Paul
Mon Jun 02 13:52:43 PDT 2008
EdwardATeller wrote:
> On May 17, 10:10 pm, EdwardATeller <sorry_no_em...@yahoo.com> wrote:
>> On May 17, 4:31 am, Paul <nos...@needed.com> wrote:
>>
>>
>>
>>> EdwardATeller wrote:
>>>> Not a standard PSU, but I was able to test the box with a PSU from
>>>> another computer.
>>>> Still freezes. I am now experimenting with different memory sticks to
>>>> see if that helps.
>>>> Using the Sony right now for this post, so far so good.
>>> It helps to use a more stressful test, than a light computing load.
>>> For example, memtest86+ from memtest.org, can be used to test the
>>> memory. Usually, this is available in floppy and also in ISO9660
>>> for a CD. The test will run forever, unless you stop it. No errors
>>> are acceptable. I've had two batches of cheap RAM, that started
>>> throwing errors after the warranty on the RAM was expired. The
>>> failed locations in the RAM were consistent from one run of
>>> memtest86+ to the next. (It is also possible for the errors
>>> to be random, when there are memory bus problems.) But the
>>> important message here, is that no errors be reported.
>>> Microsoft also makes a memory tester, free for download.
>>> If you pass that test, then getting a copy of Prime95 from mersenne.org
>>> is a second test. It runs in Windows or in Linux. There are a few
>>> different versions (including one that spawns a test thread per
>>> core on multicore processors). It uses hand coded FFT (fast fourier
>>> transform) code, where the author of the code knows what the answer
>>> of the math calculation should be. It is a pretty sensitive test of
>>> not only the memory, but also the processor and Northbridge.
>>> This version is suitable for even quad core processors, and is a
>>> Windows version. Select the stress test option (rather than joining
>>> GIMPS), and accept the default test conditions for your first test
>>> run. On a mis-adjusted computer, this program will report a problem
>>> in as little as ten seconds. I use a four hour run, error free, as
>>> my current success criterion.
>>>
http://www.mersenne.org/gimps/p95v255a.zip
>>> Sometimes memtest86+ passes, and Prime95 fails. Prime95 tends to be
>>> a more sensitive test, mainly because the computer gets so much
>>> warmer, and there is also more electrical noise from the test. But
>>> you really shouldn't be booting into Windows, if the memory is
>>> extremely bad, which is why testing first with memtest86+ is
>>> recommended as a first step after a memory changeout.
>>> Paul
>> The computer seems really f***ed now. It ended up freezing after
>> writing that. Now it doesn't even consistently POST. Tried a
>> different video card
>> in a different slot, and it still didn't POST. I should take the
>> thing
>> apart and test the components separately. I have another Socket 478
>> mobo,
>> so I can even test the CPU. Sort of a fun puzzle, but I think I'm
>> going to put it on the back burner for now since I have other
>> computers to use.
>>
>> Thanks for the help.
>
> Went to a computer repair store, and the tech suggested that the CPU
> might be overheating. Took the heat sink off, clean off the thermal
> paste, put new paste on, and reseated the heatsink. Seems much better
> now. Had some freezes after doing this, did it again more carefully,
> and now I haven't had a freeze yet. Knock on wood.
Depending on whether Speedfan understands the hardware monitor in your
Sony, you may be able to chart your progress on improving the CPU
temperature.
How well you're doing, is determined by subtracting the room temperature,
from the other readings. These are my current temperature readings.
Room temperature is actually a thermistor I installed, at the intake
vent of the computer. The case temperature is the sensor on the motherboard,
that is not supposed to be next to any hot components. And the CPU temperature
is based on reading the silicon diode on the processor die itself.
Room temp 24C
Case temp 29C
CPU temp 41C (sitting idle)
29C - 24C is 5C. A well cooled computer case should have a 7C or less delta
between case temp and room temp. Too hot a case temp, means the CPU cooler
has no where to dump the heat.
For an absolute CPU temperature, if I didn't know what kind of
processor you had, I'd suggest keeping it below 65C when the CPU
is under load. OK, I'm firing up Prime95 now.
Room temp 24C
Case temp 29C
CPU temp 47C (Prime95 program is running two test threads)
I'm under 65C on the CPU. If I wanted to keep track of how well
my CPU cooler was doing, I subtract 47C - 29C, or a difference of
18C. The TDP rating of my processor is 70W, and in lieu of measuring
the actual power (which I can do, because I have a meter for it),
my Zalman cooler has achieved 18C/70W = 0.26C/W. That number is
called theta_R, and would represent the thermal resistance of the
heatsink compound plus the heatsink cooler to ambient (case) air.
Only a few companies measure and rate their coolers in those units,
but the measured value in this case, is in the right ballpark for
the cooler. Some of the retail coolers that come with
processors, are in the 0.33-0.35C/W ballpark.
So taking deltas, is the best way to track the performance,
whether it is the ability to keep the computer case air cool,
or keep the CPU cooled off. If I slowed my rear exhaust fan
down, I could probably make my CPU temperature a lot
higher than it currently is.
Some prebuilt computers, use one fan to cool both the CPU heatsink,
and push air out the back of the computer. So a lot rests on that
fan working properly, and the fins on the heatsink being clean.
Is your computer now memtest86+ error free, and Prime95 error free ?
That is how you determine it is fixed. Gamers like to run a
3D game, as a final stability check, but you can also use
3DMark2001SE if you don't own any games and aren't interested
in doing so. You'll need properly installed video drivers for
this to work. If you wanted an overnight test, selecting the
"demo loop" in 3DMark, causes the demo to run over and over.
If the computer is still running it the next day, you're in
good shape. I selected this particular version, as it is
only a 40MB download. Some of the newer versions are huge.
http://www.majorgeeks.com/download99.html
Paul