in the arm due to a spurt in supercomputer business,
but it wasn't enough to restart the company, really.
Jim, have you followed the recent threads in the news,
to use the GPU
processor fabric as a general purpose engine, not just for pixels?
See <http://www.gpgpu.org/>
I have Stardent (Kubota) Dore' running today and
porting my old app to it.
Is this just the API or do you have a machine?
I knew a guy that left E&S to go work for Kubota Pacific when they
started up -- Kent was his first name, I am drawing a blank on his
last name right now... He worked on the microcode for the Evans &
Sutherland Freedom series of graphics accelerators that were made for
Sun, HP and IBM machines. The microcode was written for the AMD 29000
processor.
--
"The Direct3D Graphics Pipeline" -- DirectX 9 draft available for download
<http://www.xmission.com/~legalize/book/download/index.html>
Legalize Adulthood! <http://blogs.xmission.com/legalize/>
2:30 p.m.
New subject: Modern Marvels: Computers ?? no graphics supers
On 10 Mar 2007 at 12:03, Randy Dawson wrote:
Todays Ghz processors cannot beat a vector machine in
computation, Titan had a 16 Mhz 1K floating point vector ALU.
I am currently working with Ardent/Stellar/Stardent Dore' visualization
library from these machines.
I strongly suspect that the GPU from a modern PC video card or game
can probably run rings around an old Stardent box.
That's not to belittle vector boxes--even modern Intel CPUs have SIMD
instructions.
Personally, I've worked in compiler development for the CDC STAR and
the ETA 10, so I'm not bashing vector machines.
Cheers,
Chuck
2:42 p.m.
New subject: Modern Marvels: Computers ?? no graphics supers
Without Sun, HP, Apollo the next gen CPU ICs could not
be placed and routed.
Engineers needed to interact in realtime with the design at a graphics
workstation to get the job done, to create the IC masks.
Shhh...don't tell the mainframe CAD guys...
--
Will
6497
days inactive
6497
days old
9 comments
6 participants
participants (6)
-
brad＠heeltoe.com -
cclist＠sydex.com -
frustum＠pacbell.net -
legalize＠xmission.com -
rdawson16＠hotmail.com -
wdonzelli＠gmail.com