🚀 go-pugleaf

In article <1993Jan07.115610.21080@crash>, uzun@crash.cts.com (Roger Uzun) 
writes...
> 
>I am curious, if anyone knows, what the performance of the
>Cyber series of computers was in the mid 1980's.
>How would their integer or FP performance compare to a
>50 Mhz 80486, in running a single, CPU intensive task.
> 
>-Roger
>--------------------------------------------------------------
>bix: ruzun
>NET: uzun@crash.cts.com


CYBERs were 60 bit machines except the newer NOS/VE machines which were
64 bits.

I have all sorts of numbers somewhere at home, mostly comparing CYBERs
to VAXes.  Plus I did a fairly comprehensive history of the CYBER models
over time with their approximate speeds (again somewhere at home).

About the only number I can remember off the top of my head is that a
dual cpu CYBER 865 was good for about 26 mips.  I think that is about 
right.  We old CDC gurus are beginning to lose our memories.

Stan

In article <1993Jan07.115610.21080@crash>, uzun@crash.cts.com (Roger Uzun)
wrote:
> 
> I went to undergraduate school at the Univ of IL(83), and to 
> graduate school at San Diego State Univ.(85).  Both campuses
> had CDC Cyber machines I used in a limited capacity.
> I recall I was impressed with the speeds of the machines
> at the time, but I do not recall the exact models that were
> used.  They were 1's complement machines, and used 60 or perhaps
> it was 63 bit integers.
> 
> I am curious, if anyone knows, what the performance of the
> Cyber series of computers was in the mid 1980's.
> How would their integer or FP performance compare to a
> 50 Mhz 80486, in running a single, CPU intensive task.
> 
> -Roger
> --------------------------------------------------------------
> bix: ruzun
> NET: uzun@crash.cts.com

The LINPACK performance table can be obtained by sending to netlib@ornl.org
a message saying:
send performance from benchmark

Here is what I found:
CDC Cyber 2000V 32 Mflops
CDC Cyber 205(4-pipe) 17 Mflops
CDC Cyber 205(2-pipe) 17 Mflops
CDC Cyber 990E 12 Mflops
CDC Cyber 4680 9.4 Mflops
CDC Cyber 875 4.8 Mflops
CDC Cyber 176 4.6 Mflops
CDC Cyber 760 2.6 Mflops
CDC Cyber 175 2.1 Mflops
CDC Cyber 180-860 2.1 Mflops

The first 486 listed is
Compaq Deskpro 486/331-120w/487  1.4 Mflops

You can get the benchmarking program and run it on your machine.


Alain Martineau
Hydro Quebec
martineau@macmartineau.ccr.hydro.qc.ca

In article <martineau-080193081105@macmartineau.ccr.hydro.qc.ca>, martineau@MacMartineau.ccr.hydro.qc.ca (Alain Martineau) writes:
|> In article <1993Jan07.115610.21080@crash>, uzun@crash.cts.com (Roger Uzun)
|> wrote:
|> > 
|> > I went to undergraduate school at the Univ of IL(83), and to 
...
|> The LINPACK performance table can be obtained by sending to netlib@ornl.org
...
|> CDC Cyber 175 2.1 Mflops

The U of Illinois machine was a 175, so I've highlighted the applicable
entry from Alain's reports.

This could be slightly off, as the UIUC machine was an "A" model, and in
the "C" models there were multiple memory speeds offered.  The LINPACK
summaries do not appear to specify which model of 175 was measured.

-- 
Gordon K. Chace (former UIUC computer center employee)
Control Data Systems, Inc.   phone (612) 482-6524   fax 482-2791
email: gordon.chace@cdc.com or gkc@udev.cdc.com
RISC: Really Invented on the Sixty-six-hundred by CDC

In article <martineau-080193081105@macmartineau.ccr.hydro.qc.ca> martineau@MacMartineau.ccr.hydro.qc.ca (Alain Martineau) writes:
...in response to request for info about CDC Cybers vs 486PCs...
>
>The LINPACK performance table can be obtained by sending to netlib@ornl.org
>a message saying:
>send performance from benchmark
>
>Here is what I found:
>CDC Cyber 2000V 32 Mflops
>CDC Cyber 205(4-pipe) 17 Mflops
>CDC Cyber 205(2-pipe) 17 Mflops
>CDC Cyber 990E 12 Mflops
>CDC Cyber 4680 9.4 Mflops
>CDC Cyber 875 4.8 Mflops
>CDC Cyber 176 4.6 Mflops
>CDC Cyber 760 2.6 Mflops
>CDC Cyber 175 2.1 Mflops
>CDC Cyber 180-860 2.1 Mflops
>
>The first 486 listed is
>Compaq Deskpro 486/331-120w/487  1.4 Mflops
>

As a further clarification for those not familiar with the Linpack
benchmark suite, these numbers reference the 100x100 version of Linpack,
which requires an all Fortran solution.  The 1000x1000 version allows
architecture specific algorithms and assembly language tuning.  The
differences can be substantial.  For example, the Cyber 205(4-pipe) is listed
at 195 Mflops for the 1000x1000 problem set vs 17 Mflops for 100x100 problem
set, an improvement of over 1000%.  The range of difference varies strongly
with architecture, with vector and parallel architectures typically showing
much larger gains for the larger problem than scalar architectures.  For
example, the DEC VAX 6000/410 (1 processor, scalar) shows 1.2 Mflops on
100x100, and only 1.5 Mflops on 1000x1000, an improvement of only 25%.
Vendors who do not expect significant gains on the 1000x1000 problem over
the 100x100 problem are inclined to not submit results for the 1000x1000
problem.  So, my conclusion is that the linpack comparison shown above tends
to understate the floating point performance advantage of the Cybers over a
PC.  Readers should also be aware that Linpack performance says nothing
about integer or I/O performance.

If this discussion whets your appetite for more architecture/performance
comparisions, read comp.benchmarks for a few weeks.  Also be prepared for
the constant cavet 'your mileage will vary'.

In article <martineau-080193081105@macmartineau.ccr.hydro.qc.ca> martineau@MacMartineau.ccr.hydro.qc.ca (Alain Martineau) writes:
>
>Here is what I found:
>CDC Cyber 2000V 32 Mflops
>CDC Cyber 205(4-pipe) 17 Mflops
>CDC Cyber 205(2-pipe) 17 Mflops
>CDC Cyber 990E 12 Mflops
>CDC Cyber 4680 9.4 Mflops
>CDC Cyber 875 4.8 Mflops
>CDC Cyber 176 4.6 Mflops
>CDC Cyber 760 2.6 Mflops
>CDC Cyber 175 2.1 Mflops
>CDC Cyber 180-860 2.1 Mflops
>
>The first 486 listed is
>Compaq Deskpro 486/331-120w/487  1.4 Mflops
Thanks for the numbers.
I have linpack and have run it on my Amiga 4000 (68040/25) it gives me
1.24 MFlops.  Which is fine for Floating Point, but I wonder how
the Integer performance of a Cyber vs an 040/486 is.

-Roger
--------------------------------------------------------------
bix: ruzun
NET: uzun@crash.cts.com

>>>>> On 7 Jan 93 19:56:10 GMT, uzun@crash.cts.com (Roger Uzun) said:
Roger> I went to undergraduate school at the Univ of IL(83), and to 
Roger> graduate school at San Diego State Univ.(85).  Both campuses
Roger> had CDC Cyber machines I used in a limited capacity.
Roger> I recall I was impressed with the speeds of the machines
Roger> at the time, but I do not recall the exact models that were
Roger> used.  They were 1's complement machines, and used 60 or perhaps
Roger> it was 63 bit integers.
Roger> I am curious, if anyone knows, what the performance of the
Roger> Cyber series of computers was in the mid 1980's.
Roger> How would their integer or FP performance compare to a
Roger> 50 Mhz 80486, in running a single, CPU intensive task.

I remember old (1972) CDC6400, which needed 0.6us for 60-bit integer
add (register to register). Because 80486 doesn't have 60-bit registers
and only 7 32-bit registers can be used for arithmetic operations, the
old CYBER can show better speed than 50MHz 80486 if most of operations
can be done on its registers and require more than 32-bit precision.
Probably an example may be converting word(s) from binary to octal
or reverse, on CYBER it can be done by straight code without loop and
without processing every byte separately (quess what way :-).
Of course CDC computers made in the 1980'th were much better (10x?).