Mersenne Digest Wednesday, April 19 2000 Volume 01 : Number 721 ---------------------------------------------------------------------- Date: Tue, 18 Apr 2000 20:56:59 +0000 From: "Steinar H. Gunderson" <[EMAIL PROTECTED]> Subject: Mersenne: Re: Facelift (round 2) On Tue, Apr 18, 2000 at 08:50:10PM +0200, Hoogendoorn, Sander wrote: >If you use a seperate frame for the menu you only need to download the gifs >once This should have been done by the browser cache anyway. A browser without a cache today is, well, quite useless. /* Steinar */ - -- Homepage: http://members.xoom.com/sneeze/ _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Tue, 18 Apr 2000 20:58:23 +0000 From: "Steinar H. Gunderson" <[EMAIL PROTECTED]> Subject: Mersenne: Re: Facelift (round 2) On Tue, Apr 18, 2000 at 03:24:40PM -0400, George Woltman wrote: >How about www.microsoft.com which has dropdown menus from the >banner at the top of the page. Isn't that some weird kind of ActiveX or other Microsoft proprietary tech? >I read this somewhere <br clear="all"> or some such. Not `all', `both'. /* Steinar */ - -- Homepage: http://members.xoom.com/sneeze/ _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Tue, 18 Apr 2000 13:57:32 -0700 From: Kevin Sexton <[EMAIL PROTECTED]> Subject: Re: Mersenne: Re: Facelift (round 2) "Steinar H. Gunderson" wrote: > > What about CSS? Take a look at http://zicon.cjb.net/ -- no frames, no FONT > tags, no tables, but it still looks great. > > /* Steinar */ > -- Which browser did you look at that site with? It looks pretty bad in Netscape 4.7. It looks good in IE 5, and Netscape 6 should support it, but that is only an early beta now. _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Tue, 18 Apr 2000 14:32:01 -0700 (PDT) From: John R Pierce <[EMAIL PROTECTED]> Subject: Re: Mersenne: Re: Facelift (round 2) > > On Tue, Apr 18, 2000 at 03:24:40PM -0400, George Woltman wrote: > >How about www.microsoft.com which has dropdown menus from the > >banner at the top of the page. > > Isn't that some weird kind of ActiveX or other Microsoft proprietary tech? actually, I believe its done with client side JavaScript. - -jrp _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Tue, 18 Apr 2000 22:11:33 +0000 From: "Steinar H. Gunderson" <[EMAIL PROTECTED]> Subject: Mersenne: Re: Zicon On Tue, Apr 18, 2000 at 01:30:25PM -0700, Luke Welsh wrote: >Looks like shit in Netscape 4.7 :-( I know -- she's fixing it ATM. My own page (take a look at the `secret' URL http://members.xoom.com/sneeze/redesign/) should work in 4.7, but not in IE. However, I've got a version that works in _both_ IE5, NS4 _and_ Opera here :-) /* Steinar */ - -- Homepage: http://members.xoom.com/sneeze/ _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Tue, 18 Apr 2000 22:18:57 +0000 From: "Steinar H. Gunderson" <[EMAIL PROTECTED]> Subject: Mersenne: Re: Facelift (round 2) On Tue, Apr 18, 2000 at 01:40:36PM -0700, John R Pierce wrote: >unluckily, only MSIE seems to implement CSS properly. Netscape is way behind. Netscape is buggy, buggy, very buggy... You'll have to work a bit. NS6, on the other hand, implements almost total CSS1 (very bug-free) and very much of CSS2... /* Steinar */ - -- Homepage: http://members.xoom.com/sneeze/ _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Tue, 18 Apr 2000 19:58:37 -0400 From: [EMAIL PROTECTED] Subject: Mersenne: Re: Mersenne Digest V1 #720 >From: Pierre Abbat <[EMAIL PROTECTED]> >To tell when you're on line, look for a gateway in the "Flags" column, which >says 4003 for the gateway and 1 for an Ethernet connection. (I don't know whic >h >bit means gateway.) This will work only if the dialup machine is the one >running mprime. If mprime is behind an IPmasq box, you have to ask the IPmasq >box if it's online. None of my routes fit this description. However, as you can see, the machine is online. :) On a side note, alot of multipart posts (especially html) have been floating around lately. I'd like to remind everyone that at least for some of us reading the digest, that's a real nusiance. What disturbed me most this time though, was some non latin1 characters in the digest. The html I can at least scroll past, but could we please keep the control characters to a minimum? _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Tue, 18 Apr 2000 20:39:40 -0500 From: Ken Kriesel <[EMAIL PROTECTED]> Subject: Re: Mersenne: Just curious When pentium pro 200's were the hot new processor (in speed, more so than in wattage), I began running some dual-ppro-200 systems with two prime95 instances each. Those processors are still running it. I've never had to replace a cpu or motherboard (though occasionally a motherboard power connector had to be replaced because it burned up). I'm not sure but I think that's three years. Uptimes for these NT systems were averaging 6 months between reboots, though that has dropped some since the UPSes that power them are aging and so power is less reliable now. I've had a dual-pentium-200-mmx running NT4, and dual prime95 instances, 2 years solid also; the last boot of that system was August 12. I'm sure you'll hear from others, that these durations are not remarkable. Some may advocate other OS's. (I've also run Vaxes for 6-9 months uptime, and power and hardware reliability & application of OS updates was similarly controlling there. Even network switches will occasionally get in funny modes after some months.) The error detection built into prime95 has been useful in identifying some systems where memory simms or motherboards were going flaky, months before the end user noticed it. Ken At 08:24 AM 4/18/2000 +0100, you wrote: >I'm just curious really, but how durable are Intel >processors to continuous number crunching, in other words >has anyone been able to keep the same processor running for >2, 3 or even more years, on a 24/7 basis. I do realise that >Windows itself needs to be rebooted from time to time, but >what about other O/S? Anyone care to throw a few stats in? > >Tony Gott >Shetland > >_________________________________________________________________ >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: Tue, 18 Apr 2000 22:18:16 +0000 From: "Steinar H. Gunderson" <[EMAIL PROTECTED]> Subject: Mersenne: Re: Facelift (round 2) On Tue, Apr 18, 2000 at 02:32:01PM -0700, John R Pierce wrote: >actually, I believe its done with client side JavaScript. Anyways, it doesn't work in NS, and NS _invented_ JS ;-) /* Steinar */ - -- Homepage: http://members.xoom.com/sneeze/ _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Tue, 18 Apr 2000 20:57:19 -0700 From: Eric Hahn <[EMAIL PROTECTED]> Subject: Re: Mersenne: Re: Facelift (round 2) Steinar H. Gunderson wrote: >On Tue, Apr 18, 2000 at 02:32:01PM -0700, John R Pierce wrote: >>actually, I believe its done with client side JavaScript. > >Anyways, it doesn't work in NS, and NS _invented_ JS ;-) That's because the way MS wrote the JS... MS has a variable for the drop-down toolbar menu that's initialized as false. They then check to see if the browser is MSIE, and if it is, changes the variable to true. If is isn't, it leaves it as false, and the drop-down toolbar isn't displayed... BTW, MS also claims it isn't displayed because of a bug in NS, which isn't true, since they don't check and change the variable if NS (or any other browser) is used... >From MS' JS: ==================== var ToolBar_Supported = false; if (navigator.userAgent.indexOf("MSIE") != -1 && navigator.userAgent.indexOf("Windows") != -1 && navigator.appVersion.substring(0,1) > 3) { ToolBar_Supported = true; } ==================== Eric _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Tue, 18 Apr 2000 22:14:01 -0700 From: "John R Pierce" <[EMAIL PROTECTED]> Subject: Re: Mersenne: Just curious > When pentium pro 200's were the hot new processor > (in speed, more so than in wattage), > I began running some dual-ppro-200 systems with two prime95 instances each. > Those processors are still running it. > I've never had to replace a cpu or motherboard > (though occasionally a motherboard power connector > had to be replaced because it burned up). > I'm not sure but I think that's three years. ... Until last August, my *original* Prime95 participant, a Pentium-100 running first Win95, later Win98, faithfully chugged along 24/7. I started this CPU back when the very first Mersenne article came out in the San Jose Mecury News. This was long before GIMPS had found a prime. Since this win95 box's only other duty was print-server for a old inkjet, and the very occasional fax, it went a month or more between reboots regularly. Said machine is still alive and well, only now its a 133MHz 64MB ram linux based internet server for my DSL connection. http://hogranch.com :) The P100 was new when the first 133Mhz pentiums were becoming available and the 90s and 100s got a lot cheaper. Off the top of my head, I think it might be 5+ years old. And, yes, I have a dual PPro-200 which has been running prime95 24/7 since it was built 3 years ago. - -jrp _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Wed, 19 Apr 2000 01:42:51 -0500 From: Ryan McGarry <[EMAIL PROTECTED]> Subject: Re: Mersenne: Just curious All this talk about PC's running 24/7 has convinced me of the reliability of processors. It's been my thoughts that if your computer is on, it's always running at full speed, whether or not you're running Prime95. I've always left my computers on all the time, and never had a problem. This reliability got me thinking about any other appliances that have the same level of reliability. For example, I remember reading an article about light bulbs which said that if you leave a light bulb on continuously, it will last much longer due to the fact that the the filament doesn't contract when it gets cold. I suppose my question is whether or not it's more of a risk to your processor in allowing it to cool off regularly than leaving it on 24/7? Thanks, Ryan McGarry John R Pierce wrote: > > > When pentium pro 200's were the hot new processor > > (in speed, more so than in wattage), > > I began running some dual-ppro-200 systems with two prime95 instances > each. > > Those processors are still running it. > > I've never had to replace a cpu or motherboard > > (though occasionally a motherboard power connector > > had to be replaced because it burned up). > > I'm not sure but I think that's three years. > ... > > Until last August, my *original* Prime95 participant, a Pentium-100 running > first Win95, later Win98, faithfully chugged along 24/7. I started this > CPU back when the very first Mersenne article came out in the San Jose > Mecury News. This was long before GIMPS had found a prime. Since this > win95 box's only other duty was print-server for a old inkjet, and the very > occasional fax, it went a month or more between reboots regularly. Said > machine is still alive and well, only now its a 133MHz 64MB ram linux based > internet server for my DSL connection. http://hogranch.com :) The P100 > was new when the first 133Mhz pentiums were becoming available and the 90s > and 100s got a lot cheaper. Off the top of my head, I think it might be 5+ > years old. And, yes, I have a dual PPro-200 which has been running prime95 > 24/7 since it was built 3 years ago. > > -jrp > > _________________________________________________________________ > 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: Wed, 19 Apr 2000 07:01:11 -0000 From: "Brian J. Beesley" <[EMAIL PROTECTED]> Subject: Re: Mersenne: Re: Facelift (round 2) On 18 Apr 00, at 14:32, John R Pierce wrote: > > > > On Tue, Apr 18, 2000 at 03:24:40PM -0400, George Woltman wrote: > > >How about www.microsoft.com which has dropdown menus from the > > >banner at the top of the page. > > > > Isn't that some weird kind of ActiveX or other Microsoft proprietary > > tech? > > actually, I believe its done with client side JavaScript. If it's a choice between ActiveX, JavaScript or frames, I'll take frames _any_ day. I _strongly_ believe in disabling "active" content wherever possible on grounds of security. Regards Brian Beesley _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Wed, 19 Apr 2000 01:11:08 -0600 From: "Aaron Blosser" <[EMAIL PROTECTED]> Subject: RE: Mersenne: Just curious >All this talk about PC's running 24/7 has convinced me of the >reliability of processors. It's been my thoughts that if your computer >is on, it's always running at full speed, whether or not you're running >Prime95. >I've always left my computers on all the time, and never had a problem. >This reliability got me thinking about any other appliances that have >the same level of reliability. I've had many machines running NTPrime for years now... Sometimes those NT servers are so rock solid they'll run for months at a time between reboots and I've had nary a problem. >For example, I remember reading an article about light bulbs which said >that if you leave a light bulb on continuously, it will last much longer >due to the fact that the the filament doesn't contract when it gets >cold. That's true...but only as far as total number of hours of life. There's a reason that light bulbs almost always burn out right when you turn them on...only VERY rarely will they burn out while already on. Just that having a cold filament which suddenly gets really hot...it can put alot of stress on that poor thing... But of course, you're much better off turning on a light when you need it because *you* will get more use out of it that way than if you just left your lights on all day, all night, all the time. :) I mean, you might get the full 3000 hours out of a bulb if you left it on all the time...but that's only 125 days. Now...take that same bulb and then think that you maybe turn it off and on 3-4 times a day for a grand total of maybe 5 hours a day. Okay... 5 hours a day would normally be 600 days, but the cycling of it will probably reduce it's life by nearly half, but you still get 300 days worth of use out of it. Okay, so I'm definitely over analyzing it... Of course, alot of people leave their computers on all the time because they're not as fast as a light bulb when it comes to turning it on (not usually anyway)... And some businesses need to leave them on all the time for doing software distributions during off hours. So, with that being said, you gotta figure hey...we're wasting all that electricity anyway, let's at least *do* something with it! Sigh...even if a company used the Prime95 time of day stuff to only let it run during certain hours, that'd be a big plus...oh well. I don't suppose George could just program something into the code to have it check for the user being idle (like the screen saver check does, but independent of the system screen saver routines) such that if the user doesn't hit a key or move the mouse for xx minutes, it would begin it's calculations (still at whatever priority you set it to...idle by default), but when the user is hitting keys or moving the mouse, it'll stop calculations altogether? That may allay the (unfounded) fears of some that Prime95 somehow steals cycles from other running programs. Just some thoughts... Oh, and while I'm at it...running Prime95 does put a heavier load on a CPU than if you just let it sit there doing nothing while powered on...merely because the FPU is churning away whereas normally it doesn't do much. That'll increase the heat output some, draw a bit more power... But the CPU's are made to take it, so might as well. And, like we've all been saying, we've had CPU's running Prime95 for years straight with no ill effects. >From my US WEST experience on a nice sample of thousands of machines, I saw that if a CPU was bad at all, it would show up as errors in the prime.log within an hour. If it could make it past that, that CPU is good! :) Aaron _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Wed, 19 Apr 2000 01:45:22 -0700 From: Will Edgington <[EMAIL PROTECTED]> Subject: RE: Mersenne: Just curious Aaron Blosser writes: I don't suppose George could just program something into the code to have it check for the user being idle (like the screen saver check does, but independent of the system screen saver routines) such that if the user doesn't hit a key or move the mouse for xx minutes, it would begin it's calculations (still at whatever priority you set it to...idle by default), but when the user is hitting keys or moving the mouse, it'll stop calculations altogether? That may allay the (unfounded) fears of some that Prime95 somehow steals cycles from other running programs. Unfortunately, I have personal experience in this area, though not with Prime95. My own (UNIX-based) Mersenne programs and scripts, from before GIMPS started, included checks not only that all logged on users were idle for at least three hours, including their terminal, mouse, and keyboard, but also that the load was only the 1.0 due to the program itself (and less than 0.1 or so when starting). These checks themselves (that the load remained low and any users were still idle), when performed every two minutes under SunOS 4.x on the SPARCstation 1's and 2's that were available at the time, usually kicked the load average up another 0.5 or so. But two minutes is quite a while to wait for something hogging your CPU to stop. At least according to the ten or so people that complained out of the roughly 100 computers my scripts were running on for a couple of years. And the scripts were careful to start only after hours, even if the computer appeared idle during the day. And the programs that did the actual work (almost always trial factoring because I didn't have an FFT-based LL program) always ran at the absolute lowest priority UNIX offers. Note further that checking to start things was done remotely; there was _no_ process of mine on the local machine when it was not idle, not merely a process only checking for idleness: the load average and user list could be checked without any local process. There were _still_ complaints. Even though the only thing that some of them could point at that indicated "slowness" was the load average being 1.0 instead of 0.0. So, no matter how much CPU you think this sort of change could gain GIMPS, I must suggest that it _not_ be done. Except - perhaps - under the control of another .ini variable and the default is to do things the current way. Will _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Wed, 19 Apr 2000 10:41:36 +0100 (BST) From: Chris Jefferson <[EMAIL PROTECTED]> Subject: RE: Mersenne: Just curious >.... > So, no matter how much CPU you think this sort of change could gain > GIMPS, I must suggest that it _not_ be done. > .... I'm sure this has come up HUNDREDS of times before, but why not attach it to the screensaver? I'm not sure if there is a call in windows for the screensaver, alternativlywe could write a prime95.scr. If this was done, could I advise that all that is done is a simple plug-in option for another screen saver or something? The only problem with this is most people on modems are unlikely to ever connect to the internet when their screen savers come on. However it would work in a corporate situation. How long would it take to edit prime95.exe to prime95.scr, give it a nice set-up window, give option to load in another .exe file when it starts, and get it to die when the .exe file it loaded died? Also, how long would it take prime95 to save all files, and would running prime for 10 mins actually do anything useful? _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Wed, 19 Apr 2000 09:49:06 -0500 From: Ken Kriesel <[EMAIL PROTECTED]> Subject: Re: Mersenne: Just curious Monitors, on the other hand, seem to like to be shut off regularly. At work, we bought 9 Nanao F750's & &60's in 1993. Only two survive, and one sits on my desk and is turned on and off daily. Those that were left on 24/7 did not survive the last cycle of cpu upgrades. Ken At 01:42 AM 4/19/2000 -0500, Ryan McGarry wrote: >All this talk about PC's running 24/7 has convinced me of the >reliability of processors. It's been my thoughts that if your computer >is on, it's always running at full speed, whether or not you're running >Prime95. >I've always left my computers on all the time, and never had a problem. >This reliability got me thinking about any other appliances that have >the same level of reliability. > >For example, I remember reading an article about light bulbs which said >that if you leave a light bulb on continuously, it will last much longer >due to the fact that the the filament doesn't contract when it gets >cold. > >I suppose my question is whether or not it's more of a risk to your >processor in allowing it to cool off regularly than leaving it on 24/7? > >Thanks, >Ryan McGarry > >John R Pierce wrote: >> >> > When pentium pro 200's were the hot new processor >> > (in speed, more so than in wattage), >> > I began running some dual-ppro-200 systems with two prime95 instances >> each. >> > Those processors are still running it. >> > I've never had to replace a cpu or motherboard >> > (though occasionally a motherboard power connector >> > had to be replaced because it burned up). >> > I'm not sure but I think that's three years. >> ... >> >> Until last August, my *original* Prime95 participant, a Pentium-100 running >> first Win95, later Win98, faithfully chugged along 24/7. I started this >> CPU back when the very first Mersenne article came out in the San Jose >> Mecury News. This was long before GIMPS had found a prime. Since this >> win95 box's only other duty was print-server for a old inkjet, and the very >> occasional fax, it went a month or more between reboots regularly. Said >> machine is still alive and well, only now its a 133MHz 64MB ram linux based >> internet server for my DSL connection. http://hogranch.com :) The P100 >> was new when the first 133Mhz pentiums were becoming available and the 90s >> and 100s got a lot cheaper. Off the top of my head, I think it might be 5+ >> years old. And, yes, I have a dual PPro-200 which has been running prime95 >> 24/7 since it was built 3 years ago. >> >> -jrp >> >> _________________________________________________________________ >> 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: Wed, 19 Apr 2000 16:22:40 -0000 From: "Brian J. Beesley" <[EMAIL PROTECTED]> Subject: Re: Mersenne: Just curious On 19 Apr 00, at 1:42, Ryan McGarry wrote: > I suppose my question is whether or not it's more of a risk to your > processor in allowing it to cool off regularly than leaving it on 24/7? CRT monitor is definitely better OFF when not needed. The electron beams definitely age, and the very high tension circuitry contained in a CRT does represent a (small) fire risk which is eliminated by having it switched off. Power-saving "snooze" mode may reduce electricity costs but has no effect on fire safety. I strongly reccomend using the front panel power switch on a CRT monitor. LCD monitor, & most other electronic components, are definitely better left switched ON. Hard disk drives seem to last much longer if they're left running continously. I'd reccomend disabling "power saving" modes on HDD unless power consumption is critical (e.g. a notebook computer when running on internal power). However, HDDs often fail if they've been running continously for years, then switched off & left off long enough to cool right down. I think the heads get "glued" to the platters. The best advice for HDDs which must be turned off is to turn off, wait for 2 mins, turn on, wait for 10 mins, turn off, wait for 10 mins, turn on, wait for 2 mins & finally turn off. The idea is to dissipate the "glue" which accumulates with constant use. Power supply units seem to have a life which is governed primarily by the number of times they're turned on & off. I've never heard of one failing in service whilst being fed a clean mains supply, but, given a mains glitch, it's common to have to replace a proportion of PSUs. (Like light bulbs, they seem to fail at the instant power is applied) I find the commonest failure in PC systems is _cooling fans_. Like HDDs it seems to be the case that the bearing will "glue up" if an always-running fan is switched off & left to cool right off. Sometimes this makes them very noisy for a few minutes when power is restored, sometimes they just plain fail. Broken cooling fans are definitely very bad for the reliability of PC systems! Regards Brian Beesley _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Wed, 19 Apr 2000 11:18:29 -0600 From: "Aaron Blosser" <[EMAIL PROTECTED]> Subject: RE: Mersenne: Just curious >Hard disk drives seem to last much longer if they're left running >continously. I'd reccomend disabling "power saving" modes on HDD >unless power consumption is critical (e.g. a notebook computer when >running on internal power). However, HDDs often fail if they've been >running continously for years, then switched off & left off long >enough to cool right down. I think the heads get "glued" to the >platters. The best advice for HDDs which must be turned off is to >turn off, wait for 2 mins, turn on, wait for 10 mins, turn off, wait >for 10 mins, turn on, wait for 2 mins & finally turn off. The idea is >to dissipate the "glue" which accumulates with constant use. For what it's worth... I've heard that referred to as "stiction" :) There's a decent enough solution to the problem of "stuck" drive heads, but you really should be absolutely sure that's what the problem is... Drives have a "landing zone" or parking area where the heads will move to when it's powered down. There's no data on that part of the track, so if your heads do get stuck there when it's turned off, there's something you can try... Again, be sure that's really what the problem is before trying this... :) But, in short, with the drive powered on and making the "hey, my heads are stuck!" noise, you gently rap the drive on a hard surface. Rap it harder and harder until finally you hear the heads moving about normally. Usually, the head will unstick itself and as long as the head wasn't actually damaged, you may have just enough time to get your data backed up pronto. Jeremy and I used to do that alot on those first generation IDE drives (which seemed to have this problem much more often) back when we were computer techs... It sounds funny, I know, but it worked great most of the time. For what it's worth, modern drives rarely have this problem. Even the 10,000 RPM drives which get QUITE hot during use have good landing zone areas where the heads aren't likely to come into contact with the hot platters. >Power supply units seem to have a life which is governed primarily by >the number of times they're turned on & off. I've never heard of one >failing in service whilst being fed a clean mains supply, but, given >a mains glitch, it's common to have to replace a proportion of PSUs. >(Like light bulbs, they seem to fail at the instant power is applied) For power supplies, having a decent UPS or even just a good line conditioner is a MUST when you want to prolong it's life. Anyone who cared to could hook an scope to a power line (make sure the scope is protected from overvoltage! :) and if it's a nice digital scope, you can see the surges and sags that happen *all the time*. Of course, not many folks have digital scopes... But a decent UPS does it's own logging...the APC Smart-UPS for instance. It'll keep track of the peaks and valleys through the day and it really is amazing what your poor little power supply has to deal with all the time. Sags can be just as damaging to your supply as a spike, by the way. >I find the commonest failure in PC systems is _cooling fans_. Like >HDDs it seems to be the case that the bearing will "glue up" if an >always-running fan is switched off & left to cool right off. >Sometimes this makes them very noisy for a few minutes when power is >restored, sometimes they just plain fail. Broken cooling fans are >definitely very bad for the reliability of PC systems! I'll second that. Good servers (compaq servers for instance) monitor temps at key points and actually have redundant, hot swappable cooling fans. Gotta love that stuff. Aaron _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Wed, 19 Apr 2000 20:09:50 +0200 (CEST) From: Henrik Olsen <[EMAIL PROTECTED]> Subject: RE: Mersenne: Just curious On Wed, 19 Apr 2000, Aaron Blosser wrote: > For power supplies, having a decent UPS or even just a good line conditioner > is a MUST when you want to prolong it's life. Anyone who cared to could > hook an scope to a power line (make sure the scope is protected from > overvoltage! :) and if it's a nice digital scope, you can see the surges and > sags that happen *all the time*. > > Of course, not many folks have digital scopes... But a decent UPS does it's > own logging...the APC Smart-UPS for instance. It'll keep track of the peaks > and valleys through the day and it really is amazing what your poor little > power supply has to deal with all the time. Sags can be just as damaging to > your supply as a spike, by the way. I keep hearing this stuff about power problems, and while I can understand the need for an UPS, the need for a line conditioner evades me, possibly because I live in Denmark which judging from the stories have much cleaner power. Though this is getting way off topic, does anyone know where it's possible to get comparable data on the quality of the mains power in different countries? If possible, collected neither by the power suppliers nor by UPS manufacturers. :) - -- Henrik Olsen, Dawn Solutions I/S URL=http://www.iaeste.dk/~henrik/ Example is better than following 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 ------------------------------ Date: Wed, 19 Apr 2000 14:55:37 -0500 From: Jeremy Blosser <[EMAIL PROTECTED]> Subject: RE: Mersenne: Just curious - -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Aaron Blosser Sent: Wednesday, April 19, 2000 12:18 PM To: Mersenne@Base. Com Subject: RE: Mersenne: Just curious <SNIP> >For what it's worth... > >I've heard that referred to as "stiction" :) There's a decent enough >solution to the problem of "stuck" drive heads, but you really should be >absolutely sure that's what the problem is... > >Drives have a "landing zone" or parking area where the heads will move to >when it's powered down. There's no data on that part of the track, so if >your heads do get stuck there when it's turned off, there's something you >can try... > >Again, be sure that's really what the problem is before trying this... :) > >But, in short, with the drive powered on and making the "hey, my heads are >stuck!" noise, you gently rap the drive on a hard surface. Rap it harder >and harder until finally you hear the heads moving about normally. Usually, >the head will unstick itself and as long as the head wasn't actually >damaged, you may have just enough time to get your data backed up pronto. > >Jeremy and I used to do that alot on those first generation IDE drives >(which seemed to have this problem much more often) back when we were >computer techs... It sounds funny, I know, but it worked great most of the >time. > If I remember correctly, it seemed to happen with certain batches of drivers. Like a batch of WD drivers and then later maybe Maxtor or whatever. And there was *always* the same problem in that it would spin up and then right back down again at boot (usually the first boot). So I think it was more of a shipping or drive problem or something more than anything. But, as a last ditch effort it works... In reality tho, I think that what is *more* common is the actual controller or PCB on the drive starts to flake out before an actual problem w/ the platters etc. Usually, if it is a platter/head problem its usually due to abuse (such as dropping something heavy on your drive while its reading/writing). I'll never forget the times when we would do data recovery of a bad drive by putting it in the freezer for 30 mins, which we *theorized* shrank the PCB on the HD thus fixing some stress fracture or whatever temporarily (long enough to get the data off the drive before it heated back up again). Just remember the following is ONLY recommended if you are trying last ditch type of things and don't want to spend the $$$ sending it off to a *REAL* data recovery type of place. >For what it's worth, modern drives rarely have this problem. Even the >10,000 RPM drives which get QUITE hot during use have good landing zone >areas where the heads aren't likely to come into contact with the hot >platters. <SNIP> _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Wed, 19 Apr 2000 16:31:36 EDT From: [EMAIL PROTECTED] Subject: Mersenne: factoring Brian Beesley (Mersenne Digest 715, 7. April) wrote: >There are also factoring programs available in portable >high-level-language source format, in particular look >for Mfactor.c If it wasn't for the fact that factoring >is so far ahead of LL testing, I'd probably switch my >Alpha system to factoring - with its 64 bit integer >registers and quad-issue pipeline, the architecture is >much better suited to factoring than LL testing, and >the performance is somewhat startling for factoring >(up to 63 bits) even though the program is neither >tuned to the hardware nor hand optimized. True, the generic-C component of Mfactor is not tuned to any particular hardware. But Peter Montgomery (author of Mfactor) has written assembly code supplements for two architectures (Alpha and MIPS) which support 64 x 64 ==> 128-bit integer multiply, and thus on those platforms the program is very speedy. On the MIPS and pre-21264 Alphas (which don't fully pipeline integer multiply) the single functional unit that does integer multiply is saturated and thus Mfactor may not perform as well on a cycle-by-cycle basis as one would hope for, but the performance should still be quite good. On the Alpha 21264 (which fully pipelines IMUL) Mfactor should really scream, but of course the 21264 is extremely good at the floating-point ops that dominate an Mlucas run, too. The Sparc, alas, has a very poor integer multiply capability and should be used for floating-point work (i.e. LL testing) only. I think the suggestion to have one machine (whether that be a PC running Prime95 or a MIPS or Alpha running Mfactor) do all the factoring needed to keep multiple non-PC machines well-fed with exponents is a good one, since otherwise, juggling factoring and LL work becomes a pain. Note that Mfactor (as currently configured) doesn't support putting multiple exponents into a to-do file, so the best way to trial factor lots of exponents is probably just to paste the one-line inputs needed for the exponents, one after another, into the same window - when the program finishes the current exponent, it'll read the next input line from the buffer. Of course, this doesn't permit one to log out, so is best done on a machine which one owns (then one can at least lock the display when one needs to leave.) Mfactor (source and precompiled binaries for MIPS/Irix and Alpha/Unix) is available at my ftp site: ftp://209.133.33.182/pub/mayer/README.html Happy hunting, - -Ernst _________________________________________________________________ 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 #721 ******************************
