Re: SV: SV: Mersenne: Drifting UP(!) in Top Producers ranking?
- Original Message - From: "Torben Schlüntz" <[EMAIL PROTECTED]> To: "Brian J. Beesley" <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]> Sent: Monday, November 25, 2002 10:04 PM Subject: SV: SV: SV: Mersenne: Drifting UP(!) in Top Producers ranking? > I'd rather not like the "penalty"/ punishment. A reward equal to > the full effort of doing the TF would be much better - and under those > circumstances no one would try to cheat because a factor found at eg. 63 > bits would reward very well. That would allow another cheat. Current Factoring assignments are prefactored to 2^57, and are intended to be factored to 2^67. If someone were instead just to factor to 2^58, they would have about a 1/58 chance of getting a full credit for less than 1/500th of the effort. If not, then the exponent could be abandoned. This would also have the advantage (from the cheat's POV) of 'poisoning' potential competitors' factoring efforts. IMO people should expect (in the mathematical sense of the word) to get the same amount of credit irrespective of what type of work they do. Also credit should be given for work (honestly) done irrespective of whether the search was successful. The first criterion (only) could be met by crediting only found factors, and giving a higher credit for larger ones, up to the TF limit. I do not think there is any way to allocate credit that meets both criteria, which wouldn't reward cheating in some way. Brian's suggestion is a good one, but I would add that perhaps each user could get an allowance, proportionate to the number of TF assignments returned, that would be deemed to be 'honest' errors, and not penalised. P-1 (which I do almost exclusively) seems to be woefully ill-rewarded. > tsc Daran G. _ Unsubscribe & list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
RE: Mersenne: christmas computer system?
From: John R Pierce [mailto:[EMAIL PROTECTED]] > ok, true. I forgot VIA is allowing this. AFAIK, > the i845pe doesn't support memory faster than the > CPU FSB (it allows SLOWER memory, the older i845e > only supported 100/200Mhz DDR w/a 133/266/533 CPU bus). Look again... the 845PE and 845GE do, in fact, support DDR333 as well as DDR266 memory with a 533 MHz FSB. See: http://www.intel.com/design/chipsets/845pe/index.htm?iid=ipp_dlc_chip+de skchip_845pe& > I have had so many bad experiences with VIA based > systems and poor PCI and AGP I/O performance that > it will be a long time before I consider one again, > regardless of claimed CPU benchmark performance. I used to agree with you. Heck, I used to have the same "stability" bias against AMD-based systems. However, the small guys usually offer new chipset chipset features and performance levels several months before Intel, so we gave a few 3rd party boards another tryout. I've had no trouble with any of the recent (last 2 years or so) VIA-based or SiS-based motherboards my company has purchased. We always buy Asus, who seems (in my experience) to make the most stable MoBos besides intel. You might want to give them another chance. Regards, -ryan- _ Unsubscribe & list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Re: SV: SV: SV: Mersenne: Drifting UP(!) in Top Producers ranking?
At 11:04 PM 11/25/02 +0100, =?utf-8?Q?Torben_Schl=C3=BCntz?= wrote: No, and I am not the GIMPS police. It would offcourse be quite easy simply to check all accounts having done 5+ years TF and having more than 0,6 years pr. foundfactor. On the other hand some accounts could be very old and back in those days a factor could have been found in less effort than now a days appr. 0,5 y/ff. NetForce and Challenge seems to be good candidates for accounts with a very low effort pr. ff. Well, you'd nail me. I do expired exponents for the most part, which makes it much less likely that I will find a factor because almost all of those expired exponents have already been done part way, and if there had been a factor in the parts already done, they wouldn't have expired. So I have 8.783 P90 years in factoring, and only 6 factors found. Unless you count the pre-factoring work I turn in manually to George. Lots of factors found there for much less CPU expended. _ Unsubscribe & list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Re: Mersenne: christmas computer system?
> > the fastest P4 cpus have a rated FSB of 533 (which is 266*2). running the > > bus at 333(666) would be overclocking. > > Ah, but most modern mobos can run the memory asynchronously to the processor > bus. I think this goes back some way - certainly the Abit KT7A board (Athlon) > could run the memory bus at either FSB or FSB+PCI rate ... > > So you _can_ run memory at 166 MHz double-pumped and the FSB at 133 MHz > quad-pumped without over- or under-clocking anything. ok, true. I forgot VIA is allowing this. AFAIK, the i845pe doesn't support memory faster than the CPU FSB (it allows SLOWER memory, the older i845e only supported 100/200Mhz DDR w/a 133/266/533 CPU bus). I have had so many bad experiences with VIA based systems and poor PCI and AGP I/O performance that it will be a long time before I consider one again, regardless of claimed CPU benchmark performance. _ Unsubscribe & list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
RE: Mersenne: christmas computer system?
From: John R Pierce [mailto:[EMAIL PROTECTED]] >the fastest P4 cpus have a rated FSB of 533 >(which is 266*2). running the bus at 333(666) >would be overclocking. No, it wouldn't. I don't think you understand how the memory architecture of a modern Intel chipset works. The memory clock speed is independent of the FSB speed. The front-side bus connects the CPU to the Northbridge of the chipset. The chipset, in turn, has a separate bus to connect to memory. See this link for an example diagram of a modern P4 chipset: http://www.tomshardware.com/mainboard/02q2/020514/p4x333-04.html Ryan Malayter Sr. Network & Database Administrator Bank Administration Institute Chicago, Illinois, USA PGP Key: http://www.malayter.com/pgp-public.txt ::: Twas a woman who drove me to drink. I never had the courtesy to thank her. -W.C. Fields _ Unsubscribe & list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
SV: SV: SV: Mersenne: Drifting UP(!) in Top Producers ranking?
-Oprindelig meddelelse- From: Brian J. Beesley Sendt: lø 23-11-2002 13:23 >This is not a particularly effective cheat; you still end up having to do >significantly more than half of the computational work. Is there any evidence >that this may be happening? No, and I am not the GIMPS police. It would offcourse be quite easy simply to check all accounts having done 5+ years TF and having more than 0,6 years pr. foundfactor. On the other hand some accounts could be very old and back in those days a factor could have been found in less effort than now a days appr. 0,5 y/ff. NetForce and Challenge seems to be good candidates for accounts with a very low effort pr. ff. >Does it make sense to impose a "penalty clause" i.e. if someone subsequently >finds a factor in a range you claim to have sieved, you lose 10 times the >credit you got for the assignment? N.B. There will be _occasional_ instances >where an "honest" user misses a factor, possibly due to a program bug, >possibly due to a hardware glitch. I'd rather not like the "penalty"/ punishment. A reward equal to the full effort of doing the TF would be much better - and under those circumstances no one would try to cheat because a factor found at eg. 63 bits would reward very well. >> The exponents above >> 79.300.000 are still candidates, though George has chosen to limit his >> program to this size and I think with very good reason. >Hmm. As it happens, one of my systems has just completed a double-check on >exponent 67108763. This took just over a year on an Athlon XP1700 (well, >actually it was started on a T'bird 1200). The fastest P4 system available >today could have completed the run in ~3 months. The point is that running LL >tests on exponents up to ~80 million is easily within the range of current >hardware. Yes, but that kind of hardware was not at the market in 1995. But regarding Moores law George should have predicted the P4 and SSE2? >Personally I feel it is not sensible to expend much effort on extremely large >exponents whilst there is so much work remaining to do on smaller ones. I >justify running the DC on 67108763 as part of the QA effort. Sure. Let's get a new prime and let us have it fast. >> BTW, the list of found factors contains 2.500.000+ but the "top >> producers list" only contains 30.000- of these. GIMPS must be >> responsible for far more than only 30.000 factors. Any explanation for >> that? >Well, there are a lot of factors which can be found by algebraic methods >rather than by direct computation: e.g. if p+1 is evenly divisible by 4, and >p and 2p+1 are both prime, then 2^p-1 is divisible by 2p+1. Evenly? What about 11, 83, 131 and 251 giving: 3,21,33 and 63. Are these just plain luck or does it exist one p+1 / 4 is not even and the factor 2p+1 does not fit? Have a nice day tsc _ Unsubscribe & list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Re: Mersenne: christmas computer system?
> Get a P4 motherboard that supportes DDR333 (PC2700). Then buy the > "Extreme" PC2700 memory from www.corsair.com. It has a significant > performance-enhancing feature: 2.0 cycle latency versus the more > standard 2.5. This high-performance Corsair memory only costs a few > bucks more than standard PC2700 sticks. (I'm sure other people make > CL2.0 sticks, I just haven't found them anywhere for purchase.) the fastest P4 cpus have a rated FSB of 533 (which is 266*2). running the bus at 333(666) would be overclocking. _ Unsubscribe & list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Re: Mersenne: Re: christmas computer system?
On Monday 25 November 2002 12:36, you wrote: > > One should basically not use a CD-R/CD-RW as a general CD reader, since it > usually has way lower MTBF than a normal CD/DVD reader, and is more > expensive. Ie. it breaks a lot earlier if you use it a lot, and it's more > expensive to replace :-) Did you check out the manufacturer's claimed MTBF? A sample I checked out showed no significant difference between CD, CD-R & CD-RW drives. I've _never_ had any problems with CD-RW drives on linux systems, except for an old model which refuses to write some high-speed CD-R media above 2x, though it appears to write medium-speed media at its limit of 4x quite happily. On Windows systems, woe & despondency. Basically I gave up Roxio as a total disaster. Nero is better, but every so often after writing a CD it refuses to read anything in the same drive until system reboot. Never had any genuine hardware problems, though. Playing a lens cleaning CD occasionally does no harm. Most PCs tend to suck air and dust in through gaps in the front panel, this can cause a buildup of crud on the RW head. Regards Brian Beesley _ Unsubscribe & list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
RE: Mersenne: christmas computer system?
Personally, I just ordered a brand spankin' new Compaq server (ugh... HP server now) with dual P4 Xeon 2.8 GHZ processors, 3GB of RAM. I can't wait to get my hot little hands on that and see just how well it crunches the #'s. It's memory is 200MHz DDR (FSB is 400MHz), advanced ECC (Compaq's version of ECC... Detect 4 bit and corrects 2 bit errors, as opposed to detect 2, correct 1 with normal ECC). Anyone on here had any hands on experience with the 2.8 GHz Xeon's yet? Have they been benchmarked? Aaron _ Unsubscribe & list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
RE: Mersenne: christmas computer system?
Get a P4 motherboard that supportes DDR333 (PC2700). Then buy the "Extreme" PC2700 memory from www.corsair.com. It has a significant performance-enhancing feature: 2.0 cycle latency versus the more standard 2.5. This high-performance Corsair memory only costs a few bucks more than standard PC2700 sticks. (I'm sure other people make CL2.0 sticks, I just haven't found them anywhere for purchase.) Any decent white-box manufacturer should be able to find you PC2700 DDR with CAS2.0. Ryan Malayter Sr. Network & Database Administrator Bank Administration Institute Chicago, Illinois, USA PGP Key: http://www.malayter.com/pgp-public.txt ::: Twas a woman who drove me to drink. I never had the courtesy to thank her. -W.C. Fields -Original Message- From: Russel Brooks [mailto:[EMAIL PROTECTED]] Sent: Sunday, November 24, 2002 9:55 AM To: [EMAIL PROTECTED] Subject: Mersenne: christmas computer system? I'm giving my brother's family a new computer for christmas. He'll buy it from a local (to him) 'white box' pc store and I'll pay for it. I am a little concerned about performance because the pc will probably be running GIMPS and I'd like to get my money's worth. It's easy to request a P4 in the 2.0-2.5 range and 256M-512M of memory but I've read of the bottleneck caused by slow memory and the bus between memory and cpu and I don't know what to specify or how to evaluate components in this area or measure performance after the pc is built. Any comments or suggestions? I want them to have a fast machine but not on the bleeding (expensive) edge. Other than GIMPS I'm sure my nephew will be the heaviest load when he plays games on the new machine. I think I should also request the video card has all of it's own memory, right? I don't want the video to share the main memory for performance reasons, right? Cheers... Russ DIGITAL FREEDOM! --> http://www.eff.org/ _ Unsubscribe & list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _ Unsubscribe & list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Mersenne: Re: christmas computer system?
On Sun, Nov 24, 2002 at 08:56:30PM -0800, John R Pierce wrote: > those are very fast CD readers too... I dunno CD-RW, never had much use for > them, but the bulk TDK blanks I get at Costco seem to burn 100% AOK at 40X, > and you can make a 700MB data backup in just a couple of minutes. One should basically not use a CD-R/CD-RW as a general CD reader, since it usually has way lower MTBF than a normal CD/DVD reader, and is more expensive. Ie. it breaks a lot earlier if you use it a lot, and it's more expensive to replace :-) /* Steinar */ - already destroyed two of them, with not very much usage ;-) -- Homepage: http://www.sesse.net/ _ Unsubscribe & list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
