[RC5] d.net Pride

Eric Wikman eric at wikman.com
Sun Jul 26 15:15:19 EDT 1998


I am sure everyone was disappointed at not winning our recent 
DES contest, and I am as ready as anyone to get to v3 and I don't 
want to wait 2+ years for RC5 to finish.  However we have some 
things to be proud of.  For one it looks like we have had a 40% 
increase in daily blocks from our June average to the previous 
week.  

But to me one of the coolest things is being part of the largest 
computer in the world.  I do not know how accurate the teraflop 
rating of d.net is, but it says we are at 4.4 TFlops as of July 3, 
before our 40% increase.  According to my research NEC's SX-5 is 
the fastest computer weighing in at 4 TFlops.  But in order to stay 
on top we have to grow fast.  The following is a timeline of 
supercomputer release dates:

2000 - IBM's 10 TFlop
2002 - NEC's 32 TFlop
2004 - ASCI's 100 TFlop

Now lets see what our projected growth is according to Bruce 
Ford's first and second methods.  Both methods are linear which 
may be unrealistic, but exponential growth does not fit this 
equation well.  Also I do not know if the growth rate of keyrates is 
equivalent to the growth rate in teraflops.  

Method 1 - 
2000 - 6.1 TFlops
2002 - 8.5 TFlops
2004 - 10.8 TFlops

Method 2 - 
2000 - 7.0 TFlops
2002 - 10.6 TFlops
2004 - 14.1 TFlops

NOTE - these statistics are made with the linear equation, 
exponential growth is much more likely.  However the numbers 
achieved with the exponential equation appear unrealistic.  So I do 
believe that we will actually hit all three targets, but using the above 
figures it gives us a strong goal to try to sign up more new 
members.  



Eric Wikman
eric at wikman.com
--
To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
rc5-digest subscribers replace rc5 with rc5-digest



More information about the rc5 mailing list