Upgraded computer; Now all my games crash!

[Brian R.]

Yes, that is 2/3. If you haven't, try removing the cards as I suggested earlier or forget about the problem with test programs. By the way, what slots are occupied by what cards in your mb?

[antdude]

ALL:

I have a question about the temperature readings from Motherboard Monitor v4.18 and BIOS:

89 degrees(F) for case sensor temperature.
228 degrees(F) for CPU DIODIDE or something like that (what's that?)

Are these normal? I read diodide gets hot. Maybe I am wrong? Maybe my temperature readings are wrong? I heard ASUS motherboards have bad readings. I am all confused.


Brian: I will try the cards idea later.

[This message has been edited by antdude (edited 26 September 2000).]

[antdude]

I think I am getting somewhere!!!!!! Check out my observation so far. I am going to test more like DVD playbacks, run the Linux BurnP6 longer, etc. From what I understand, it looks like under super heavy stress the system will crash. Games didn't stress enough! I haven't had a single game crash within the last two days! What do you guys think so far?

[snipped]

Check the newer post by me for the list.

[This message has been edited by antdude (edited 27 September 2000).]

[Brian R.]

Looks like it is much more prone to lock-up at 242 than when it is cooler. Better cpu cooling might help. Still might be the ps. A different component may still be the weak point...

[antdude]

Yeah, I just read and heard that the temperatures from ASUS mobo are not accurate. Oh well, I will fiddle with the stuff more later since I am at work. Bah! Oh and I updated the test list.

[Brian R.]

Actually, unless you're worried about exceeding the Intel recommended temperature, the accuracy of the temp measurement isn't important. As long as 242 today and 242 tomorrow are the same temperature (whatever that may be), you can make decisisions based on that number relative to other temps you see measured by the same system.

[antdude]

Brian: I noticed if I use Waterfall Pro (software cooler) in Windows, it drops to like 100 degrees (CPU) as noted in the result list. The temperature only goes up when there is a load when Waterfall doesn't issue HLT commands. Obviously, the test programs max it out . hehe.

Anyways, I have a feeling it is a heat problem. I still need to run the burnp6 in Linux longer. 5 minutes is too short. We'll see what happens!

At least I am not getting crashes like the old motherboard! Games seem to be stable.

[This message has been edited by antdude (edited 26 September 2000).]

[antdude]

Here is the screen shot from Motherboard Monitor during two scenes:
http://home.apu.edu/~philpi/temp/MBM-418.gif

If there is something not right, then please let me know and tell me how to fix it.

Note: The temperature in the room was about 78 degrees(F) when capturing the screen shots.

[Brian R.]

Noticed your +3.3 V is reading 3.71 V. Sounds a little high to me. Don't know if this could be a problem, but I don't think it is normal.

[antdude]

Originally posted by Brian R.:
Noticed your +3.3 V is reading 3.71 V. Sounds a little high to me. Don't know if this could be a problem, but I don't think it is normal.

As of this reply, it is at 3.70. Hmmm... Looks suspicious indeed. I need more comments if this needs to be fixed or not.

[antdude]

If you guys really want to know what is on my system: Check http://home.apu.edu/~philpi/temp/09262K.txt ... it gets technical . I hope this will help. Sandra Professional is very cool.

[DuRaNgO]

Antdude,
I know replacing the P.S. might seem unimportant, but when I built my peltier system, it required a certain amount of power so...

I picked up a 300watt P.S. and it turned out to be inferior to my original 250watt, had to buy a higher quality one to get the peltier working at full power

You could always find a friend with a good one (output of 12amps @ 12volts) to try out, let him use yours, He'll have to be a good friend
jim

------------------
P3-700e @ 1052! Check it here!
Abit BE6-2
TwinMos 128mb pc-133
G400 32mb DH
Maxtor 15.3gb 7,200rpm
SB Live!
Winblows 98se & DX7
and 384k DSL!

[antdude]

Also, I found out that the power supply brand is generic. I will try another PS if it is not a heat problem. I never had problems with it since July 1998 on my Pentium II 300 Mhz.

[This message has been edited by antdude (edited 28 September 2000).]

[antdude]

[Deleted the old result list and reposted the updated list]:

NOTES: All CPU temperature is based on on-die diode as noted in CMOS
info and based from Motherboard Monitor v4.18 (assuming the
settings are correct). Room temperature was about 78-80
degrees(F).

Windows 98 Safe Mode Boot with BurnP6 = Hard lockup/Failed
91 degrees(F) Case Sensor
242 degrees(F) CPU Sensor
Took over 5 minutes to crash.
First reset rebooted system, but BIOS didn't show up.
Had to reset again to bring it up.

Windows 98 Regular Boot with BurnP6 = Hard lockup/Failed
91 degrees(F) Case Sensor
242 degrees(F) CPU Sensor
Took about 20-30 seconds to crash.
Reset button worked properly on the first time.

Windows 98 Clean Boot with BurnP6 = Hard lockup/Failed (twice)
91 degrees(F) Case Sensor
242 degrees(F) CPU Sensor
First reset rebooted system, but BIOS didn't show up.
Had to reset again to bring it up.

Windows 98 Regular Boot with BurnP5 = Passed
91 degrees(F) Case Sensor
228-235 degrees(F) CPU Sensor
Ran it for about 10 minutes.

Windows 98 Regular Boot with BurnMMX = Passed
91 degrees(F) Case Sensor
242 degrees(F) CPU Sensor
Ran it for about 15 minutes.

--

Linux Boot Disk with BurnP6 = Failed

1st test:
5 minutes -- No segment faults, lock ups, or any symptoms.
Forgot to check the temperatures. 5 minutes was
not long enough.

2nd test:
55 minutes -- The first 50 minutes survived while I was eating
(not near the computer). I came back and noticed
the computer is still working. I typed ls, ps,
and other basic commands during the stress test.
Finally, it hard locked up without errors when I
typed an invalid command (w doesn't exist in this
Linux boot disk). 1st reset didn't work. Had to
reset again to boot. Checked CMOS and saw: 101
degrees(F) case and 212 degrees for CPU.

--

In Windows Regular Boot:

Q3A v1.17 Demos (1024x768 32bit) for 10 minutes = Passed
Two demos: 91 degrees(F) for case & 217 degrees for CPU

Played UT v428 intro demo (1024x768 32bit) in a loop = Passed
overnight and downloading on a 28800 modem connection
at the same time (about 5.5 hours)
91 degrees(F) for Case & 220 degrees(F) CPU when I
played four games with 7 bots.

Diablo 2 for 1+ hour = Passed
91 degrees(F) for Case & 221 degrees(F) CPU

PowerDVD v2.55 0620 for 15 minutes = Passed
90 degrees (Case) and 160 degrees (CPU)

WinDVD 2000 v2.2 for 1:22 hours = Passed
90 degrees (Case) and 162 degrees (CPU)

--

AleGro MemTest (DOS boot) failed right after the second test. I
noticed about five different memory miswrite and misread until the
second test (random part) and it signaled CPU halted and hard
lock. I think this took about 10 minutes.

--

Idled time, I have seen it: 91 degrees (Case) and 104 degrees (CPU)
with Waterfall Pro (software cooler) in Windows 98. Without it, it
is above 200 degrees(F) idled according to BIOS and MBM readings.

I noticed the CPU temperature never goes higher than 242 degrees(F)
in MBM. Did I reach the limit/peak or something? However, I heard
the CPU sensor from ASUS is not accurate.

[antdude]

Finally (I think)! I removed the thermal pad and put thermal paste/grease on the CPU. Ran burnp6 for about 40 and 30 minutes (two tests) in a regular Windows 98 boot. NO LOCK UPS AND NO SYMPTOMS! The CPU temperature was around 130 degrees(F). The case temperature was about 86 degrees(F) even in a 79 degrees(F) room.

I guess it is time for me to celebrate! YAY! Sheesh, computers are annoying. I am keeping crossing my fingers/feelers that this finally solved all of my computer problems! I hope this thread dies!

[This message has been edited by antdude (edited 01 October 2000).]