I to had questions and concerns about the temps my cpu was generating and rather than terminating the program I decided to do a bit of research concerning the limitations of my computer and its cpu. You can find out the max operating temp of your cpu by contacting the cpu manufaturers web site. I am running an AMD 64 3700+ and my max operating temp is 70c. Keeping in mind that the onboard measurement facilities are often inaccurate and may report temps that are too low I feel comfortable running any temp under 60c. My cpu temp does not go over 53c under full load. If your cooler is insufficient and the temp exceeds the maximum operating temp, then this does not mean that the cpu is automatically damaged. With AMD cpu's you will usually encounter crashes if the cpu is overheated; but these will go away as soon as the cpu is cooler again. In the long term, running the cpu at a temp that is too high may reduce the cpu life, since an overheated cpu is more prone to electromigration. With P4 cpu's the cpu will turn its speed down automatically when it overheats thus making the system run slower and no damage to the cpu will occur. I had to install a copper/aluminum heatsink to acheive my cooling solution. After doing a little research, I found that BOINC was not my problem but rather the inadequate cpu cooler.

---

Cleaning the inside of your Pc regularly is an important task whether you are currently suffering heat related problems or not. I have a couple of machines running BOINC under Windows and a web server running Linux (I'd have BOINC running on this one too if it had enough RAM).

I would say that I do a cleanup every 4-6 months or so. Open the case take a PCB vacuum and a can of compressed air to the insides. Full details including a photographed walkthrough can be seen here.

---

Hello and Good holliday to everyone:

Well It has been a little over 30 days NONE STOP, and I only reboot to Clear my memmery once a week, Seti_BOINC runs in the background with permision to run at 100% of the time, Yes 100% 24/7, even while I am running other CPU hungrey apps and programsn. ie: While i am Flying in my VAirlines online or running other Job related Graphis. Hear are a few screen shots of the Temps and fanspeeds in the system as Seti is running just click onto the little thumbs to advance to next shot:

http://www.frappr.com/?a=viewphoto&id=2085322&pid=3271512&myphotos=1

I have done this to try and calm down all the Post that I have been reading sence I have returned to running seti last month, I have been away from seti on and off, lol my job lol. Anyway seti is NOT the only app. I run on this rig. that runs the CPU at full, and That is why I do not worry nor do I turn off seti when running other programs. That is What the CPU on your system Board is Made for, to do just that Crunch numbers as fast as it can, NOT to sit in the parking lot looking all nice whith No real action :) I call that Show Boat, All glem noo action :O lol, for all U gamers out there, comeon speak up let the rest of the world know the things our rigs can really do lol, JUST a little word of Becareful, To Keep ALL vints in the case away from the wall or other things that block the holes in the case. and go get a fine Brush and some Dust-off air in a can and take your time cleaning out the Dust-bunnies once every 60 days, BECARFUL

---

Delta Sigma Phi ΔΣΦ

Hi all,

Please note that ANY software, that does work on the CPU, will cause more heat to be produced. So, why are you acting surprised that SETI (a bulk calculation program) does so? Of course your fan will speed up to a higher rpm, it has to compensate the extra heat with more airflow. My CPU farm has laptops in it too, some of them 24/7, and when given the chance to crunch (e.g. not on batteries), they get hotter and their fans make noise. It is a fact of life, and I can live with it.

Please, do not judge the SETI application by what you think it should do. Read what it really does and if you don't like it, stand back for a while and skip up until the good part comes :)

Running a computer hot and beyond specs shortens its life.

In any enterprise environment this would completely ruin life-cycle planning.

Also BOINC is NOT "just like any other program" in its use of the CPU.. it beats the heck out of it by running it full bore all the time.

Few "average users" ,who are solicited to provide free computing time for the paid academics who want this gift, have any real idea just how hard it is on a CPU to run it hot all the time.

Additionally.. this program is intended for "average users" and they simply are NOT going to crawl around and blow out dust bunnies..their machine is gonna be killed by BOINC.

It is a horrible piece of software that requires constant attention in one form or another... it completely fails in its intended purpose.. which is to recruit "average users" to safely and easily aid in some scientific computing.

As soon as we see all of this instruction on the care and feeding of BOINC..including the need to load still more software to control CPU usage.... we have seen failed software. It is utterly failing to meet its stated purpose.

If you are hacker-hobbyist who goes thru computers like a flu patient goes thru Kleenex you will never notice the destruction because you acquire new toys so often.

But in an enterprise environment.. life-cycle planning does not presume hard, hot, use 24/7.

Neither does an "average user" plan on buying a new machine every few months.. and running things hot and heavy 24/7 will definitely destroy a machine far ahead of schedule.

---

I have several "ancient" systems that have been running SETI Classic when it was still going, and have run BOINC after Classic closed down. Classic ran the CPU at full bore too, and I have not had one iota of a problem with these systems.

If the system was built properly in the first place, every system is designed to run at full speed, otherwise, what would be the point of owning the system? People don't buy computers just to simply sit idle the majority of the time. Graphics professionals, servers, database centers all use their CPU at 100% without issue using other software. This is no different.

---

I have several "ancient" systems that have been running SETI Classic when it was still going, and have run BOINC after Classic closed down. Classic ran the CPU at full bore too, and I have not had one iota of a problem with these systems.

If the system was built properly in the first place, every system is designed to run at full speed, otherwise, what would be the point of owning the system? People don't buy computers just to simply sit idle the majority of the time. Graphics professionals, servers, database centers all use their CPU at 100% without issue using other software. This is no different.

No.. CPUs are NOT designed to run at 100% all the time. if you monitor CPU usage on a typical machine.. as I do on about two dozen machines in an enterpise environment.. they are usually operating at far less than 25% of full bore and 100% is rare and an indication that some malware has taken over... or you have opened too many windows at once.

Life cycle calcs do not presume 100% usage at all.. and never did.

Running things full bore all the time shortens lifespan.. it is that simple. ...there is no room for debate... and anecdotal reports from hobbyists simply do not override the laws of physics and engineering.

Additionally "ancient systems" run much slower and much cooler than faster newer ones.. and so the newer machines are more at risk than the old ones.

---

No.. CPUs are NOT designed to run at 100% all the time. if you monitor CPU usage on a typical machine.. as I do on about two dozen machines in an enterpise environment.. they are usually operating at far less than 25% of full bore and 100% is rare and an indication that some malware has taken over... or you have opened too many windows at once.

Quite the contrary, CPUs are designed to run at 100% all the time. Just because the average system doesn't run at 100%, doesn't mean the manufacturer didn't design it to do so. (PS - Simply opening "too many windows" does not increase CPU usage). It is left up to the PC designer to make adequate designs for cooling to allow such usage.

Any attempt to claim CPUs are not designed to run at 100% can easily be refuted by every CPU manufacturer. Running at 100% is not necessarily rare, it simply depends on the application (and I'm not referring to software application - I'm talking about the role of the PC and it's intended usage).

Life cycle calcs do not presume 100% usage at all.. and never did.

Life cycle calcs do not presume 20% or some other arbitrary usage either. Life cycle calcs simply take into account the average use of a PC before it is considered obsolete (and not the typical joke of being obsolete when you open the box). Of course, there are places that cannot afford to upgrade every five years, but five years is the average for most American companies. Five years is plenty of time to run at 100% without issue.

Running things full bore all the time shortens lifespan.. it is that simple. ...there is no room for debate... and anecdotal reports from hobbyists simply do not override the laws of physics and engineering.

Not true. Before you start claiming this, I suggest you get confirmation from Intel or AMD or even VIA first. Otherwise, there is room for debate - even though I'm more than just a hobbyist. And anyone that has actually taken a class in physics and engineering knows this - not just people who like to accuse everyone else of being hobbyists so as to cast off aspersions from themselves.

Additionally "ancient systems" run much slower and much cooler than faster newer ones.. and so the newer machines are more at risk than the old ones.

Quite true, but my point was, that they handled "full bore" with no problem and it hasn't shortened their lifespan. There is no conclusive evidence that it does so on newer systems either. How can there be conclusive evidence if these "newer" systems have not become old enough to stand the test of time? Prematurely claiming that it will definitely shorten lifespan is pure conjecture and not based on definitive fact.

---

No.. CPUs are NOT designed to run at 100% all the time. if you monitor CPU usage on a typical machine.. as I do on about two dozen machines in an enterpise environment.. they are usually operating at far less than 25% of full bore and 100% is rare and an indication that some malware has taken over... or you have opened too many windows at once.

Quite the contrary, CPUs are designed to run at 100% all the time. Just because the average system doesn't run at 100%, doesn't mean the manufacturer didn't design it to do so. (PS - Simply opening "too many windows" does not increase CPU usage). It is left up to the PC designer to make adequate designs for cooling to allow such usage.

[quote]

Any attempt to claim CPUs are not designed to run at 100% can easily be refuted by every CPU manufacturer.

Wrong. Running them at 100% will simply overheat them.. and cooling is a constant issue.

[quote]Life cycle calcs do not presume 100% usage at all.. and never did.

Life cycle calcs do not presume 20% or some other arbitrary usage either. Life cycle calcs simply take into account the average use of a PC before it is considered obsolete (and not the typical joke of being obsolete when you open the box). Of course, there are places that cannot afford to upgrade every five years, but five years is the average for most American companies. Five years is plenty of time to run at 100% without issue.

Wrong...It is not just "obsolete" (as in out of date) but broken or about to break or requiring so much tech time to keep it running it is cheaper to buy new than repair it.

And no.. virtually NO CPU runs at 100% all the time for any purpose.. nor are they intended to. Additionally servers that run hot and heavier than a workstation are provided with extra coolng capacity precisely because heat DOES kill off the machine and because their heavier usage pushes the temps up.

Running things full bore all the time shortens lifespan.. it is that simple. ...there is no room for debate... and anecdotal reports from hobbyists simply do not override the laws of physics and engineering.

Not true. Before you start claiming this, I suggest you get confirmation from Intel or AMD or even VIA first. Otherwise, there is room for debate - even though I'm more than just a hobbyist.

Additionally "ancient systems" run much slower and much cooler than faster newer ones.. and so the newer machines are more at risk than the old ones.

Quite true, but my point was, that they handled "full bore" with no problem and it hasn't shortened their lifespan. There is no conclusive evidence that it does so on newer systems either. How can there be conclusive evidence if these "newer" systems have not become old enough to stand the test of time? Prematurely claiming that it will definitely shorten lifespan is pure conjecture and not based on definitive fact.

You know.. you make a whole lot of utterly false statements.

In nearly every point you make you are not only wrong but present a case that you do not understand what you think you do.

---

NO... CPUs are not designed to run at 100% all the time.

And opening too many windows DOES increase usage.. each window is a program running and it increases CPU usage accordingly.

For crying out loud.. do you really think you can bluff your way through everything?

Every single point you try to make is either wholly incorrect or demonstrates incomplete understanding of the subject.

For example..servers are built with more powerful cooling systems because they are typically under heavier usage and generate more heat that tends to destroy the chip. This example itself puts the lie to your assertion that they are designed to run at 100%. A CPU running at 100% is a sign something is wrong.

---

Wrong. Running them at 100% wil simpl overheat them.. an dcooling is a constant issue.

Simply stating that I am wrong does not do well for your case. You need to back it up with better facts. Yes, cooling is a constant issue, and if the PC manufacturer designed their case well, it will be less of a problem.

Wrong...It is not just "obsolete" (as in out of date) but broken or about to break or requiring so much tech time to keep it running it is cheaper to buy new than repair it.

Only true in some cases, but many companies actually utilize leases on their PCs, so they replace them every five years regardless if they are broken or about to break. The average lease is for about two to five years, and so companies can afford to stay on the current edge without worrying about systems breaking.

Please, try to consider that there are other methods to run an IT department, so don't simply claim information as wrong simply because your organization doesn't do it the way I or others do. Your method isn't definitive or absolute.

And no.. virtually NO CPU runs at 100% all the time for any purpose.. nor are they intended to. Additionally servers that run hot and heavier than a workstation are provided with extra coolng capacity precisely because heat DOES kill off the machine and because their heavier usage pushes the temps up.

So graphics workstations don't run at 100% all the time? Even when rendering large graphic sets? Yes, heat is the enemy and can kill a machine. That is not what I am refuting. What I am refuting is the fact that if the PC was well designed in the first place, regardless if it's a server or workstation, it won't be such an issue. If people become more careful about what they buy, then they won't have to worry about it in the first place.

Further, simply because the average home user's PC doesn't run at 100% all the time, doesn't mean the CPU wasn't designed to do so by the manufacturer. You still haven't (and can't) deny that fact.

---

NO... CPUs are not designed to run at 100% all the time.

I ask you to prove yourself. Provide technical information from the manufacturer that states that CPUs are not designed to run at 100%. The burden of proof lies upon you.

And opening too many windows DOES increase usage.. each window is a program running and it increases CPU usage accordingly.

I don't know what type of programs you run, but I can open over 100 instances of Windows Explorer and it doesn't increase my CPU usage one iota. As long as the program isn't actively doing something, it will not increase usage at all. If it is waiting for user input, then it will not take up CPU usage at all.

Your statement is false.

For crying out loud.. do you really think you can bluff your way through everything?

Do you think you can accuse everyone else of not knowing what they are talking about forever? If you think I'm "bluffing", then prove it. Provide technical details proving me wrong - not just your word. Otherwise, we can toss personal insults back and forth all day and not get anywhere.

Every single point you try to make is either wholly incorrect or demonstrates incomplete understanding of the subject.

Funny, I could say the same thing about you, but I'm trying to remain civil and hold a discussion. I won't resort to such tactics for fear it would show me in a bad light as if I were wrong or something. Perhaps you should do the same?

---

So tell us. .what graphics workstations are in use at 100% 24/7?

Few? Or none? They are in use while they are rendering graphics.

I might also remind you that BOINC is designed for the "average user" and it fials miserably in serving that audience.

The sheer volume of tender-care-and-feeding that is advised in this board demonstrates that more than adequately.

So, to state that it is somehow normal and "okay" to run things overheated for 24/7 on an average machine by an average user is simply a falsehood.. be it mistake or lie.. it is a falsehood.

---

Oh.. and no.. kndegarten "debating tactics" really don't impress me much.

You are quite simply full of it, honey.

---

So tell us. .what graphics workstations are in use at 100% 24/7?

Few? Or none? They are in use while they are rendering graphics.

Now you're avoiding the burden of proof and the actual problem. Many graphics workstations are in use 100% 24/7, depending on their role in an organization.

I might also remind you that BOINC is designed for the "average user" and it fials miserably in serving that audience.

Your opinion is noted. Now you're starting an entirely different discussion. An easy squirm tactic once you've been proven wrong.

The sheer volume of tender-care-and-feeding that is advised in this board demonstrates that more than adequately.

I figure the "sheer volume" was a small percentage, which every piece of software has problems with. I'm not aware of any software package that doesn't have it's share of issues or it's share of users needing technical help. The several dozen or so is small apples compared to the actual installed base of users, so I think the percentage is just fine. But leave it to everyone else to twist the percentages to further their own political parade.

And you're still off topic.

So, to state that it is somehow normal and "okay" to run things overheated for 24/7 on an average machine by and average user is simpyl a falsehood.. be it mistake or lie.. it is a falsehood.

I never said it is OK to run things "overheated". It is not OK to run things overheated at all. Now you're twisting my words. What I said was, if the machine was designed with adequate cooling in the first place, then heat isn't such an issue. Please don't add words that I never said.

---

Trying to wear me out with a never ending supply of false or misleading statements and demanding I somehow prove you wrong time and time again.. as I keep doing.. is a kindergarten debate tactic.

For example.. it is just plain a nonsensical lie to state that opening a window.. which means opening a program.. doesn't use CPU cycles.

To even state that shows you have no interest in facts but are playing a silly little kiddy game with someone who simply outclasses you.

---

I also note you provide zero evidence for your assertions while whining about mine.

You know kiddo.. people who actually know about this stuff will read your words and laugh at ya.

---

Oh.. and no.. kndegarten "debating tactics" really don't impress me much.

You are quite simply full of it, honey.

Sounds to me like a poor attempt to make your arguments relative. You can't refute my logic, so you attack the person in an attempt to discredit them (as evidenced by calling everyone here a hobbyist, a claim to which you have no definitive proof of, but you go on doing so anyway so as to make yourself look better). Sounds like I have you pegged well, my dear.

Oh, and if I'm so full of it, prove it. That's all I ask. Don't just sit there and call me pet names, or throw your "expertise" around, or other insults either. Provide technical proof to your claims and you will immediately make yourself look better. Otherwise, your tactics are relegated to troll-ish levels.

---

Listen child, you don't know what you are talking about and it is also clear you don't know much about evidence, nor debate nor proving points... and so trotting out the same tired lame "debate" as I have seen you do before really doesn't prove much to me except you are a parrot who says the same things every time you are shown up.

This line of nonsensical assertion is something I've seen you do before. It is lame.

I hope you impress yourself.. you don't impress me.

---

Trying to wear me out with a never ending supply of false or misleading statements and demanding I somehow prove you wring time and time again.. as I keep doing.. is a kindergarten debate tactic.

To even state that shows you have no interest in facts but are playing a silly little kiddy game with someone who simply outclasses you.

LOL "Wear you out"? My dear, I'm afraid I don't have to. My words speak for themselves. Your insistent on tossing insults instead of providing proof of your claims. I'm afraid only one of us is engaging in "kindergarten debate tactics", and it isn't me. You are only digging yourself a larger hole and I'm giving you enough rope to hang yourself with.

If you outclass me, then I'm afraid the world's standards for classes has fallen tremendously. Be warned, you are very closely falling into abusing rule number three: No messages intended to annoy or antagonize other people, or to hijack a thread.

For example.. it is just plain a nonsensical lie to state that opening a window.. which means opening a program.. doesn't use CPU cycles.

Windows Explorer is a program. Open it 100 times and see if your CPU usage increases - because mine doesn't. Simply opening a program doesn't increase usage, especially if that program is waiting for user input - it goes idle and doesn't do anything.

---