Mersenne Digest Sunday, May 7 2000 Volume 01 : Number 730 ---------------------------------------------------------------------- Date: Sun, 7 May 2000 08:58:21 -0000 From: "Brian J. Beesley" <[EMAIL PROTECTED]> Subject: Re: Mersenne: The sound of M(M(19)) On 6 May 00, at 22:45, Nathan Russell wrote: > While running ECM on this exponent I have noticed a throbbing sound, > somewhat similiar to a ceramic plate being spun on a hard surface; this > differs from the usual continuous hum of running an LL test. I have read > the FAQ entries relating to noise, but none of them mention a noise related > to ECM. Much more likely to be related to a resonance with the speed at which the FFT works at the exponent you're testing. ECM will be on a small exponent (memory requirements!) so the FFT will run very quickly compared with the same code running on a larger exponent. My guess is that if you run ECM on an exponent say 30% bigger, or smaller, the noise will vanish. > > In any event, though, I cannot help being somewhat curious about the source > of the sound. Does it relate to the writing to and from memory? Probably something loose (maybe inside the processor cartridge) is excited by an oscillating electric field, causing mechanical vibration, which can get large enough in amplitude to make itself obvious by causing contact with a non-moving part if the loose object's natural frequency is close enough to the frequency of the driving field. The cheap & nasty way in which Intel join the heat sink/fan unit into the processor cartridge proper on their SECC2 processors (cartridge slot PIIIs) seems to have the potential to resonate seriously. If you're feeling brave, it would be possible to disassemble this & refit the heatsink properly, eliminating mechanical sloppiness as well as getting better thermal contact. I doubt it's much to do with memory access; for exponents typical for ECM tests, the transforms will run inside the processor's L2 cache, maybe even in the L1 cache. Also, the memory loading for ECM is very different between Stage 1 and Stage 2; if the noise is present during both stages, we can pretty well rule out memory access as its cause. Incidentally, this makes processors with 133 MHz FSB particularly suited to ECM. The point is that, unless you have RDRAM fitted (at enormous expense), your effective memory bus speed will be 100 MHz - _even if you have PC133 SDRAM fitted_ - since there is a bottleneck in the memory subsystem in the VIA Apollo Pro chipset, and also in the Intel i820 / i840 chipset when a memory translation hub is fitted to allow the use of SDRAM instead of RDRAM. If you're working from the L2 cache, the memory bottleneck isn't going to affect you much. There is still some memory access requirement - more in Stage 2 than in Stage 1, but still a lot less than the demands of LL testing an exponent well into 7 figures - but running ECM simply isn't dominated by memory bus throttling in the same way that LL testing can be. Regards Brian Beesley _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Sun, 7 May 2000 08:58:21 -0000 From: "Brian J. Beesley" <[EMAIL PROTECTED]> Subject: Re: Mersenne: Overclocking On 6 May 00, at 23:52, keith wrote: > I have an athlon 700, and i overclocked it to 800 w/o soldering the cache. > Now when I run prime95, I get errors. I clocked it back down to 700, and > prime runs fine. But is there any way i can run it overclocked and run > prime too? Everything else runs perfectly fine exept for that. Thanks all! It's not unusual to find that a system will apparently take more overclocking than Prime95 is comfortable with. Prime95 works some parts of the system (particularly the FPU and the memory) very hard & also reports some errors which would not normally be obvious. Especially when it's being run in self-test or torture test mode. If you elect to dump Prime95 so you can continue to run overclocked to 800 MHz, don't be surprised if your system isn't as stable as it should be. In fact you will probably come across some applications which will crash the system for reasons which are entirely consequential to the excess overclocking. IMHO Prime95 is worth at least $49.95 more than you paid for it as a hardware testing tool alone. If a system will run the full (16-hour) Prime95 self-test without complaining, when run in an environment that's as hot as it ever gets, it's probably not going to have much difficulty running anything reliably. Unless the reliability problem is inherent in the software... BTW - Athlon overclockers please note - the Asus K7M M/B allows one to set the (100 MHz nominal) bus speed in 1 MHz increments. For reasons of reliabilty I'd reccomend finding the fastest bus speed which will reliably run the Prime95 selftest & then reducing by 1 MHz just to be on the safe side. Don't be disappointed if you don't get away with more than 2% or 3% - _anything_ is a bonus! Personally I would not resort to any trick involving increasing core voltages. If you really need the extra performance that much, then either bite the bullet & pay the extra for a chip with a higher rating, or simply wait a few weeks for the price of the faster part to fall. I would reccomend that _any_ overclocked system has a full self-test run regularly - say every 3 months at the outside - as a check that the system is still operating reliably. Of course, this strategy will cost about 1% of the total system performance. In my opinion, better safe than sorry... the hardware's replaceable, and at a cost which will tend rapidly towards zero, but the _personal_ cost of missing a prime due to a hardware error caused by excess overclocking would be unbearable! Regards Brian Beesley _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: 7 May 2000 16:51:29 +0000 From: "Adam Atkinson" <[EMAIL PROTECTED]> Subject: Re: Mersenne: Overclocking There's something funny going on here. Am I the only person who's getting Henrik's message again and again and again with longer sigs each time? On 07-May-00 08:30:37, Henrik Olsen said: >On Sat, 6 May 2000, keith wrote: >> Heya, >> I have an athlon 700, and i overclocked it to 800 w/o soldering the cache. >> Now when I run prime95, I get errors. I clocked it back down to 700, and >> prime runs fine. But is there any way i can run it overclocked and run >> prime too? Everything else runs perfectly fine exept for that. Thanks all! >Consider it an indication that the CPU really can't work reliably at that >speed and don't use it so fast. >-- >Henrik Olsen, Dawn Solutions I/S URL=http://www.iaeste.dk/~henrik/ > Speak of the Devil and he will hear about it. > The Devil's Dictionary, Ambrose Bierce. >_________________________________________________________________ >Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm >Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers >_________________________________________________________________ >Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm >Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers >_________________________________________________________________ >Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm >Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers >_________________________________________________________________ >Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm >Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers >_________________________________________________________________ >Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm >Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers >_________________________________________________________________ >Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm >Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers I've quoted the sigs because, er, it's the whole point of the message. Henrik says it's nothing to do with him. Actually... the same thing's started happening with a Brian J. Beesley message as well. Argh. - -- Adam Atkinson ([EMAIL PROTECTED]) "I got the first three wrong" he said, forthrightly. _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Sun, 7 May 2000 08:58:21 -0000 From: "Brian J. Beesley" <[EMAIL PROTECTED]> Subject: Re: Mersenne: The sound of M(M(19)) On 6 May 00, at 22:45, Nathan Russell wrote: > While running ECM on this exponent I have noticed a throbbing sound, > somewhat similiar to a ceramic plate being spun on a hard surface; this > differs from the usual continuous hum of running an LL test. I have read > the FAQ entries relating to noise, but none of them mention a noise related > to ECM. Much more likely to be related to a resonance with the speed at which the FFT works at the exponent you're testing. ECM will be on a small exponent (memory requirements!) so the FFT will run very quickly compared with the same code running on a larger exponent. My guess is that if you run ECM on an exponent say 30% bigger, or smaller, the noise will vanish. > > In any event, though, I cannot help being somewhat curious about the source > of the sound. Does it relate to the writing to and from memory? Probably something loose (maybe inside the processor cartridge) is excited by an oscillating electric field, causing mechanical vibration, which can get large enough in amplitude to make itself obvious by causing contact with a non-moving part if the loose object's natural frequency is close enough to the frequency of the driving field. The cheap & nasty way in which Intel join the heat sink/fan unit into the processor cartridge proper on their SECC2 processors (cartridge slot PIIIs) seems to have the potential to resonate seriously. If you're feeling brave, it would be possible to disassemble this & refit the heatsink properly, eliminating mechanical sloppiness as well as getting better thermal contact. I doubt it's much to do with memory access; for exponents typical for ECM tests, the transforms will run inside the processor's L2 cache, maybe even in the L1 cache. Also, the memory loading for ECM is very different between Stage 1 and Stage 2; if the noise is present during both stages, we can pretty well rule out memory access as its cause. Incidentally, this makes processors with 133 MHz FSB particularly suited to ECM. The point is that, unless you have RDRAM fitted (at enormous expense), your effective memory bus speed will be 100 MHz - _even if you have PC133 SDRAM fitted_ - since there is a bottleneck in the memory subsystem in the VIA Apollo Pro chipset, and also in the Intel i820 / i840 chipset when a memory translation hub is fitted to allow the use of SDRAM instead of RDRAM. If you're working from the L2 cache, the memory bottleneck isn't going to affect you much. There is still some memory access requirement - more in Stage 2 than in Stage 1, but still a lot less than the demands of LL testing an exponent well into 7 figures - but running ECM simply isn't dominated by memory bus throttling in the same way that LL testing can be. Regards Brian Beesley _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Sun, 7 May 2000 08:58:21 -0000 From: "Brian J. Beesley" <[EMAIL PROTECTED]> Subject: Re: Mersenne: Overclocking On 6 May 00, at 23:52, keith wrote: > I have an athlon 700, and i overclocked it to 800 w/o soldering the cache. > Now when I run prime95, I get errors. I clocked it back down to 700, and > prime runs fine. But is there any way i can run it overclocked and run > prime too? Everything else runs perfectly fine exept for that. Thanks all! It's not unusual to find that a system will apparently take more overclocking than Prime95 is comfortable with. Prime95 works some parts of the system (particularly the FPU and the memory) very hard & also reports some errors which would not normally be obvious. Especially when it's being run in self-test or torture test mode. If you elect to dump Prime95 so you can continue to run overclocked to 800 MHz, don't be surprised if your system isn't as stable as it should be. In fact you will probably come across some applications which will crash the system for reasons which are entirely consequential to the excess overclocking. IMHO Prime95 is worth at least $49.95 more than you paid for it as a hardware testing tool alone. If a system will run the full (16-hour) Prime95 self-test without complaining, when run in an environment that's as hot as it ever gets, it's probably not going to have much difficulty running anything reliably. Unless the reliability problem is inherent in the software... BTW - Athlon overclockers please note - the Asus K7M M/B allows one to set the (100 MHz nominal) bus speed in 1 MHz increments. For reasons of reliabilty I'd reccomend finding the fastest bus speed which will reliably run the Prime95 selftest & then reducing by 1 MHz just to be on the safe side. Don't be disappointed if you don't get away with more than 2% or 3% - _anything_ is a bonus! Personally I would not resort to any trick involving increasing core voltages. If you really need the extra performance that much, then either bite the bullet & pay the extra for a chip with a higher rating, or simply wait a few weeks for the price of the faster part to fall. I would reccomend that _any_ overclocked system has a full self-test run regularly - say every 3 months at the outside - as a check that the system is still operating reliably. Of course, this strategy will cost about 1% of the total system performance. In my opinion, better safe than sorry... the hardware's replaceable, and at a cost which will tend rapidly towards zero, but the _personal_ cost of missing a prime due to a hardware error caused by excess overclocking would be unbearable! Regards Brian Beesley _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Sun, 7 May 2000 10:30:37 +0200 (CEST) From: Henrik Olsen <[EMAIL PROTECTED]> Subject: Re: Mersenne: Overclocking On Sat, 6 May 2000, keith wrote: > Heya, > I have an athlon 700, and i overclocked it to 800 w/o soldering the cache. > Now when I run prime95, I get errors. I clocked it back down to 700, and > prime runs fine. But is there any way i can run it overclocked and run > prime too? Everything else runs perfectly fine exept for that. Thanks all! Consider it an indication that the CPU really can't work reliably at that speed and don't use it so fast. - -- Henrik Olsen, Dawn Solutions I/S URL=http://www.iaeste.dk/~henrik/ Speak of the Devil and he will hear about it. The Devil's Dictionary, Ambrose Bierce. _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Sun, 7 May 2000 08:58:21 -0000 From: "Brian J. Beesley" <[EMAIL PROTECTED]> Subject: Re: Mersenne: The sound of M(M(19)) On 6 May 00, at 22:45, Nathan Russell wrote: > While running ECM on this exponent I have noticed a throbbing sound, > somewhat similiar to a ceramic plate being spun on a hard surface; this > differs from the usual continuous hum of running an LL test. I have read > the FAQ entries relating to noise, but none of them mention a noise related > to ECM. Much more likely to be related to a resonance with the speed at which the FFT works at the exponent you're testing. ECM will be on a small exponent (memory requirements!) so the FFT will run very quickly compared with the same code running on a larger exponent. My guess is that if you run ECM on an exponent say 30% bigger, or smaller, the noise will vanish. > > In any event, though, I cannot help being somewhat curious about the source > of the sound. Does it relate to the writing to and from memory? Probably something loose (maybe inside the processor cartridge) is excited by an oscillating electric field, causing mechanical vibration, which can get large enough in amplitude to make itself obvious by causing contact with a non-moving part if the loose object's natural frequency is close enough to the frequency of the driving field. The cheap & nasty way in which Intel join the heat sink/fan unit into the processor cartridge proper on their SECC2 processors (cartridge slot PIIIs) seems to have the potential to resonate seriously. If you're feeling brave, it would be possible to disassemble this & refit the heatsink properly, eliminating mechanical sloppiness as well as getting better thermal contact. I doubt it's much to do with memory access; for exponents typical for ECM tests, the transforms will run inside the processor's L2 cache, maybe even in the L1 cache. Also, the memory loading for ECM is very different between Stage 1 and Stage 2; if the noise is present during both stages, we can pretty well rule out memory access as its cause. Incidentally, this makes processors with 133 MHz FSB particularly suited to ECM. The point is that, unless you have RDRAM fitted (at enormous expense), your effective memory bus speed will be 100 MHz - _even if you have PC133 SDRAM fitted_ - since there is a bottleneck in the memory subsystem in the VIA Apollo Pro chipset, and also in the Intel i820 / i840 chipset when a memory translation hub is fitted to allow the use of SDRAM instead of RDRAM. If you're working from the L2 cache, the memory bottleneck isn't going to affect you much. There is still some memory access requirement - more in Stage 2 than in Stage 1, but still a lot less than the demands of LL testing an exponent well into 7 figures - but running ECM simply isn't dominated by memory bus throttling in the same way that LL testing can be. Regards Brian Beesley _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: 7 May 2000 16:51:29 +0000 From: "Adam Atkinson" <[EMAIL PROTECTED]> Subject: Re: Mersenne: Overclocking There's something funny going on here. Am I the only person who's getting Henrik's message again and again and again with longer sigs each time? On 07-May-00 08:30:37, Henrik Olsen said: >On Sat, 6 May 2000, keith wrote: >> Heya, >> I have an athlon 700, and i overclocked it to 800 w/o soldering the cache. >> Now when I run prime95, I get errors. I clocked it back down to 700, and >> prime runs fine. But is there any way i can run it overclocked and run >> prime too? Everything else runs perfectly fine exept for that. Thanks all! >Consider it an indication that the CPU really can't work reliably at that >speed and don't use it so fast. >-- >Henrik Olsen, Dawn Solutions I/S URL=http://www.iaeste.dk/~henrik/ > Speak of the Devil and he will hear about it. > The Devil's Dictionary, Ambrose Bierce. >_________________________________________________________________ >Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm >Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers >_________________________________________________________________ >Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm >Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers >_________________________________________________________________ >Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm >Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers >_________________________________________________________________ >Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm >Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers >_________________________________________________________________ >Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm >Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers >_________________________________________________________________ >Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm >Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers I've quoted the sigs because, er, it's the whole point of the message. Henrik says it's nothing to do with him. Actually... the same thing's started happening with a Brian J. Beesley message as well. Argh. - -- Adam Atkinson ([EMAIL PROTECTED]) "I got the first three wrong" he said, forthrightly. _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Sun, 7 May 2000 10:30:37 +0200 (CEST) From: Henrik Olsen <[EMAIL PROTECTED]> Subject: Re: Mersenne: Overclocking On Sat, 6 May 2000, keith wrote: > Heya, > I have an athlon 700, and i overclocked it to 800 w/o soldering the cache. > Now when I run prime95, I get errors. I clocked it back down to 700, and > prime runs fine. But is there any way i can run it overclocked and run > prime too? Everything else runs perfectly fine exept for that. Thanks all! Consider it an indication that the CPU really can't work reliably at that speed and don't use it so fast. - -- Henrik Olsen, Dawn Solutions I/S URL=http://www.iaeste.dk/~henrik/ Speak of the Devil and he will hear about it. The Devil's Dictionary, Ambrose Bierce. _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Sun, 7 May 2000 08:58:21 -0000 From: "Brian J. Beesley" <[EMAIL PROTECTED]> Subject: Re: Mersenne: Overclocking On 6 May 00, at 23:52, keith wrote: > I have an athlon 700, and i overclocked it to 800 w/o soldering the cache. > Now when I run prime95, I get errors. I clocked it back down to 700, and > prime runs fine. But is there any way i can run it overclocked and run > prime too? Everything else runs perfectly fine exept for that. Thanks all! It's not unusual to find that a system will apparently take more overclocking than Prime95 is comfortable with. Prime95 works some parts of the system (particularly the FPU and the memory) very hard & also reports some errors which would not normally be obvious. Especially when it's being run in self-test or torture test mode. If you elect to dump Prime95 so you can continue to run overclocked to 800 MHz, don't be surprised if your system isn't as stable as it should be. In fact you will probably come across some applications which will crash the system for reasons which are entirely consequential to the excess overclocking. IMHO Prime95 is worth at least $49.95 more than you paid for it as a hardware testing tool alone. If a system will run the full (16-hour) Prime95 self-test without complaining, when run in an environment that's as hot as it ever gets, it's probably not going to have much difficulty running anything reliably. Unless the reliability problem is inherent in the software... BTW - Athlon overclockers please note - the Asus K7M M/B allows one to set the (100 MHz nominal) bus speed in 1 MHz increments. For reasons of reliabilty I'd reccomend finding the fastest bus speed which will reliably run the Prime95 selftest & then reducing by 1 MHz just to be on the safe side. Don't be disappointed if you don't get away with more than 2% or 3% - _anything_ is a bonus! Personally I would not resort to any trick involving increasing core voltages. If you really need the extra performance that much, then either bite the bullet & pay the extra for a chip with a higher rating, or simply wait a few weeks for the price of the faster part to fall. I would reccomend that _any_ overclocked system has a full self-test run regularly - say every 3 months at the outside - as a check that the system is still operating reliably. Of course, this strategy will cost about 1% of the total system performance. In my opinion, better safe than sorry... the hardware's replaceable, and at a cost which will tend rapidly towards zero, but the _personal_ cost of missing a prime due to a hardware error caused by excess overclocking would be unbearable! Regards Brian Beesley _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Sun, 7 May 2000 16:53:47 -0000 From: "Brian J. Beesley" <[EMAIL PROTECTED]> Subject: Re: Mersenne: Overclocking On 7 May 00, at 16:51, Adam Atkinson wrote: > There's something funny going on here. Am I the only person who's > getting Henrik's message again and again and again with longer sigs > each time? No, you're not the only person. I think maybe the mail list system needs "the laying on of hands", or maybe a sharp boot if that doesn't work. Actually mail hub administrators all over the world are tearing their hair out trying to upgrade anti-virus software to cope with You Know What, so it might be a Good Idea not to add fuel to the fire for a while. Regards Brian Beesley _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Sun, 7 May 2000 08:58:21 -0000 From: "Brian J. Beesley" <[EMAIL PROTECTED]> Subject: Re: Mersenne: The sound of M(M(19)) On 6 May 00, at 22:45, Nathan Russell wrote: > While running ECM on this exponent I have noticed a throbbing sound, > somewhat similiar to a ceramic plate being spun on a hard surface; this > differs from the usual continuous hum of running an LL test. I have read > the FAQ entries relating to noise, but none of them mention a noise related > to ECM. Much more likely to be related to a resonance with the speed at which the FFT works at the exponent you're testing. ECM will be on a small exponent (memory requirements!) so the FFT will run very quickly compared with the same code running on a larger exponent. My guess is that if you run ECM on an exponent say 30% bigger, or smaller, the noise will vanish. > > In any event, though, I cannot help being somewhat curious about the source > of the sound. Does it relate to the writing to and from memory? Probably something loose (maybe inside the processor cartridge) is excited by an oscillating electric field, causing mechanical vibration, which can get large enough in amplitude to make itself obvious by causing contact with a non-moving part if the loose object's natural frequency is close enough to the frequency of the driving field. The cheap & nasty way in which Intel join the heat sink/fan unit into the processor cartridge proper on their SECC2 processors (cartridge slot PIIIs) seems to have the potential to resonate seriously. If you're feeling brave, it would be possible to disassemble this & refit the heatsink properly, eliminating mechanical sloppiness as well as getting better thermal contact. I doubt it's much to do with memory access; for exponents typical for ECM tests, the transforms will run inside the processor's L2 cache, maybe even in the L1 cache. Also, the memory loading for ECM is very different between Stage 1 and Stage 2; if the noise is present during both stages, we can pretty well rule out memory access as its cause. Incidentally, this makes processors with 133 MHz FSB particularly suited to ECM. The point is that, unless you have RDRAM fitted (at enormous expense), your effective memory bus speed will be 100 MHz - _even if you have PC133 SDRAM fitted_ - since there is a bottleneck in the memory subsystem in the VIA Apollo Pro chipset, and also in the Intel i820 / i840 chipset when a memory translation hub is fitted to allow the use of SDRAM instead of RDRAM. If you're working from the L2 cache, the memory bottleneck isn't going to affect you much. There is still some memory access requirement - more in Stage 2 than in Stage 1, but still a lot less than the demands of LL testing an exponent well into 7 figures - but running ECM simply isn't dominated by memory bus throttling in the same way that LL testing can be. Regards Brian Beesley _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ End of Mersenne Digest V1 #730 ******************************
