still chasing around stability, but getting somewhere...

Batch codes, RAM specs, BIOS settings, etc..
cavity
Posts: 66
Joined: Sun Jun 27, 2004 9:09 pm
Location: Victoria BC, canada
Contact:

still chasing around stability, but getting somewhere...

Post by cavity »

I've been chasing sround stability issues, and i've got it so prime95 will run under windows or mprime under linux for about 12-17 hours before getting a rounding error. this is acceptable with me, but when the computer is warmer during the day i get reading from temp file errors within minutes.
its a bp6 with 2x400@558, and i cant clock it anywhere inbetween here (93) and 66mhz without huge problems.
right now, everything is sitting at 38c. during the day when its unstable its around 42-44c
the cpus have nice big heatsinks, and the chipset has a bigger heatsink and fan than the greenie.
the vtt is stable to within .2v
i think the cpus are around 2.1v or 2.05v

is that temp difference likely to make it unstable because of the clock speed? i havnt found a more stable speed between here and 66mhz.

i cant afford to buy waterblocks like i've always wanted to because im in college right now. and i really dojnt want to be stuck at 66mhz fsb.
any ideas :?:
Last edited by cavity on Wed Dec 08, 2004 1:15 am, edited 1 time in total.
i dont have enough free time anymore... college will do that.
I've learned from experience that hpt sucks
Derek
Site Admin
Posts: 2489
Joined: Tue Jul 23, 2002 3:55 pm
Location: Canada
Contact:

Post by Derek »

cavity wrote:I've been chasing sround stability issues, and i've got it so prime95 will run under windows or mprime under linux for about 12-17 hours before getting a rounding error. this is acceptable with me, but when the computer is warmer during the day i get reading from temp file errors within minutes.
its a bp6 with 2x400@558, and i cant clock it anywhere inbetween here (93) and 66mhz without huge problems.
right now, everything is sitting at 38c. during the day when its unstable its around 42-44c
the cpus have nice big heatsinks, and the chipset has a bigger heatsink and fan than the greenie.
the vtt is stable to within .2v
i think the cpus are around 2.1v or 2.05v

is that temp difference likely to make it unstable because of the clock speed? i havnt found a more stable speed between here and 66mhz.

i cant afford to buy waterblocks like i've always wanted to because im in college right now. and i really dojnt want to be stuck at 66mhz fsb.
any ideas :?:
Check the capacitors. The motherboard is old and some leaking or bulging capacitors could fail with an increase in temperature.
-Derek
cavity
Posts: 66
Joined: Sun Jun 27, 2004 9:09 pm
Location: Victoria BC, canada
Contact:

Post by cavity »

i think i did a quick check a few months ago... but ill have another look
i dont have enough free time anymore... college will do that.
I've learned from experience that hpt sucks
davd_bob
Confused
Posts: 1043
Joined: Fri Feb 13, 2004 2:30 am
Location: Houston, TX

Post by davd_bob »

Is the increased heat because of increased work load or just room temp increasing?

Do you have good air flow throuth the system? You may have a stick of ram or a card(like your video) overheating. The pci/agp clock divider may be getting you. If I calculated right you are running on 92fsb which would put the cards underclocked which should be ok, but if you are at 91fsb the cards would be overclocked and that may be the problem.
There are *almost* no bad BP6s. There are mostly bad caps.

No BP6s remaining
Athlon 2800
Sempron 2000
ViaCPU laptop with Vista.(Works great after bumping ram to 2Gig)
P-III 850@100
cavity
Posts: 66
Joined: Sun Jun 27, 2004 9:09 pm
Location: Victoria BC, canada
Contact:

Post by cavity »

the increased temp is due to both work load and room temp, mostly load though. i have quite good air flow, there are intake ducts on both cpus so they always get cold air, that saved 6 degrees. and there are no sides on the case. i stuck my hand in there just now and it was barely above room temp, but the cpus are at 37 due to an open window and very cold air getting into the intake ducts.
the video card is an ati radeon 9200 with a fan stuck on the stock heatsink,and it feels cool, though i havnt been gaming for a few hours.
Im on 93 fsb, and i dont know anything about pci/agp dividers.
i dont have enough free time anymore... college will do that.
I've learned from experience that hpt sucks
InactiveX
BeOS Forever
Posts: 1385
Joined: Wed Jul 24, 2002 8:25 am
Location: UK

Post by InactiveX »

cavity wrote:Im on 93 fsb, and i dont know anything about pci/agp dividers.
You should be fine with the divider. Put simply, at <92MHz your PCI devices are running at 1/2 the FSB speed. At 92MHz and over, the PCI devices will run at 1/3 of the FSB speed.
cavity wrote:when the computer is warmer during the day i get reading from temp file errors within minutes.
I'm not quite sure what you mean here, but I wonder if your hard disk is getting a little hot. Does it feel overly warm after being run for a while?
Like BP6.com? Not a member?
Then why the hell not? It's great!
-> BP6.com Membership <-
cavity
Posts: 66
Joined: Sun Jun 27, 2004 9:09 pm
Location: Victoria BC, canada
Contact:

Post by cavity »

when the computer is warmer during the day i get reading from temp file errors within minutes.
sorry, i mean prime 95 has read from temp file errors when it gets hot.
The hard drive is indeed a little hot right now, and its been almost idle for a while. i'll see if i can put a small fan next to it.
i dont have enough free time anymore... college will do that.
I've learned from experience that hpt sucks
purrkur
Linux Guru
Posts: 687
Joined: Fri Dec 12, 2003 5:57 pm
Location: Sweden
Contact:

Post by purrkur »

Cavity: You said you were running Linux. That is a good thing :)

What harddrive have you got in there (as in type) and what Linux distro are you using?

Some harddrives (like Seagatess) have temperature sensors inside the harddrive. You can monitor those with a program called "hddtemp" under Linux. It works really well. I use it to monitor all the different Seagate disks I have. Don't know of a Windows equivalent program though since I don't do Windows unless I am forced to :)

Code: Select all

thor ~ # hddtemp /dev/hda
/dev/hda: ST360015A: 39 C
Also, my .02c worth on stability issues on overclocked systems: If you are getting stability issues after overclocking and you are worried/concerned that they might be caused by your overclock then put your CPU's down to their rated speed and rerun your tests. If your tests go through as they should then you know the source of your problems! From there you can work your way up, a step at a time until you reach the stability vs speed that you want.

I also had a pair of 400's before I bought my 533's that I now run at 576MHz. I did overclock my 400's as well but I kept them at 466MHz. I probably could have clocked them higher but at the time I didn't have decent coolers so I decided not to push it. I love overclocking but I love stable systems as well.

And btw, I have seen temps much higher than 44C on my 400's without stability issues....
2x533MHz@544MHz, 2.0V
640MB PC100 memory
Realtek RTL-8139 NIC
Maxtor 6Y080L0 80GB hdd
Debian Linux stable with 2.4.8 kernel
cavity
Posts: 66
Joined: Sun Jun 27, 2004 9:09 pm
Location: Victoria BC, canada
Contact:

Post by cavity »

I'm jumping back and forth between linux and windows so i can do some homework with it.
the hd is a western digital, and it does have a temp sensor. i've used it in windows before.
Im using xandros/debian but to make a long story short, i have no compiler anymore and it says it requires itself to install.
i'll have a look at the hd temp in windows, just because the program for it is already there.
i dont have enough free time anymore... college will do that.
I've learned from experience that hpt sucks
Wolfram
Posts: 401
Joined: Tue Jul 30, 2002 3:19 am
Location: Germany

Post by Wolfram »

If your system was perfectly stable at 66 Mhz and unstable at 67 Mhz FSB, this would be very unusual. I would guess your system is unstable even at 66 Mhz FSB, if it's at 67 Mhz. Did you run prime at 66 Mhz FSB?

But I've heard there were PCI cards (especially sound cards, IIRC) that do not like other PCI speeds than 33 Mhz.

I'm not surprised that your system is unstable at 558 Mhz- your CPUs might just be beyond their maximum stable speed.
BP6, RU BIOS, XP SP3, ACPI, 2x366@523(1,95V), Pentalpha HS + 1x 12cm fan @5V, 768MB, Powercolor Geforce 3, RTL8139D NIC, Terratec EWS64L, Samsung M40 80GB (2,5''), LiteOn CDRW
cavity
Posts: 66
Joined: Sun Jun 27, 2004 9:09 pm
Location: Victoria BC, canada
Contact:

Post by cavity »

im running prime95 at 66mhz right now... we'll see how it does.
i dont have enough free time anymore... college will do that.
I've learned from experience that hpt sucks
cavity
Posts: 66
Joined: Sun Jun 27, 2004 9:09 pm
Location: Victoria BC, canada
Contact:

Post by cavity »

prime95 ran all nite fine, but thismorning after using the computer for a few minutes it came up with a rounding error. 15h 30min.
and this is at 66mhz. time to go to school anyway
i dont have enough free time anymore... college will do that.
I've learned from experience that hpt sucks
purrkur
Linux Guru
Posts: 687
Joined: Fri Dec 12, 2003 5:57 pm
Location: Sweden
Contact:

Post by purrkur »

cavity wrote:prime95 ran all nite fine, but thismorning after using the computer for a few minutes it came up with a rounding error. 15h 30min.
and this is at 66mhz.
So what are you aiming for?? Will any old computer that is not overclocked run Prime95 into all eternity without errors? I would think not. Your BP6 ran under continuous 100% load for 15 hours and 30 minutes without a hitch. What would be considered normal? If there is a "normal" figure out there somewhere then how old was the computer that was used to get that value?

And are you sure that your caps etc are in good working condition??
2x533MHz@544MHz, 2.0V
640MB PC100 memory
Realtek RTL-8139 NIC
Maxtor 6Y080L0 80GB hdd
Debian Linux stable with 2.4.8 kernel
cavity
Posts: 66
Joined: Sun Jun 27, 2004 9:09 pm
Location: Victoria BC, canada
Contact:

Post by cavity »

the 15 hours is fine with me, its when i was getting read from disk errors i was concerned. i think its the hd, and the cpus are just as stable at 558 as 400. I'll do something to keep the hd cooler, but i think i'll overclock it again.

I guess the problem was just the hd getting hot, so it could be considered a solved problem. thanks for the help 8)
i dont have enough free time anymore... college will do that.
I've learned from experience that hpt sucks
Wolfram
Posts: 401
Joined: Tue Jul 30, 2002 3:19 am
Location: Germany

Post by Wolfram »

cavity wrote:the 15 hours is fine with me, its when i was getting read from disk errors i was concerned. i think its the hd, and the cpus are just as stable at 558 as 400. I'll do something to keep the hd cooler, but i think i'll overclock it again.

I guess the problem was just the hd getting hot, so it could be considered a solved problem. thanks for the help 8)
But beware: There are harddrives that don't like an overclocked IDE bus. I once had an IBM DHEA 8 GB, which seemed to run fine with the PCI bus at 41,5 Mhz. After a few days, the whole data on the drive was corrupted. A newer harddrive (20GB WD, IIRC) worked fine.
BP6, RU BIOS, XP SP3, ACPI, 2x366@523(1,95V), Pentalpha HS + 1x 12cm fan @5V, 768MB, Powercolor Geforce 3, RTL8139D NIC, Terratec EWS64L, Samsung M40 80GB (2,5''), LiteOn CDRW
purrkur
Linux Guru
Posts: 687
Joined: Fri Dec 12, 2003 5:57 pm
Location: Sweden
Contact:

Post by purrkur »

Wolfram wrote:But beware: There are harddrives that don't like an overclocked IDE bus. I once had an IBM DHEA 8 GB, which seemed to run fine with the PCI bus at 41,5 Mhz. After a few days, the whole data on the drive was corrupted. A newer harddrive (20GB WD, IIRC) worked fine.
I'll go with Wolfram on this one since this is my experience as well. Certain harddrives are simply not safe in an overclocked computer. Make sure you don't have any data that you can't loose if you intend to overclock it!
2x533MHz@544MHz, 2.0V
640MB PC100 memory
Realtek RTL-8139 NIC
Maxtor 6Y080L0 80GB hdd
Debian Linux stable with 2.4.8 kernel
cavity
Posts: 66
Joined: Sun Jun 27, 2004 9:09 pm
Location: Victoria BC, canada
Contact:

Post by cavity »

i know about overclocking old hard drives, i killed the data on an old 4gig by running it a few mhz over. now i have a fairly new 80gb, so it shouldnt be a problem. but i do backup data once in a while.
i dont have enough free time anymore... college will do that.
I've learned from experience that hpt sucks
cavity
Posts: 66
Joined: Sun Jun 27, 2004 9:09 pm
Location: Victoria BC, canada
Contact:

Post by cavity »

right now the hd is reporting 32c after hours of idling...
i dont have enough free time anymore... college will do that.
I've learned from experience that hpt sucks
Wolfram
Posts: 401
Joined: Tue Jul 30, 2002 3:19 am
Location: Germany

Post by Wolfram »

Looking at your sig, I assume the hd is on the standard IDE, not the HPT366?
BP6, RU BIOS, XP SP3, ACPI, 2x366@523(1,95V), Pentalpha HS + 1x 12cm fan @5V, 768MB, Powercolor Geforce 3, RTL8139D NIC, Terratec EWS64L, Samsung M40 80GB (2,5''), LiteOn CDRW
cavity
Posts: 66
Joined: Sun Jun 27, 2004 9:09 pm
Location: Victoria BC, canada
Contact:

Post by cavity »

yup, i also go the bios without hpt
i dont have enough free time anymore... college will do that.
I've learned from experience that hpt sucks
Dave Rave
G'Day Mate!
Posts: 894
Joined: Thu Sep 26, 2002 4:28 am
Location: Sydney Australia
Contact:

Post by Dave Rave »

(( I haven't read all of this .. so .. ))

a 7200rpm drive is a beast.
it runs hot.
if it has no air flow it will go super-hot !!
yet ...
it takes a crappy little fan running a crappy small amount of air to keep it back to an acceptable temp ...
while the Seagate tech thingies say that an (now) older drive is okay running 60+ degrees (ouch)

I think that if a drive goes over-hot it will go pear shaped and die, without notice due to who knows what problem....
get a copy of HDD to monitor better the SMART reports form the HDD.

(damn, hope that's legible... and intelligible)
cavity
Posts: 66
Joined: Sun Jun 27, 2004 9:09 pm
Location: Victoria BC, canada
Contact:

Post by cavity »

you guessed right, its a 7200, and i've got a fan under it now. with the fan it dropped 10c both under load and idle. at idle its 2c above room temp instead of 12. the highest it gets now is 27 while defragging, and it used to get near 40c.

where's that damn degrees symbol? useless kayboard
i dont have enough free time anymore... college will do that.
I've learned from experience that hpt sucks
hyperspace
Board Admin
Posts: 1395
Joined: Wed Jul 24, 2002 7:39 am
Location: Lincoln, NE USA
Contact:

Post by hyperspace »

&deg (& + deg)

:D
Quantum WormHole

Image
lost in hypertime...
headseed
Posts: 165
Joined: Thu Dec 26, 2002 5:45 pm

Post by headseed »

Not really on topic, but I don't believe that Xandros installs a SMP kernel by default. Might want to check on that.
Dual Barton Mobile 1.8ghz
Venice 3000+ @ 2.6ghz
lots of BP6s and two VP6s all apart currently
cavity
Posts: 66
Joined: Sun Jun 27, 2004 9:09 pm
Location: Victoria BC, canada
Contact:

Post by cavity »

if it wasnt a smp kernel, then wouldnt i see one cpu doing all the work and the other idle?
i dont have enough free time anymore... college will do that.
I've learned from experience that hpt sucks
Post Reply