CPU problem?

[blairm]

Hey everyone. I have two questions/problems that I hope someone can help me with. The first is my system itself. When I bought my computer I was told the FSB speed was 800. However, recently I have been looking into the bios and inside the computer and I think it may actually be only 400.
I downloaded the Intel Processor Identification application and ran it. Below are the relevant results from that.

Expected Processor Frequency- the highest speed at which the processor was manufactures to operate: 2.2 GHz

Reported Processor Frequency- the speed at which the processor is running: 2.2 GHz

Expected System Bus- the highest frequency at which the system bus on the processor was manufactured to operate: 400 MHz

Reported System Bus- the actual system bus speed at which the processor is currently operating: 400 MHz

Are FSB and system bus the same thing?

So I am guessing that the sytem that I thought I was buying (800 Mhz FSB) and the one that I actually got are different?
Aside from this fact though, is this the reason why things were fine for several months and now all of a sudden I am having problems? To elaborate on my problem, my system is hanging a lot and varies a lot in the time taken to boot up . One time it could be normal and the next time it could take up to 2 minutes, although the longer times are happening more frequently. As well, it doesn't multitask as well anymore. A lot of times if I close one program, it hangs a lot and kind of superimposes over another open program and I have to go into task manager and end the application or process to close it. Also when the program finally ends, the program name hangs in the task bar for 5-10 seconds and then disappears.

Here are the specs on my computer that I got from the bios and by taking the case off:

Motherboard- Intel D865 Perl
CPU- P4 2.2 Ghz
System Bus Speed- 400 Mhz
System Memory Speed- 266 Mhz
RAM- 512 Mb DDR 333. The memory mode is single channel and is operating in single channel dynamic paging mode. It is in one 512 Mb stick and occupies one slot with three empty ones. The slots are designated as channel A slot 0, channel A slot 1, channel B slot 0, and channel B slot1
Bios- RL86510A.86A.0056.P07
Hard Drive- Samsung 160Gb
Processor Zone temperature- 34'C/95'F
System Zone 1 temperature -28'C/82'F
System Zone 2 temperature -30'C/86'F
Processor Fan Speed 3655 RPM

You probably don't need all this, but if it helps....Appreciate any help you can give me.

 

[BrentUnitedMem]

Are FSB and system bus the same thing?

No. FSB is the bus between the memory controller and the CPU. System bus usually refers to the bus that connects to the memory modules and other items via I/O bus.

When an application freezes, it is most likely a problem with memory or CPU overheating, check your fan to see if it is working properly. Problems can also be caused by corruption to the hard drive. If the program crashes, or your computer turns off, your best bet is to point at the power supply.

To test for corrupted hard drive and to narrow the problem as hardware specific:
-boot windows from a new hard drive and run the same application.
If it runs fine, you have a hardware problem.
If the application still freezes, you have a hardware problem: either CPU overheating, or memory.

You can test your memory using Memtest86.
Your CPU temperatures can be monitored using MBM5.

My guess is that you have a memory problem. Running Memtest86 should clear up this question.

<font color=green>*****
"Memory with lifetime warranty? So, whose lifetime is that?"
<A HREF="http://www.brentcrowley.com/" target="_new">homepage</A>
<font color=red>AIM BrentUnitedMem

 

[blairm]

I tried running memtest86 from both the floppy and cd but both times it froze up at task 3 and wouldn't respond to any commands. I then ran Microsoft Windows Memory Diagnostic and everything came up error free. I will check the CPU temperatures next. Also, is there an application you can use to test how your power supply is running?
I forgot to mention before that I have been getting warnings about virtual memory when sevearl applications are running. I don't know if this may indicate something?
Thanks for the help.

 

[BrentUnitedMem]

Use MBM5 to log the voltages for your Power supply.

Virtual memory is managed by the operating system, and really has little to nothing to do with your hardware setup.

What memory brand are you using?
If no label on the memory, are there numbers on the chips?

I still susupect the memory... See if you can test your setup with another memory module- maybe you can borrow one?

Also, what OS are you using?

<font color=green>*****
"Memory with lifetime warranty? So, whose lifetime is that?"
<A HREF="http://www.brentcrowley.com/" target="_new">homepage</A>
<font color=red>AIM BrentUnitedMem

 

[Crashman]

First of all, Intel IS using the FSB for the System Bus report. Second Intel never made an 800 bus P4 2.2GHz. Third, the P4 2.2 wasn't even a production item with 800 bus was introduced. What you have is an extremely old CPU, it was introduced 3 or 4 years ago, and has a bus speed that wasn't even fast back then, the 533 bus was current back then, 400 bus was a leftover for upgrades.

<font color=blue>Only a place as big as the internet could be home to a hero as big as Crashman!</font color=blue>
<font color=red>Only a place as big as the internet could be home to an ego as large as Crashman's!</font color=red>

 

[P4Man]

Two things;

First, as crashman pointed out, your cpu is supposed to run at a 400 MHz FSB. Your motherboard, however, is most likely capable of running 800 MHz FSB chips, which is probably how they fooled you. Thats an old dog of a cpu with a fairly good motherboard you have there.. strange mix. Did you buy this new as a system ?

As for the crashing.. my intial thoughts where a virus or dying harddisk. memory errors usually result in spontaneous reboots, not long delays. But memtest86 hanging is odd, especially when the MS tool works fine ? Not quite sure you have a memory problem, but as someone else pointed out, try another memory stick, or try changing memory timings or voltage in the BIOS.

I would still scan for virusses or spyware though. Get AVG antivirus and I also recommend Microsoft Antispyware. Not often I recommend a MS product, but this one is very good.

= The views stated herein are my personal views, and not necessarily the views of my wife. =

 

[blairm]

When I was configuring MBM there were two choices for my mobo (D865PERL), one gave me a reading for the cpu and case and the other gave me a reading for my cpu, case and chipset so this is the one I went with. The results are as follows:
CPU - 34'C
Case - 32'C
Chipset - 32'C
For the power supply I am not sure what is relevant but here are all results
Core 0 - 1.49V
CPU/IO - 1.42V
+3.3 - 3.29V
+5.00 - 5.23V
+12.00 - 12.29V
-5.00 and - 12.00 are both 0.00V
According to the graphs these look alright?

I am using Kingston Ram (333) 512Mb and am using Windows Xp Pro.
I am not sure when I can test another RAM module. I may buy some yet.
Another point I forgot to mention was that my hard drive sometime makes a strange noise when it is booting. It is a mechanical sound and to use an old analogy, sounds similar to the way old lp's used to play (the arm slides over and drops the lp). It sounds like a platter is dropping or something, if that makes sense. it is a quick click and then a drop. I have checked some places on the Internet which said it could be a minor problem and some hard drives are just noisy. I was considering getting a new hard drive anyway. These problems might be some extra incentive. Thanks for the help

 

[blairm]

Yes, this was bought as a new system and yes I think I may have been fooled. I think a call to the store may be in order but anyways...I have Mcafee Internet Security and scan on a regular basis. My hard drive is making strange sounds periodically as I mentioned in my previous post, but I just thought it was noisy. My hard drive is a Samsung 160Gb 7200RPM.

 

[BrentUnitedMem]

A damaged hard drive can cause problems too.

If you suspect your hard drive, the best way to test is to switch the drive. It's always a good idea to have an extra drive lying around around or in your computer for two reasons:
A) You can keep the second drive as a storage medium or to keep program files on it during bootup.
B) In your case, you'd be able to troubleshoot your current drive by replacing it with one that works. If freezing persists, you will know the problem exists elsewhere.

Unfortunately, some troubleshooting requires additional hardware to test.

What is strange in your case is that Memtest86 wont complete. As you know, Memtest86 runs prior to booting from the hard drive. This would point to a problem with the memory, power supply or CPU overheating.

Also, I find it strange that you get a 0-volt reading for your -5 and -12 volt lines. Maybe MSM5 is not configured correctly. Read through the help files in MBM5 to make sure the program is setup correctly.

One other thing I would try...
Clear your CMOS settings. You might be running settings that are too tight for your system which may be causing crashing.
Look in your motherboard manual on how to clear the CMOS settings.



<font color=green>*****
"Memory with lifetime warranty? So, whose lifetime is that?"
<A HREF="http://www.brentcrowley.com/" target="_new">homepage</A>
<font color=red>AIM BrentUnitedMem

 

[BrentUnitedMem]

As Crashman pointed out you have an old CPU which is probably folding under it's current load.

Since memtest86 is freezing, that would most likely eliminate the hard drive as the primary issue. But it could also be a combination of more than one problem; you do hear gremlins running around in your hard drive

<font color=green>*****
"Memory with lifetime warranty? So, whose lifetime is that?"
<A HREF="http://www.brentcrowley.com/" target="_new">homepage</A>
<font color=red>AIM BrentUnitedMem

 

[blairm]

I'll try some of things you mentioned and see what happens. Do the temperatures for the cpu look ok? Is there a maximum temperature for p4's? Thanks

 

[BrentUnitedMem]

MBM5 can log temperatures for you. You need to log them while you are running your applications.

You aren't trying to play DOOM3 are you? hehe

Anyway, you should still be able to run most applications fine with your current setup; I have an 800MHz CPU that runs WINXP, and current office software + accounting applications with no crashes.

<font color=green>*****
"Memory with lifetime warranty? So, whose lifetime is that?"
<A HREF="http://www.brentcrowley.com/" target="_new">homepage</A>
<font color=red>AIM BrentUnitedMem

 

[blairm]

Ha. No, I am not much of a gamer. However, I do do a lot of video editing/conversion, photo vcd's, convert vhs to dvd etc. that have high stream rates and such.

i have been monitoring the temps while running applications and they range from 35-40'C.

I can run pretty much all of my apps, it is just the darn delays and freezes and occasional lockups that get annoying.
Thanks for all your help.

 

[slvr_phoenix]

I can run pretty much all of my apps, it is just the darn delays and freezes and occasional lockups that get annoying.

My instincts say to watch your power supply. When things get undervolted (RAM, hard drive) they act funny. So if you can, monitor your voltages over a period of time (hopefully until a lockup) and then see if anything 3.3v, 5v, or 12v dip below spec in tandem with the weirdness.

<pre><font color=purple><i>Jesters do oft prove prophets.</i> -Regan in
King Lear (Act V, Scene iii) by William Shakespear</font color=purple></pre><p>@ 187K -> 200,000 miles or bust!

 

[blairm]

Hey Guys,
I ran a hard drive diagnostic and from the report saw no errors and the status for the different areas being ok. I ran spybot and did find a couple of cookie entries and removed them.

I did some more looking inside and the power supply is a 350W Pentium 4 power supply. I don't see a specific company on it, but the case is a Vision Pentium 4 ATX case so I assume the power supply came with the case?

I also wanted to check my CPU to get some numbers to look for some info. After I removed the clips and went to remove the heatsink, I found a lot of resistance. I continued with even pressure not forcing it and the heatsink came out with the processor stuck to it. The processor came right out of the socket and stayed attached to the heatsink. I know about the theremal interface material Intel provides to go between the heatsink and processor, but this substance was quite "gummy" and it seems the processor was stuck to the heatsink. I don't know if this is normal, I can't see how it would be, but I was wondering if this could be some indicator to my processor overheating at one point and somehow "melting" the substance and sticking. I took a picture of it in case it could help. If you would like to see it and think it may help just let me know. Thanks

 

[slvr_phoenix]

Intel's thermal interface material is a phase change substance that more or less is a solid that melts when it gets hot. AMD prefers the same stuff because it's nice and clean to pre-attach it to the stock HSF and makes installation easy as hell. The downside is that on occasion some people have reported your problem, that it kind of epoxies the CPU to the HS. :O Of course sometimes the same thing happens with thermal pastes when they dry up. (Especially stuff like Arctic Silver that 'sets'.)

At least this is what I've read and heard from other people. I've never run into it myself.

But they say that usually if you're careful with a good knife you can widge it between the HS and the CPU and pry them apart. Some people report it working even better if you drop the stuck components into your freezer first. Just be careful not to use force between the CPU's PCB and the HS to pull them apart, as there's a small possibility that the CPU's protective cover can come unglued instead of the CPU-to-HS bond breaking. That's why it's better to pry them apart with a knife between the cover and the HS.

So I've heard anyway.

As for your power supply, it sounds like a rebadged generic PoS. It probably did come with the case. It'd be my first suspect. All components are powered by it. It's like building a house. Give it a bad foundation and god only knows what will go wrong over time. So were it me, I'd probably replace it with a good dependable PSU just on principle alone, and if problems still continued I'd know that at least it isn't my power supply. **ROFL**. But of course that's just my advice. It can be expensive to replace components, so maybe you'll want more proof of the actual culprit first and replace that only.

<pre><font color=purple><i>Jesters do oft prove prophets.</i> -Regan in
King Lear (Act V, Scene iii) by William Shakespear</font color=purple></pre><p>@ 187K -> 200,000 miles or bust!

 

[slvr_phoenix]

i have been monitoring the temps while running applications and they range from 35-40'C.

Oh, and your temps sound fine. I doubt that you're having overheating problems.

Unless, of course, what is overheating is your PSU. With a good exhaust fan on the case, a PSU fan can fail and yet the rest of the system can be cooled well. Is your power supply's fan still running?

<pre><font color=purple><i>Jesters do oft prove prophets.</i> -Regan in
King Lear (Act V, Scene iii) by William Shakespear</font color=purple></pre><p>@ 187K -> 200,000 miles or bust!

 

[blairm]

Thanks for all ther responses. I did manage to seperate the CPU and the HS. I have put them back together now and nack in my machine and they seem to be no worse for the wear.

I put my RAM in another slot to see any difference and memtest still froze up at the same spot.

The fan on my power supply is working fine. I was wondering why in MBM5 tha I am getting a 0.00V reading for -5.00 and -12.00. Is this normal?

 

[slvr_phoenix]

I was wondering why in MBM5 tha I am getting a 0.00V reading for -5.00 and -12.00. Is this normal?

It happens quite a bit in MBM5. I'm not sure exactly why, and there may be a way to fix it in the settings somewhere, or there may not be. **shrug** It's usually nothing to worry about though since the negative rails are hardly ever a concern.

<pre><font color=purple><i>Jesters do oft prove prophets.</i> -Regan in
King Lear (Act V, Scene iii) by William Shakespear</font color=purple></pre><p>@ 187K -> 200,000 miles or bust!

 

[apesoccer]

It sounds like you're having problems all over your system...

The 'clicking' noise you're hearing from your hd, normally i would say is a good indicator of your problem. In addition, since you're having problems with other things (memory...), i would look pretty hard at your psu. If your power supply isn't working properly then it can cause problems with any number of components in your system. As a warning, if a bad psu is left in a system (that is supplying inconstant voltages to components) it can eventually destroy those components.

This can end up being pretty costly overall...since nobody here can reliably tell you it's <i>definately</i> one thing over another. This is where you will save money by taking it to a tech (or using your warrenty if it's still good) to let them try to fix it. They should have all of the components in shop to try testing these different problems. Keep in mind though, if this was caused by some sort of power flicker/voltage problem/psu issue, that any number of your components <i>could</i> wear out and begin to cause similar problems because of the stress put on them by the issues described above in the upcoming year or so.

F@H:
AMD: [64 3500+ down][64 3000+][2500+][2000+][1.3x2 down][366]
Intel: [X 3.0x3][P4 3.0x2][P4 2.4x5 down][P4 1.4]

"...and i'm not gay" RX8 -Greatest Quote of ALL Time

 

[blairm]

Hey everyone,
I think I agree you are saying about the power supply.I was thinking about doing an upgrade bit by bit anyways so I will eventually come to some sort of conclusion as I go.

I am going to start with the power supply. I am going to look on the web tonight but would appreciate any suggestions on things to look for such as specs and brands when it comes to buying one. Thanks

 

[endyen]

Your 2 biggest problems are memory band width, and too much shite on your start menu.
If you want your old speed back, try a re-install.
If you want a little faster, add another stick of the same type of ram, for dual channel.
You biggest benefit would be from getting a P4c chip, with dual channel ram.

 

[P4Man]

I'm probably the only one here, but I don't think its your PSU. A faulty PSU can cause reboots, instability, but doesn't cause slowdowns, and won't make memtest lock up at same spot each time. Furthermore, your voltages look perfect. CPU overheating is another issue that is *not* causing your issues. First, your temps are fine, second, P4s will throttle when overheating, but that doesn't explain any of the effects you are seeing.

So what IS your problem ? Sound like you may have more than one, but my biggest concern is the harddisk. I'm not sure what utility you ran, but did it test S.M.A.R.T ? Do you have SMART enabled in your bios (if it supports it ?). SMART is not fail proof, but it does really help in detecting hard drive problems before the drive dies. But even if SMART doesn't detect anything, I'd be very warry, get a new drive, and move your data to it.

The memtest issue.. I've never seen memtest lockup, even with bad memory modules or extreme overclocks. It will usually either report an error, or reboot the machine. are you sure you are waiting long enough ? Some test are really slow. Either way, while there might be an issue here or not, once more, it doesn't explain the behaviour you are seeing. Bad memory will also reboot the machine, cause exceptions or blue screens, but not slow response.

I saw you ran spybot, but you might also want to run a registry cleaner. I tried tweaknow regclear, and I was very impressed with the speedup I saw, but the program isnt free (got a serial from some porn ridden keygen site). Maybe someone can point you (and me) to a free alternative ?

Anyway, in short, I'd ignore the memtest issue for now. If windows memory diagnostics doesn't report anything, any problem you might have is not likely huge. Ignore the PSU and CPU, both seem fine (though a slow pcu). Focus on your harddisk and windows setup. Once that works out, have a closer look at the memtest thing, once you figured that out, consider what you'll upgrade next.

= The views stated herein are my personal views, and not necessarily the views of my wife. =

 

[slvr_phoenix]

A faulty PSU can cause reboots, instability, but doesn't cause slowdowns

I hate to tell you this, but you're <i>really</i> wrong on this one. I've seen a faulty PSU cause slowdowns on a number of systems.

When hard drives get slightly undervolted. ATA, like so many communications standards, is designed to repeat a transmission on a miscommunication. So when the hard drive is sending weak signals and has to repeat itself a lot, this causes a significant and noticable slowdown. Ditto for ethernet, especially with Windows kernels, and also for memory. Hell, really, for pretty much everything in a PC these days. The more PCs have been designed to handle bad transmissions by repeating the communication, the more a bad PSU can cause slowdowns without bringing the system to a halt.

While I agree with you that the voltages don't look <i>low</i>, what is strange is that the 5V line is almost dangerously <i>over</i>volted. 5.23V is only .02V from the 5.25V allowed maximum variation in the ATX spec, which itself is a bit scary IMHO. So it's entirely possible that the power supply has burned or is burning something out.

What is also possible is that the power supply is fluxuating, which is why we've been pushing for a log of the voltages over time, preferable in association to adverse behavior. Fluxuations would certainly explain a lot. But even without any fluxuations occuring, the overvolting is a little worrisome.

And while possible, it seems highly unlikely that it's the hard drive because:
1) The drive already passed a test and by the description of the noise there probably would have been head clashes that damaged a platter were that really the problem, so it shouldn't have passed any tests if it were failing.
2) The memtest failure couldn't be caused by a failing hard drive, but memtest flat out failing like that would indicate a serious problem.

So my first concern is the PSU. My second concern is what you've already sagely suggested earlier, to nudge the VDIMM up a little.

<pre><font color=purple><i>Jesters do oft prove prophets.</i> -Regan in
King Lear (Act V, Scene iii) by William Shakespear</font color=purple></pre><p>@ 187K -> 200,000 miles or bust!

 

[P4Man]

>I hate to tell you this, but you're really wrong on this one.

Why would you hate it ? Nah, quite on the contrary, you'd just love to prove me wrong on something, anything, .. it would be a shame for the original poster however, if you abuse this thread to try and do just that, creating alternative universes again to prove the unprovable.

>When hard drives get slightly undervolted. ATA, like so many
> communications standards, is designed to repeat a
>transmission on a miscommunication.

Thats the first time I ever heard of that. Got a link for me ? I know there is an ECC algoritm in ATA, but that wouldn't cause (noticable) slowdown, and no retransmission, it works pretty much transparently like ECC RAM. In fact, I've never heard ATA would retransmit under any circumstance. If it where as fault tolerant as you seem to suggest, bad ATA cables or too high overclocks wouldn't cause the HD corruption that is seen so often.

>While I agree with you that the voltages don't look low,
>what is strange is that the 5V line is almost dangerously
>overvolted.

Considering the error in measuring such values, these seems perfectly healthy to me. Further more, harddisks are powered by the 12v rail afaik. 12.29V is a-ok.

>So it's entirely possible that the power supply has burned
>or is burning something out.

Nothing would burn at those voltages.

>And while possible, it seems highly unlikely that it's the
>hard drive because:
>1) The drive already passed a test

Yes, but surely you remember deskstar 60XPs ? they would pass any test when the click of death began, IBM would go as far as saying the sound was normal, but a couple of months later, the drive was toast. I haven't heard his disk, so I don't know if it really has a problem (some disks make odd noises without it being troublesome) but if a ball bearing is wearing out for instance, no diagnose pogram will pick it up until it causes the disk to fail.

BTW, I assume he just ran a surface scanner, which would certainly not detect these kind of errors, and will keep claiming the disk is okay until the day it fails to spin up.

> and by the description of
> the noise there probably would have been head clashes that > damaged a platter were that really the problem,

Head clashes ? Damaged platter ? What on earth makes you think that ? A head crash will kill the drive, or at least data, and would have been picked up even by a surface scan. Highly unlikely that happens every time he boots. Doesn't mean his disk is okay though.

>so it shouldn't have passed any tests if it were failing.

well there are so many different kinds of failing, other than head crashes. Worn out ball bearing is another common one, but it could be *anything*. Might as well be the mechanism moving the head over the platter.

>2) The memtest failure couldn't be caused by a failing hard
>drive,

Agreed, but since none of his symptons seem to indicate a ram error, and MS memory diag test completed without error, I don't believe bad ram is his real problem either. Might be an incompatibility between memtest and his motherboard or videocard as well for all I know. And even if the memtest failing is correct, he cearly has another issue as bad ram doesn't cause long delays in windows either.

>So my first concern is the PSU. My second concern is what
>you've already sagely suggested earlier, to nudge the VDIMM
>up a little.

Neither makes sense to me. One or the other *might* cure memtest failing (but I doubt it already), but neither will cure his slowdowns or HD clicking sounds.

= The views stated herein are my personal views, and not necessarily the views of my wife. =