Author: Robert Hyatt
Date: 10:42:45 12/17/02
Go up one level in this thread
On December 17, 2002 at 11:51:29, Vincent Diepeveen wrote: >On December 17, 2002 at 11:33:53, Robert Hyatt wrote: > >>On December 17, 2002 at 11:25:10, Vincent Diepeveen wrote: >> >>>On December 17, 2002 at 10:58:51, Bob Durrett wrote: >>> >>> >>>Indeed you are correctly seeing that DIEP, which runs well on >>>cc-NUMA machines as well, is a very good program from intels >>>perspective, because even a 'second' processor on each physical >>>processor which runs slower will still give it a speedboost, >>>where others simply slow down a lot when you do such toying. >>> >>>So where many programs which will be way slower when running at >>>4 processes/threads at a 2 processor Xeon, the software is the >>>weak chain. >> >>What program fits this description? Not mine... > >*many* programs Bob. Crafty and DIEP aren't the only thing on the planet >which gets used by most people who need a dual to perform multithreading >for them. And majority of them uses NT4 server, 2000 server or XP pro/server. > Yes, but you paint with a broad brush and say many will be way slower" and I can't find a single one. I have run crafty. I have run several parallel programs for things like molecular modeling, a simulation, and _none_ do worse with HT on than with HT off. Not a single one so far. Some do _far_ better with SMT than Crafty. Some do about the same. But I haven't found a one yet that does worse unless you take the known exception for two processes running on one physical cpu while the other is idle. That will be fixed shortly in linux, and is already fixed in windows .net >For sure not some non existing OS that is seeing the clear difference >between physical and split processors! Windows .net is not "non-existing". I don't follow the linux development kernels but wouldn't be surprised if they already have fixes being tested... > >> >>> >>>In case of DIEP the bottleneck is the hardware clearly. Even >>>something working great on cc-NUMA doesn't profit too much from >>>the SMT/HT junk from intel. >>> >>>Though it is a great sales argument, the hard facts (11.4% >>>speedboost) are not lying. >>> >>>So they need to press 2 cpu's which results in a cpu price >>>2 times higher *at least* than an AMD cpu, the result >>>is that you win 11.4% in speed. >> >>What are you talking about? SMT doesn't "press 2 cpus". > >the size of a P4 processor is a lot bigger than the AMD core, that's >explaining for a big part why P4 is so much more expensive than a K7. What does that have to do with SMT? > >>> >>>Though i am not a hardware engineer, i can imagine the problems >>>they had getting this to work. >>> >>>Instead of a P4-Xeon cpu clocked at 2.8Ghz which can split itself >>>into 2 physical processors, i would have preferred a P3-Xeon cpu >>>which splitted itself into 2 real processors (so each having its >>>own L1 and L2 caches) clocked at 2.0Ghz. >>> >>>That would have kicked anything of course from speed viewpoint as >>>it scales 1 : 1.2 to a K7 (k7 20% faster for each Ghz than the P3). >>> >>>Now we end up with a very expensive cpu which is 1 : 1.4 and a bad >>>working form of HT/SMT. >>> >>>So it's not DIEP having a problem here. But the hardware very clearly. >>>Intel optimistically claims 20% speed boost here and there. Others >>>claim 11% for database applications. >>> >>>I see 11.4% for DIEP. So that's a market conform viewpoint. >>> >>>The not so amazing thing of this all is that a 2.8Ghz Xeon being not >>>deliverable yet here is very expensive (even a 3.06Ghz P4 is already 885 >>>euro in the shops here also not yet deliverable) and the MP2200 which >>>DOES get offered for sales here is 290 euro. the fastest Xeon i see >>>getting offered socket 603 is a 2.0Ghz Xeon for 829 euro at alternate.nl >>> >>>a dual motherboard for the P4 i see here is several: >>> 789 euro for a dual xeon motherboard called: 860d pro (msi) >>> 549 euro for a tyan S2720GN is by far the cheapest i see >>> >>>then you gotta buy ecc registered DDR ram for it. >>> >>>a dual motherboard for K7 i see at the same alternate.nl is: >>> 259 euro for A7M266-D/U >>> 299 euro chaintech 7KDD (dual; U-DMA/133 RAID en sound) AMD-762MPX >>> 289 euro tiger MPX S2466N-4M >>> >>>The last mainboard (tiger) for sure needs registered DDR ram. but lucky >>>not ECC ram. >>> >>>the P4 dual motherboards need for sure ecc registered stuff. >>> >>>The only good news is that ddr ram ecc registered is a lightyear cheaper >>>than ecc registered RDRAM. >>> >>>RDRAM RIMM 256 MB (ValueRAM, ECC) voor PC PC1066 EUR 239,00 >>>now you can't need 256MB at all. You need more RAM than that. which is >>>exponential more expensive i fear. >>> >>>You get better served with DDR ram though: >>> kingston 1GB DIMM 1 GB (Registered) for PC PC266 EUR 599,00 >>> >>>It is amazing how many professors and others still throw away money >>>to get that dual 2.8Ghz P4 which is over 2 times more expensive than >>>AMD dual at the moment is. >>> >>> >>> >>> >>>>On December 17, 2002 at 10:10:46, Vincent Diepeveen wrote: >>>> >>>>>Hello, >>>> >>>><snip> >>>> >>>>> >>>>>Best regards, >>>>>Vincent >>>> >>>>Vincent: >>>> >>>>Please help me to understand this. I had the impression that a software >>>>package's design makes a huge difference in how well that software will utilize >>>>a given hardware package. In the past, you talked about the advantage of >>>>portability. The "ideal" chess engine would run optimally on "anything." >>>> >>>>It seems to me that evaluation of the suitability, of a particular hardware >>>>configuration, for chess purposes must be measured using several or many >>>>different chess software packages. >>>> >>>>How do you know, for sure, that your program will run properly on the hardware >>>>you're discussing? How do you separate out the evaluation of the hardware from >>>>the software? Doesn't performance depend on both? If you get poor performance, >>>>how do you isolate the problem to the hardware? >>>> >>>>Bob D.
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.