(in my case, it's a palomino) when you have to put your processor out to pasture? I guess what I mean is this... When a processor becomes old and mature, does it tend to show signs of fatigue, huge temperature swings, irratability, etc., just not as productive as they once were in general? Mine is beginning to show these signs here just lately. It's 7 years old.

YO

Just a guess, but I would say NO. There are plenty of people still running old 486 machines & early Pentiums - if your theory was correct, their machines would have withered away a long time before your Palomino.

Asus A7N8X-X
1800+ @ 8 x 210MHz
512MB PC3200
Asus Ti4200 128MB
WinME/WinXP Pro

Jam,

I have tried everything I know of to keep the temps down to no avail. I have used every bios update around, just reinstalled the cpu 3 weeks ago with new white grease, proccessor fan (volcano6) runs at 5400RPM, intake and blowhole run at 2600RPM, and everything is clean. Have just recently noticed the inability to OC fsb. It crashes everytime when doing nothing, IMO, if I try too. It has even started to crash while running folding with no OC. And the heatsink is hot to the touch. It was running at 54-55 degrees c while running folding and now it's going up to 60-64 degrees c. And I don't know what to try next, other than buy a new cpu.

Also, do the mobo drivers have anything to do with heat?

TIA!

YO

'Also, do the mobo drivers have anything to do with heat?'

Not sure?But I do know that when I installed the drivers for my soundblaster 5.1 card,the temps rose by 10c!(In my old system).

Tt Lanfire
nf7-s v2.0
XP-m 2500@209x11
SP 97
512mb pc3200
Jou Jye 550w psu
FX5600
WDCaviar 160gb sata
WDCaviar 160gb sata

;~}

I'd be looking at the motherboard first. Some develop voltage regulation problems and exhibit symptoms like this.

Skip

The answer to the original question is "yes".

Metal migration- micro circuit in a PC may be subjected to different environmental elements; humidity; temp variation (hot&cold). The effects are many that may cause more leakage.

lazyman, until a threshold is reached the device works like always. After that it stops working.
It either works or it does not work. If CPUs die, they die in one shot, not in an agonising way.
having said that I know of a case where a 166MHZ MMX o/c to 233 MHz had its cache burnt off (the reports showed no on chip cache) but the cpu itself was running - though slow. I dont know how much of this is true, but i'd say the reasons for the CPU "feeling" slow in many cases will the OS that the user has not taken care of, fragmentation, too many junk applns etc.

Good Luck and Happy Computing,
Kailas Shastry,

2000+ u/c to 1700+, No thermal paste :
DDR 256MB
10.2GB Samsung SV1022D

Report Offensive Follow Up For Removal

@: Kailas

If you say so,:)
I only involved in micro Chip production for 10 years and might not know as much.

So instead of the works/doesn't work thing we're used to there is a grey area (purgatory?) where a processor can go on the way to hell.

Makes sense to me...might explain some weird things.

Skip

I've always gone by the "works/doesn't work" theory...I didn't think there was a grey area.

lazyman,

Is this "Metal migration" that you mentioned the same as Electromigration?

http://www.hardocp.com/article.html?art=MjMw

Asus A7N8X-X
1800+ @ 8 x 210MHz
512MB PC3200
Asus Ti4200 128MB
WinME/WinXP Pro

jam, me too; and to be completely honest, I never gave it a second thought. I know the physical properties of metals can be changed by extremes of heat, cold, chemicals, atmosphere and so forth and it stands to reason that silicon, aluminum, copper would follow suit.

Gotta do some research...I might learn something today.

Skip

Geez! Dudes I didn't know that I was going to get this much attention today while I was at work. lol! But I have come to the conclusion that yes it is not that it works or it don't work kind of answer on all processors. Some do die a slow death IMO. The EE that I work for claims that the metal leads in the processor can become smaller as time and heat wear on them until the leads blow out. Now my question to him tomorrow will be if one lead or trace burns into, is it still possible for the processor to work, possibly causing my heating problems? Or how about this question, if one transister inside the cpu blows, will it knock out a whole bank of transisisters but still allow the processor to continue to operate? And maybe this one, would the bios recognize that something wasn't right, and report to windows that the processor was only running at 1.6ghz instead of 1.73ghz? Hmmm!!! Lots of questions in my mind about this scenario.

The EE also told me to check out my mobo tonight and see if anything looks suspicious too. Especially caps.

Thank you for all of your responses. This is a great board with great minds too. You just can't ask for much more then that IMO.

YO

Hmmm!!! Just noticed that at some point in time today the cpu temp popped up to 66 degrees C using MBM5. Maybe it won't be such a slow death after all. I will let you know when it passes away. lol! I will never buy another one of them rascals again.

YO

"When a processor becomes old and mature, does it tend to show signs of fatigue, huge temperature swings, irratability, etc., just not as productive as they once were in general"

In theory yes, but in practice no. Its like someone calls you at work and says you were robbed, and you assume that he took a cannon and blew a hole into your house's brick walls in order to get in. In theory one could do this, but in practice nobody breaks into a house using a cannon as there are dozens of easier ways in. Similarly, cpus virtually never fail, especially compared to the rest of the system, which is weak. Dust causes heat, power supplies weaken over time, mobo or psu capacitors can start to leak. CPUs are incredibly reliable. As for aging, migration and lead thinning and the like happen only when the system is run for extended periods on the threshold of stability, like if you ran that palomino at 2.0V and 70C for a couple years.

Moreover, system sluggishness is never a sign of a failing digital component, at least in a home PC. One bad bit and you hang.

The real question here is your temps. Assuming the temps are accurate (odds are in favor that they are not - a bios update may be in order), you may have a failing regulator on the mobo or psu to the end that the cpu is being sporadically overvolted. MBM5's cpu vcore reading will tell you if that is so.

johnoh,

Nice to here from you! I had been wondering what happened to you. Glad you're back.

I'm still puzzled by the operation of this PC. I switched back to an older version of the bios one or two days ago and thought it was doing better but it wasn't. So last night I reflashed the bios with the new version (F-7) and it seems to be doing a little better. Current temps are 51 & 32 degrees C (cpu to case). And it hasn't locked up yet either, which is great news to me..

I would like to know a lil' bit more about that....."One bad bit and you hang." thingy.

Processor bit's or memory bit's?

Should I set my bios to not shut down on any errors? Just wondering.

YO

From what I read the modern day processor can run for 50 years without any signs of what you are going through. Unless you overclocked it at one point and time to such an extent that it shorten the lifespan of that processor considerably, since older processor didn't have any thermal heat protection on them of any kind.

Or if you overclocked you could have did some wear and tare to your motherboard over the years and shorten that its lifespan especially motherboards back then when they wern't very overclockable for the most part and wern't made to overclock unlike 90% of the motherboards today that are made to overclock.

I dunno though have you tried getting another motherbord of the same kind to see if it does the same thing?

I once had a p1 166mhz mmx system and for some reason (didn't overclock it) after 5 years I had it i noticed alot of odd things where happening similair to what you are going through. I tried cleaning out the board and still the same problem. So finally I got a new board of the same type and it ran like it was a brand new system.

"one bad bit and you hang" is an exaggeration, since memory, motherboards and cpus all have some degree of error correction or detection built in. However it is not an exaggeration in that since a couple billion bits pass through your cpu in one second, if you have faulty digital circuitry anywhere in your system, the error correction capabilities are going to fall short very quickly, and you'll hang quickly. That's why a CPU oc'd to say 2.12ghz may run fine but try it at 2.13ghz and it freezes at boot up.

As you know a palomino does not oc very well. If yours has recently gotten worse the two probable culprits are temps and power supply. It sounds like a temp problem, but note that power supplies weaken over time. An oc'd cpu draws more current, and your psu may not be able to do so with stability any more.

All systems are to some degree a combination of digital and analog. Your car, your body, your house, your computer. Your mood is mostly analog, operating on a spectrum and moving smoothly up and down that spectrum (for most of us!). A light bulb is mostly digital, being either on or off. Analog systems change gradually and break down gradually and show wear and tear over time, as will your time in the 100-meter dash over the years. But digital systems work fine until one day they just do not work. How many 100W light bulbs do you see glowing at a 50W lumens? It does not happen unless it is about to die.

That's the premise of the one bad bit and you hang comment. And that if your system is sluggish, it is not likely due to something which is almost purely digital, the cpu. Off the top of my head I'd say the hierarchy of least digital things that most degrade performance in a home PC (that is, the ones that change steadily over time the most, slowing you down while doing so) are software (Windows registry, device driver database, spyware/adware/viruses, bios corruption), hard drive fragmentation, hard drive mechanical components, the power supply, capacitors, thermal paste movement, and dust build up.

Windows aside, if you had a system running DOS (which was as digital as an OS got) for several years and the sytem then failed, a can of compressed air and a new power supply are pretty likely to get it back into new condition. assuming the hard drive is okay.

51C for a palomino is normal.

Thanks all!

YO

johns comment about the PSU posibly going bad does seem logical.. I've seen them start to fail and noticed hangs and temp spikes.. I'd look into it..

Shuttle AN35N-Ultra
AMD 2000+ 1gb 2100ddr
Gainward Gforce 5600 Ultra 128ddr
120GB of 7200 HD space
UV Clear Acrylic Blue case
AXIO Aluminum 600watt PSU
Leadtek Winfast Tv2000xp

Just my 2 cents I think that any CPU that has been subjected to a dip or spike in voltage, Excessive heat, or condensation from being cold then started, or overclocking has the possibility to gradually fail. Tempature expecially I believe can do strange things to Solid state components. Micro fractures between insulative materials electronic leakage causing strange electromagnetic behavior.

ASUS A7V8X
AMD XP 2700+ 2.17ghz
768mb ddr 2700
nVidia 128mb FX 5200
WD 80gb SE
DVD R/RW