The video card folks can do this because
a) each successive generation of cards is derived from the past, so the
NRE is lower.. most of the card (and IC) is the same
b) they have truly gargantuan volumes
c) they have sales from existing products to provide cash to support the
development of version N+1.
And you can ofcourse use the video card's processors to do generic
numbercrunching. See e.g.
http://www.gpgpu.org
Don't know how e.g. the 16 vertex units on a mainstream graphics card
compare to the clearspeed processing units.
Mattijs Janssens
OpenCFD Ltd.
The Mews, Picketts Lodge,
Picketts Lane, Salfords,
Surrey RH1 5RG.
Tel: +44 (0)1293 821272
Email: [EMAIL PROTECTED]
URL: http://www.OpenCFD.co.uk
_______________________________________________
Beowulf mailing list, [email protected]
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf
