[rc5] Supercomputers & Bovine, Q&A

Bob Krzaczek rskpci at cis.rit.edu
Mon Aug 4 13:48:49 EDT 1997


On Sun, 3 Aug 1997, Timothy I Mattox wrote:
> I looked into porting the v1 client to our MasPar MP-1 at Purdue, but after
>[ ... ]
> Does anyone actually want a Bovine client for a MasPar?

I suspect the core developers wouldn't be interested.  Some time ago
(prior to the v2 release), I made two email offers (PGP signed and dated),
offering abide by their source distribution policies, in order to port the
then-up-and-coming v2 client to a Parsytec parallel machine, as well as
performing Digital Alpha optimizations.  At the time, I thought the
Parsytec was one of the parallel machines on the market uniquely suited to
RC5---each node in such a machine is comprised of a PPC and a Transputer. 

At the time, I felt a little rebuffed; my messages were never even ACK'ed
or NAK'ed.  Not even a single word reply of ``No''.  For a short period of
time, I entertained the idea of starting a competing effort, with publicly
disclosed protocols and sources.  I eventually dropped the idea, not on
technical grounds (an open and secure protocol for this sort of game seems
rather straightforward, and besides, my experience is that security
through obscurity never works in the long run), but because I didn't enjoy
the word ``competition'' being applied to an effort like this.  And, let's
be honest, phrases like ``alternative effort'' are mere euphamisms for
this.  We're here, first and foremost, to demonstrate the inadequacies of
56 bit keys, not to win some damn prize. 

Anyway, I later realized that the bulk of the machines available for an
effort like this are mainstream platforms based mostly on Intel and
Motorola CPUs, running mainstream OS's like Linux, MacOS, OS/2, Windows,
etc.  There may not be a great need for optimized Digital Alpha clients; 
there is undoubtedly a smaller need for esoteric parallel machines like
Parsytec's.  Compared to the immense numbers of personal computers out
there, the potential impact of such a machine (or even a small group of
them) is small.

So, as I say, I doubt they're interested.  This isn't meant as an insult
to them; I'm working under the assumption that unless they see a huge need
for a given client, they aren't going to waste whatever resources it takes
to support its development.  In the interest of remaining a ``team
player'', I've got a small handful of Alphas cranking over at a mediocre
key rate, and that's probably the way it will remain for us.  Oh well. 

Okay, I've rambled enough. 

// bob

-- 
// Bob Krzaczek
// Center for Imaging Science, RIT                        <rskpci at cis.rit.edu>



----
To unsubscribe, send email to majordomo at llamas.net with 'unsubscribe rc5' in the body.



More information about the rc5 mailing list