Author: Robert Hyatt
Date: 21:18:36 07/10/02
Go up one level in this thread
On July 10, 2002 at 12:53:11, Joshua Lee wrote: >On July 09, 2002 at 13:28:54, Robert Hyatt wrote: > >>It doesn't... For example, the C90 had a 2 nanosecond clock. The cpu >>could read two 64 bit words and write one 64 bit word per clock cycle, >>per cpu. With 16 cpus, that is 16 * 24 * 500000000 bytes per second >>and that can be _sustained_ forever. >> >>Compare that to any PC you want and you see why (a) the supercomputers are >>so expensive and (b) why the micros have absolutely no chance at catching >>them in terms of speed. >> >192,000,000,000 bytes per second >That's over 178 Gigabytes a second. Is Bandwidth refered to in this way or how >fast the memory can communicate with the cpu or both? In Athlon's it's has a 2.1 >GB/Sec bus it can execute a multiply and add on every clock cycle which gives it >a peak throughput of 3.2 gigaflops. > I don't see any way a 2.1 gigabyte per second memory bandwidth can translate into 3.2 gigaflops. A flop requires accessing two operands, doing something to them, and putting the result back... IE a flop == 12 bytes of memory traffic (cache doesn't count because big applications and arrays don't fit into cache). That translates into maybe 100 gigaflops as a more realistic number... And I don't believe any PC has a prayer of coming within a factor of 10 of that number in reality. >So i don't get confused are they talking about three different things here if so >please explain. How much does each one correspond to performance increase? >and how is it supposed to do anything for a program to run faster when alot of >problems exist anyways wasted cycles inefficient code etc... I am getting ahead >of myself let me keep it simple. > >If i went from SDR to DDR and going from 100MHZ (800MB a sec ) >(PC133 1064MB) to (DDR 200MHZ 1600MB) which would be twice as fast as the PC100 > DDR266 2128MB again twice as fast as PC133 and so on but what does this >mean for the program searching for the moves? It depends on more than bandwidth. DDR has a _higher_ latency than the good old SDRAM. That means that a random byte read takes _longer_. If a program generates a lot of memory traffic, it will suffer on _any_ PC platform. If it is more cache-friendly, then memory bandwidth/latency might not be so critical. It depends on the program... > >Then what about the hash tables themselves ? the program can only read from them >so fast and it must barely make any difference in that way in the search and >more of a difference in like that of an Opening book as it contains Usefull >Information and that's it. All you could do is change what it stores and how >it's read. > >More and More questions to follow > >Thanks
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.