🚀 go-pugleaf

In article <1993Jan20.211032.11929@hubcap.clemson.edu>, richard@meiko.com (Richard Cownie) writes:
|> knut@uni-paderborn.de (Knut Menzel) writes:
|> : 
|> : 	4.4 Gflops - (1992)
|> : 	Paderborn Center for Parallel Computing
|> : 	pcpc@uni-paderborn.de
|> : 	1) Parsytec SuperCluster, 320 Inmos T800, 900 Mflops
|> : 	2) Parsytec GC Giga Cube, 1024 Inmos T805, 4.4 Gflops
                                             ^^^^
                                             It's new
|> : -- Knut Menzel
|> 
|> This seems to imply a figure of over 4MFLOPS per T800.  Last time
|> I programmed one, it was a real struggle to achieve over 1MFLOPS
|> even for inner loops of vector routines.  Not that 4MFLOPS is
|> necessarily a lie, you might manage it if you're adding two 32-bit
|> zeros (and never storing the result).  But it's certainly stretching
|> the truth a long way: there are lies, damned lies, and statistics.

Sorry, it is the T805 (30 MHz) and the given peak (!) performance is
about 4.4 MFlops. However, I assume everybody knows about peak performance 
and its real value. In addition my opinion is that most of the given
performance values in the presented list are peak values.

-- Knut Menzel
--------------------------------------------------------------------------
 Knut Menzel                          University of Paderborn
 email:   knut@uni-paderborn.de       Department of Computer Science FB17
          ...!uunet!unido!pbinfo!knut Warburger Str. 100
 Tel:     +49 5251 603325             D-4790 Paderborn
 Telefax: +49 5251 603853
--------------------------------------------------------------------------