rdnelson@phoenix.Princeton.EDU (Roger D Nelson) (04/11/89)
Standard apology for asking what may be an old question. What is the net's experience-based wisdom on the effect of leaving a micro-computer turned on? I have an Amiga 1000 with an external floppy, a printer and modem, all powered through a spike suppressor outlet gang. How is the longevity of the computer affected when there is no activity -- merely turned on -- and when there is occasional activity prompted by timer activated invocation of programs, e.g., a one hour period every 12 hours? How about the peripherals? There are chips and diodes that are on, warm and active -- do they get "used up" like an incandescent bulb? And does the transition shock of being turned on/off cause more damage than continuous running? I can imagine a function relating MTBF to the mean length of on-time or the frequency of cold-boots, or some similar variable. While I have your attention, may I expand the question to include systems that have a hard drive? And finally, is there a difference in regard to the question of continuous running for the various types of computer? At work I use IBM PCs and ATs, and wonder about leaving them running also. Roger Nelson rdnelson@phoenix.princeton.edu
gentry@kcdev.UUCP (Art Gentry) (04/12/89)
In article <7684@phoenix.Princeton.EDU>, rdnelson@phoenix.Princeton.EDU (Roger D Nelson) writes: > Standard apology for asking what may be an old question. > > What is the net's experience-based wisdom on the effect of leaving a > micro-computer turned on? I have an Amiga 1000 with an external [......] Standard apology for using same ol' answer :-) As a general rule, we leave EVERYTHING on at all times. Part of your question shows you don't really understand what causes bulbs to blow (and likewise, other electronic devices). Bulbs pop when shocked, either from a physical shock or the current surge shock that occurs when you turn it on. Rarely will a bulb go when just left on and left alone. Same applies to other electronic devices, there is a tremendeous surge of current at poweron, and this is what puts all the strain on your equipment. While parts will deteriorate, in particular capacitors, they will last much longer if left in a constant state. Art I am fully qualified to hold intimate conversations with inanimate objects.
aberg@math.rutgers.edu (Hans Aberg) (04/12/89)
Most computers feel better when continuously left on. The reason is that the heat make the circuitry expand and contract, so you get an exhaustion problem if you often turn the computer off and on. Hard disks, on the other hand, should be left off, with the head parked, if you think that somebody might bump or move the computer. Hans Aberg, Mathematics aberg@math.rutgers.edu
jafo@hpfcdc.HP.COM (Sean Reifschneider) (04/13/89)
>What is the net's experience-based wisdom on the effect of leaving a >micro-computer turned on? I have an Amiga 1000 with an external > [...] I feel that leaving it on is the best bet. I don't have any proof, except all the equipment around here that is left on. We have OLD computers here that were left on all the time. I leave my computer on at home (a 2000 with hard disc, printer and modem). The printer I turn off because I don't use it very much (once or twice a week). Everything else I use daily. NOTE: It would be a good idea to get a screen blanker if you are going to do this. Otherwise you are liable to have a nice little bar burned into the screen from your CLI window(s). I also leave my stereo on, but that's a matter for another notes file. Alwhough I did ask there about it, too. Most people with the real expensive equipment said that the only time they had tubes go out was when they did a power on. Another good reason to leave it on, is that I don't like the sound of a hard disc going to sleep. <RRRrrreeeeeooooooooowwwwwwwwwwwwwwwwwwww> Sean
ciriello@lafcol.UUCP (Patrick Ciriello II) (04/13/89)
Would be to look at all the BBS's in the world that operate 24 hours a day, 7 days a week, all year round. I have never seen a computer die of overheating ( except when obvious abuse is involved ). I keep mine on all the time. I have two floppies and a hard disk, XT clone. Never turn it off unless I run something that hangs the system. At work, I have basically the same thing, and it has been running now for a solid three months. Other examples ... LAN servers ... 24 hour Voice-Responce Units (which I have had the pleasure of working with for about a year and half before coming to Lafayette ). .... No problems with leaving it on. Pat Ciriello II Supervisor of Networking and Tech. Services Lafayette College bitnet: ciriello@lafayett uucp: ciriello@lafayett.uucp
mjl@ut-emx.UUCP (Donald A Kassebaum) (04/13/89)
In article <7684@phoenix.Princeton.EDU> rdnelson@phoenix.Princeton.EDU (Roger D Nelson) writes: >Standard apology for asking what may be an old question. > >What is the net's experience-based wisdom on the effect of leaving a >micro-computer turned on? [stuff deleted] Leave them on all the time, BUT... Make sure the video to your monitor is turned OFF, one way or the other. If you rely on a screen blanker, beware what will happen given a power failure (say, isn't that a picture of a hand that's burned into your phosphor? :-). Be sure your components are adequately ventilated. Leaving them on all the time may make quite a difference if they're running too hot already. ("Overheating" was implicated in the demise of my 1084 monitor. The tech who performed the autopsy recommended the use of a small fan [available @ Radio Shack] attached to the vent holes, especially if you have it sitting in an enclosure). Maurice LeBrun | "So then I says to Borg, `You know, Institute for Fusion Studies | as long as we're under siege, one of us University of Texas at Austin | oughta moon these Saxon dogs.'" Internet: | mjl@fusion.ph.utexas.edu | (Far Side)
andrews@cos.com (Andrew R. Scholnick) (04/13/89)
In article <7684@phoenix.Princeton.EDU>, rdnelson@phoenix.Princeton.EDU (Roger D Nelson) writes: > Standard apology for asking what may be an old question. > > What is the net's experience-based wisdom on the effect of leaving a > micro-computer turned on? I have an Amiga 1000 with an external > floppy, a printer and modem, all powered through a spike suppressor > outlet gang. How is the longevity of the computer affected when there > is no activity -- merely turned on -- and when there is occasional > activity prompted by timer activated invocation of programs, e.g., a > one hour period every 12 hours? How about the peripherals? > There are chips and diodes that are on, warm and active -- do they get > "used up" like an incandescent bulb? And does the transition shock of > being turned on/off cause more damage than continuous running? > I can imagine a function relating MTBF to the mean length of on-time or the > frequency of cold-boots, or some similar variable. > > While I have your attention, may I expand the question to include > systems that have a hard drive? And finally, is there a difference in > regard to the question of continuous running for the various types of > computer? At work I use IBM PCs and ATs, and wonder about leaving > them running also. > > Roger Nelson rdnelson@phoenix.princeton.edu You will find all sorts of opinions on this subject, my favorite follows: Systems with Hard Disks are better left on all the time. The repeated heating and cooling of drive elements wears them out, causes substrates to flake sooner (Read/Write errors) and drives to whine before their time... Computers, well, got a coin? Heating and cooling components (which is what happens when you turn them on and off) can cause fracturing of connections and 'stress' the electronic elements needlessly. However, if you only use the thing for an hour or two once or twice a week, the extended running time may wear it out just as fast or faster... How heavily is your machine being used by these timed tasks? I use my machine for daily mail and programming activities about 3 hours a day (on weekdays) to 10 hours a day (on weekends) and run a news feed over night, so I just leave it on all the time. Of corse my Uninterruptable Power Supply and Line Noise Filters help my piece of mind considerably ;-). Just my $.02. ARS. -- - Andrew R. Scholnick @ Corp. for Open Systems, McLean, VA -- andrews@cos.com - {uunet, sundc, decuac}!cos!andrews -- Everything I write blame on me, NOT -- my employer. - "Adventure is when you toss your life on the scales of -- chance and wait for the pointer to stop." - M. Leinster (First Contact)
840445m@aucs.UUCP (Alan W. McKay) (04/14/89)
In article <11984@ut-emx.UUCP> mjl@emx.UUCP (Maurice LeBrun) writes: >In article <7684@phoenix.Princeton.EDU> rdnelson@phoenix.Princeton.EDU (Roger D Nelson) writes: >>Standard apology for asking what may be an old question. >> >>What is the net's experience-based wisdom on the effect of leaving a >>micro-computer turned on? [stuff deleted] > >Leave them on all the time, BUT... > YUP, I've left my Amiga 1000 on for as long as 2months strait. No problem at all. But I do turn my monitor off. -- % Alan W. McKay % % % Acadia University % " The world needs more Socrates % % Wolfville N.S. % walking the streets today " % % CANADA % - S. Corbett %
dwl10@uts.amdahl.com (Dave Lowrey) (04/14/89)
I can see how it is a good idea to leave your CPU and Monitor turned
on all of the time. However, I am confused about the hard disk issue.
Is having the disk spinning constantly, wearing down the bearings, REALLY
better than having it go through numerous warm/cool cycles?
If I do leave it on all of the time, should I have the heads autopark
automatically after (I think) 3 seconds?
--
-------------------------------------------------------------------
"What is another word for 'Thesaurus'?" Steven Wright
Dave Lowrey
Amdahl Corp.
Houston, Texas
(713)-850-8828
...!{ames,sun,decwrl,uunet,....}!amdahl!dwl10
[ The opinions expressed <may> be those of the author and not necessarily
those of his most eminent employer. ]new@udel.EDU (Darren New) (04/15/89)
In article <11984@ut-emx.UUCP> mjl@emx.UUCP (Maurice LeBrun) writes: > >Make sure the video to your monitor is turned OFF, one way or the other. >If you rely on a screen blanker, beware what will happen given a power >failure (say, isn't that a picture of a hand that's burned into your >phosphor? :-). I have heard, as an unsupported rumor, that the Amiga video chips may incur damage if the video is turned off while the computer is on. I've never had any problem, but then I don't turn the monitor off while the CPU is on for longer than it takes me to turn the CPU off. I would suggest turning down the brghtness instead of turning off the power, just in case. P.S., kickbench (for A1000 owners) can avoid problems with the power coming on and the hand appearing. I assume real-ROM based machines can fend for them selves via s:startup-sequence as far as screen-blankers go.
ugkamins@sunybcs.uucp (John Kaminski) (04/15/89)
In article <13123@louie.udel.EDU> new@udel.EDU (Darren New) writes: >I have heard, as an unsupported rumor, that the Amiga video chips >may incur damage if the video is turned off while the computer >is on. Slight chance of a voltage spike coming thru the monitor cabling. I suppose this is akin to shutting down of a refrigerator motor someplace close to your computer while having it on. I think it is about the same risk that is incur- red if you leave the disks in the drive(s) while powering up/down. Sure, there is a real slight chance that a pulse will go thru the write head, but that chance, in my opinion, is rather slim. Oh, BTW, I was wondering if write protect is harware or software enforced. I had the ("lowly") Timex/Sinclair 2068 and something called a MikroDrive for it (that used special little continuous loop tapes). I was feeling a little adventurous (sp?) one day while fooling around with the port. I figured that outputting 255 would set all the bits in that particular register, thereby activating the drive motor (the desired action). HOWEVER, that also turned on the write bit too. Control was somewhat like the cassette port, in that you were responsible for generating the written bit pattern by toggling the status of that bit on that port. Much to my dismay, I had a completely blank tape after doing what I wanted. The write protection was strictly managed in software by reading the status of another port, which had a bit of that port connected to a lever switch. Without trashing one of my disks, is it possible to program the disk DMA, start the motor, etc. and write the disk with the tab on "protect?" I am pretty sure on IBM-PC drives, the controller chip refuses to do it and puts a write-protect into its status register, thereby making write-protect more-or-less hardware enforced. The only way you could do it is to change the firmware or microprogramming of the controller. What say you in netland?
Sullivan@cup.portal.com (sullivan - segall) (04/15/89)
With all of this diatribe about leaving computers and other equipment on
permanently, I'd like to settle it down with a little real information.
After working in the R&D division of North American Phillips Lighting
for half a year, and burning out more than a thousand bulbs (we were
working with quartz halogen, but the principle still applies) I can make
the following generalization.
Bulbs which are turned on for 45 minutes and then off for fifteen in
a constantly repeating cycle last about 75% as long as bulbs which are
left on constantly. This is called a 75% duty cycle. At a 25% duty
cycle bulbs last about 130% as long as bulbs left constantly on. AT
a 10% duty cycle they last about three times as long.
What all of this research goes to show is that what burns out a light
bulb is how long it is left on, with some attention paid to how many
times it is power cycled. But what this means for the world of electronics
I couldn't say.
-Sullivan Segallnew@udel.EDU (Darren New) (04/16/89)
In article <5277@cs.Buffalo.EDU> ugkamins@sunybcs.UUCP (John Kaminski) writes: >In article <13123@louie.udel.EDU> new@udel.EDU (Darren New) writes: >>I have heard, as an unsupported rumor, that the Amiga video chips >>may incur damage if the video is turned off while the computer >>is on. > >Slight chance of a voltage spike coming thru the monitor cabling. I suppose What I had meant to express is "... if the video is off while the computer is on." That is, turning off the monitor and then turning on the computer could damage the chips for some reason like unbalanced loads, unsinked sources, or something like that, and not just because of the transient. >I think it is about the same risk that is incur- >red if you leave the disks in the drive(s) while powering up/down. Sure, there >is a real slight chance that a pulse will go thru the write head, but that >chance, in my opinion, is rather slim. Regarding this, I saw in a magazine where a drive manufacturer was asked this question and responded that the probability of damaging a 3.5" floppy while removing it is about 1 in 2000 and the prob of a voltage spike trashing data because you powered off is about 1 in 1000. It seems that second number would be very dependent on the machine the drive was in, tho. - Miscellaneous rememberences without references from -- Darren New
bryan@intvax.UUCP (Jon R Bryan) (04/18/89)
In article <617@lafcol.UUCP>, ciriello@lafcol.UUCP (Patrick Ciriello II) writes: > Would be to look at all the BBS's in the world that operate 24 hours a > day, 7 days a week, all year round. I have never seen a computer die of > overheating ( except when obvious abuse is involved ). My experience with data acquisition systems that run 24 hours a day has been that things _do_ break. Power supplies seem to be the weak link, and I've had computers fail, counters, you name it. And of course when there are moving parts as well ...
andy@cbmvax.UUCP (Andy Finkel) (04/18/89)
In article <5277@cs.Buffalo.EDU> ugkamins@sunybcs.UUCP (John Kaminski) writes: >Oh, BTW, I was wondering if write protect is harware or software enforced. I Hardware. Hardware. All together now, its enforced in hardware. -- andy finkel {uunet|rutgers|amiga}!cbmvax!andy Commodore-Amiga, Inc. - "Stupid of me not to have seen it before." - "Not really, deathgod." Any expressed opinions are mine; but feel free to share. I disclaim all responsibilities, all shapes, all sizes, all colors.
rwallace@vax1.tcd.ie (04/21/89)
In article <7684@phoenix.Princeton.EDU>, rdnelson@phoenix.Princeton.EDU (Roger D Nelson) writes: > What is the net's experience-based wisdom on the effect of leaving a > micro-computer turned on? The shock of turning the power off and on is always the same whereas the wear caused by leaving the power on is proportional to the length of time so the longer the time the computer will be unused for the more worth while it is to turn the machine off. Obviously it's not worth turning a computer off while you go for a cup of coffee whereas it is if you're going for a fortnight's holiday. As far as I know the break-even time is a few days so I leave mine on all the time. > While I have your attention, may I expand the question to include > systems that have a hard drive? And finally, is there a difference in > regard to the question of continuous running for the various types of > computer? At work I use IBM PCs and ATs, and wonder about leaving > them running also. Hard disks use moving parts constantly so the issue is different here. I turn hard disk machines off overnight but leave them on over lunch. Incidentally could any hardware designers out there give thought to the possibility of being able to turn off the hard disk while leaving the computer on? It's not just a matter of saving wear on the electronics; suppose you had a computing job that would take several days to do (all in memory) and you didn't want to leave the hard disk running that long. "To summarize the summary of the summary: people are a problem" Russell Wallace, Trinity College, Dublin rwallace@vax1.tcd.ie
chasse@ziebmef.uucp (Ken Chasse) (04/22/89)
In article <7684@phoenix.Princeton.EDU> rdnelson@phoenix.Princeton.EDU (Roger D Nelson) writes: >What is the net's experience-based wisdom on the effect of leaving a >micro-computer turned on? I have an Amiga 1000 with an external > WEll, I run a BBS and my Amiga has been off 32 odd and 20 odd hours two different occasions, since October 12th, 1988. It was owned by someone who used it about 3 hours a day for about a year before that. My amiga still works. I even spilled coffee in the keyboard. (I washed it OUT with water even. Then let it dry for a few days I was away by sticking it behind the amiga 1000's fan. I had the thing unplugged (no power in the KB while the BBS was up...used two drops per sugar-crystalized key. Wonder how dangerous this is, even if there is no power.) /kc chasse@ziebmef.uccp chass@ecf.toronto.edu Sysop of Sonic Interzone BBS 416-276-0552 1200, 2400, 9600 (HST), 24hrs >Roger Nelson rdnelson@phoenix.princeton.edu
mcr@julie.UUCP (Michael Richardson) (04/22/89)
Russell Wallace writes: >Hard disks use moving parts constantly so the issue is different here. I turn >hard disk machines off overnight but leave them on over lunch. > >Incidentally could any hardware designers out there give thought to the >possibility of being able to turn off the hard disk while leaving the computer >on? It's not just a matter of saving wear on the electronics; suppose you had >a computing job that would take several days to do (all in memory) and you >didn't want to leave the hard disk running that long. I've always wanted to do such a thing. Hard disks wouldn't be a good idea though: at least not for a couple of hours. Days and weeks I would understand. But this could be a question for things like printers, 1/4" tape archives, MODEMs, Bridgeboards, etc.. (And software has to realise that the subsystem was shut down---so it can sync, reinitialise when it comes back up, etc..) Less stuff on overnight when dcron is running the better---less heat, etc. If the the memory allocator realises that a whole bank of ram is unused, turn it off. With PCs this type of thing isn't really that important (the electronics required to do this costs more than replacing the components every ten years), but when Amigas wind up in space stations, interstar spaceships.. :-) (Or become portable) Most of my complaint is the turbo exhaust fan noise, which I can do something about (but not until the store's one year Warranty on the 2000 runs out, another month), but I really wish CA would put one of those two speed fans in... >"To summarize the summary of the summary: people are a problem" >Russell Wallace, Trinity College, Dublin >rwallace@vax1.tcd.ie -- :!mcr!: Michael Richardson Amiga v--------+ UUCP: uunet!attcan!lsuc!nrcaer!julie!mcr | INTERNET mcr@doe.carleton.ca Fido: Michael Richardson @ 1:163/109.10<--+ Alter @ 7:483/109.10
jim@syteke.UUCP (Jim Sanchez) (04/24/89)
There is a basic misconception about the failure modes of hard disks.
When I was last in California I went to a user's group(ATT actually) and
they were having a talk by an outfit that actually repairs hard disks.
Among the many interesting things they told us was that hard disks are
specified with regard to START/STOP cycles and not by hours of spinning.
It turns out that there are two failure modes associated with starting
and stopping. The first, and the one most know about, is the potential
media damage due to the head rubbing before it starts flying. The other
and more subtle is that the torque of the motor during starting deforms
the spindle hole of each platter which eventually causes failure.
What this outfit said is that you should turn off the computer and leave
the disk drive running if you wanted to do something special. Actually,
I turn the monitor off and leave everything else running.
--
Jim Sanchez {sun,hplabs}!sun!sytek!syteke!jim OR
Sytek Brussels mcvax!prlb2!sunbim!syteke!jim