Sign in with
Sign up | Sign in
Your question

Need help bad! BSOD and High Temps...

Last response: in Systems
Share
June 17, 2008 9:43:44 AM

When I run my system doing normal things I have encountered no problems but when I do gaming and run tv on my 3rd screen my processor goes into full load 100% usage. Anyways when that happens i recieve a BSOD that states
"A clock interrupt was not recieved on a secondary proccessor within the allocated time interval."

This is a major bummer because it took me a while to get the board to see the RAM at correct speeds. I had to call OCZ and they guided me through what voltages and timings to set on my mobo. We reached 1020mhz not quite 1066 but what can you do, especially if 1020 is the only option the mobo gives you.

The other issue is that I live in Arizona and it gets really hot here, I'm not a rich guy and cant afford to install AC or have AC so my ambient room temp is 86-90f. In idle my processor runs at 55c and under full load it goes up and down between 65-67c. Could the heat be causing my BSOD Issue or are those acceptable numbers?

The computer works great, no lag and does exactly what i wanted it to do, except when it BSODs every hour or two while doing my gaming/tv watching.

My system setup (my parents bought me the parts as a graduation gift so trust me i'm not rich lol and am really bummed out about this, close to crying... and we all know how it is when grown men are close to that)

Intel Q9450 with stock heatsink
2x OCZ Reaper 4GB DDR2 SDRAM DDR2 1066 (PC2 8500) (8gb 2gbx4)
ABIT IX38 Quad GT
2xPALiT NE/880TSXT302 GeForce 8800GT SONIC 1GB
2x750gb seagates
Ultra Pro 750w Powersupply
Running Vista x64

Please help me out i'm really desperate


Edit: I am running memtest on my new PC as we speak will update with those results.

EDIT2: Also i am checking my bios temps and the CPU is Idleing at 59-60c somtimes even jumping to 61!!!!! Can my temperature reading program be this far off??? Also if thats it's idle wth is it's full load temp!!!!! Is the mobo full of it?

More about : bad bsod high temps

June 17, 2008 11:37:05 AM

Quote:
Edit: I am running memtest on my new PC as we speak will update with those results.

EDIT2: Also i am checking my bios temps and the CPU is Idleing at 59-60c somtimes even jumping to 61!!!!! Can my temperature reading program be this far off??? Also if thats it's idle wth is it's full load temp!!!!! Is the mobo full of it?



@ Edit 1
Great! That was going to be the first thing I recomended to you!

@ Edit 2
Even with your high room temp that sounds very high. Dl coretemp and see what it is telling you under Xp. I also seem to remember that there is a problem with inaccurate temp sensors on Intel's new(er) 45nm CPU's so it could just be reading high. In either case, you should check the mounting of your stock HS. They are known to have mounting issues that lead to overheating.
June 17, 2008 12:53:03 PM

Check if your CPU is overclocked by mistake.

Try reinstalling Windows.
Related resources
June 17, 2008 12:55:41 PM

Bios temps, those during POST, are always quite high, the CPU is working quite hard during boot, don't worry about those.

Real Temp is best for measuring the Q9450 core temps. 65/67 is fine for a Q9450 working hard on the stock cooler in your hot room, and Realtemp will show lower than Coretemp.
http://www.xtremesystems.org/forums/showthread.php?p=2809778

Those error messages are still due to memory issues. Reduce the speed of the memory and set timings to "auto" and see if the errors go away. The memory should be at 1:1 fsb for best performance anyway, not the fastest possible.
June 17, 2008 12:56:10 PM

I assume you are running the latest Bios update for your MB.

1st, Great on runing Memtest86 from bootable disk. If that passes I would also recommend runing Prime 95. - After verifing temps.

2nd, If Bios is reporting close to 60 C at idle (and is correct) I would guess that your core temps are higher. I would download realtemps and/or CPUID (Hardware monitor) and read some of the threads on temps for the 45 nm cpus. With your ambient temp close to 90F, I would recomment an after market cooler - A GOOD ONE. I'm not a fan of push pin HSFs, I prefer the ones with a backing plate.

3rd - How high are your GPU temps. Removing your side panel, How much does it lower your temps.

Added:
BustedSony, Could be wrong; But Bios Load on CPU is only for about 1 min then equivalent to idle. Also (if correct) is always lower than the 4 individual core temps.
June 17, 2008 1:34:08 PM

RetiredChief said:
2nd, If Bios is reporting close to 60 C at idle (and is correct) I would guess that your core temps are higher.
Added:
BustedSony, Could be wrong; But Bios Load on CPU is only for about 1 min then equivalent to idle. Also (if correct) is always lower than the 4 individual core temps.


My Q6600 G0 shows 56 in Bios and drops to 52. Working temperatures in XP on Coretemp are 42 idle to 68 working hard, overclocked with a Tuniq Tower in a warm room like the OP's. In Bios there's no power saving control or speedstep etc. to control CPU performance, even system and CPU fan control is inactive, usually leaving the fans running slowly. They don't ramp up until Windows is booted. So I've never considered BIOS temp readings a reliable indicator of cooling efficiency.
June 17, 2008 6:48:52 PM

I ran memtest86 and the ram passed. But I tried to replicate the same stress levels and the thing BSOD again with the exact same error here is what the microsoft thing says

Problem signature
Problem Event Name: BlueScreen
OS Version: 6.0.6000.2.0.0.256.1
Locale ID: 1033

Files that help describe the problem (some files may no longer be available)
Mini061708-02.dmp
sysdata.xml
Version.txt

View a temporary copy of these files
Warning: If a virus or other security threat caused the problem, opening a copy of the files could harm your computer.

Extra information about the problem
BCCode: 124
BCP1: 0000000000000000
BCP2: FFFFFA8006CF4358
BCP3: 00000000B2000040
BCP4: 0000000000000800
OS Version: 6_0_6000
Service Pack: 0_0
Product: 256_1
Server information: d2e2d2aa-bcd1-416d-af15-a99cff3db7c3


I have now adjusted it to auto for the memory/cpu which brought the ram down to 816 instead of 1020, I will try it again and see if i get the BSOD again.

I have version 12 of the abit bios on my motherboard already which was the stock bios.
June 17, 2008 9:56:29 PM

BustedSony -
I concur in that Bios (CPU temp) at idle is NOT a good indicator of cooling effienciency. What I said is that CPU temp is less than the individual core temps. Not sure (You could be correct) But I am under the the impression that the multiplier ( and fan control ) is a BIOS call made by the CPU. This BIOS call could be made by the operationg system, or the CPU. In reference to HSF speed that can be controlled in bios by "Telling it " to run full speed, which is what I have mine set to.

Realypk.
Down load either realtemp or CPUID, run Prime95 and monitor Your core temps. The CPU should throttle down if temps get to high. I'm leaning more toward a memory problem. On my system I could run memtest (admitted I did a short test is 30 - 45 Mins) and my memory passed, But Ortho prime would lock up within 7 to 8 Minutes. I had to bump by FSB and gMCH votages up by +0.1V. I was not getting BSODs, just orthos would Lock up.

Even if your memory is the problem and you solve it, I still recommend a better HSF Because of your high ambient temp.
June 17, 2008 11:55:50 PM

RetiredChief said:
BustedSony -
I concur in that Bios (CPU temp) at idle is NOT a good indicator of cooling effienciency. What I said is that CPU temp is less than the individual core temps. Not sure (You could be correct) But I am under the the impression that the multiplier ( and fan control ) is a BIOS call made by the CPU. This BIOS call could be made by the operationg system, or the CPU. In reference to HSF speed that can be controlled in bios by "Telling it " to run full speed, which is what I have mine set to.

Realypk.
Down load either realtemp or CPUID, run Prime95 and monitor Your core temps. The CPU should throttle down if temps get to high. I'm leaning more toward a memory problem. On my system I could run memtest (admitted I did a short test is 30 - 45 Mins) and my memory passed, But Ortho prime would lock up within 7 to 8 Minutes. I had to bump by FSB and gMCH votages up by +0.1V. I was not getting BSODs, just orthos would Lock up.

Even if your memory is the problem and you solve it, I still recommend a better HSF Because of your high ambient temp.


I agree and concur with everything RetiredChief says. (Some Bioses don't have fan speed options for actually being in Bios)

I find that an overheated CPU will just cause a hard reboot, if it's allowed to get THAT hot. The problem here really does sound like memory. - Or an overheated/undervolted motherboard chipset component.
June 18, 2008 12:28:03 AM

I found realtemp and cpuid but where is prime95 for vista 64?
June 18, 2008 2:45:11 AM

Ok so i brought my memory speed/voltages down from 1020 to defualt/auto which was 816 but that was still at a 1:1.25 ratio while 1020 is 1:1.50. It worked for around 7 hours before crashing and rebooting. 1:1 would be 667, should i bring the ram down to that and see if it works? Basically at 816 it works allot longer...
June 18, 2008 3:09:27 AM

Since you are not overclocking the CPU, for now I would try the 667 @ 1::1 which would be a 1333 FSB and cpu vs memory would be synchronous.

Once you get your memory stable, then work on getting temps under control. Then you can work on getting your memory up to speed.

For what it is worth, there have been many posts dealing with memory problems on memory above DDR2 800. (PC6400). Apparently some motherboards require alot of tinkering to get the Ram to work stable.

Try doing a search in the motherboard/memory section for PC8500.
June 18, 2008 4:34:03 PM

So lets say i swapped that out for pc6400 would the mobo most likley recognize it immidiatly without any issues and I'd be chugging allong?


Also my core temp at idle is saying

Core#0 at 63C which is staying stable at that
Core#1 at 47C fluctuates a bit up and down
Core#2 at 49C fluctuates a bit up and down
Core#3 at 42C fluctuates a bit up and down


Do you think that core #0 might be broken or somthing if it's staying idle at 63 and not fluctuating?
June 18, 2008 5:09:17 PM

Sounds like a "stuck" sensor for core 0. There have been several posts on the new quads validating this and is not a real problem. Try running a program that loads your CPU and then look at how your temps respond. I think the new version of prime95 will work, not sure as I have a dual core.

Also, Verify what multiplier you are using at idle (CPU-Z). Your temps need to be verified under nomial load - At least cpu loaded to 50%

Still recommend getting a better HSF due to your High ambient (room) temps. Not a big diff at idle, BUT under load can be a big diff.

Before I would switch out memory, I would try to get the PC8500's too work. You could try to bump the FSB voltage and the Gmch voltage up a little (ie increase them both by 0.1V - May not help BUT worth a try.
June 18, 2008 5:51:17 PM

Under 100% load my temps are as follows

Core#0 70-74c
Core#1 62-63c
Core#2 64-66c
Core#3 62-63c


So seeing as the sensor whent up with load it doesnt seem like it's stuck... it's just running hotter on core 0

With cpuz the multiplier is at 6x idle 8x under load.

June 18, 2008 6:05:05 PM

Hello, Have you tried re-seating your heatsink? Your temps are high but as long as you have thermal throttling enabled in bios your chip wont fry, Retired chief is spot on about your ram, you need to use the lowest divider in your bios and leave the memory voltage on auto, Run Orthos grommacs core test for nine hours and this will test your cpu's stability at stock speed, Use the blend test for another nine hours to test your rams stability, Run core temps to log your temps, OCZ memory can cause problems on some mobos and it might be worth considering the latest bios flash if problems persist.

cheers
June 18, 2008 7:12:17 PM

I have reset the heatsink and that seems to be on their real nice and tight.
I will also put the memory on the lowest divider.

If i got some corasair 800mhz ram would i also have to clock it down to 667? And if that is the case why invest in 1066 if it has to run at 667 anyways? It may just be worth it exchanging the ram for another type with newegg?
June 18, 2008 7:49:13 PM

realypk said:
I have reset the heatsink and that seems to be on their real nice and tight.
I will also put the memory on the lowest divider.

If i got some corasair 800mhz ram would i also have to clock it down to 667? And if that is the case why invest in 1066 if it has to run at 667 anyways? It may just be worth it exchanging the ram for another type with newegg?


Run the memory at safe settings UNTIL everything else has been checked for problems, it's hard enough to get everything else right without risky memory settings confusing the issue. In fact when running extreme overclocks the memory is the last thing that should be ramped up.

Faster memory doesn't have to be run at its clock limit. If it's run well within its limit it can also be set to the standard 1.8 or even 1.7 volts so that it will run cooler. Also the latency can be tightened, which will give a bigger boost in performance than running fast with loose latency timings. As a rule always try to run the memory at 1:1, since that gives the widest bandwidth.
June 18, 2008 8:32:42 PM

The OCZ tech guy told me to up my MCH to 1.4 and my ddr voltage to 2.1 and this time i got a different BSOD
"The system encountered an uncorrectable hardware error."

Problem signature
Problem Event Name: BlueScreen
OS Version: 6.0.6000.2.0.0.256.1
Locale ID: 1033

Files that help describe the problem (some files may no longer be available)
Mini061808-02.dmp
sysdata.xml
Version.txt

View a temporary copy of these files
Warning: If a virus or other security threat caused the problem, opening a copy of the files could harm your computer.

Extra information about the problem
BCCode: 124
BCP1: 0000000000000000
BCP2: FFFFFA800714D040
BCP3: 00000000B2000040
BCP4: 0000000000000800
OS Version: 6_0_6000
Service Pack: 0_0
Product: 256_1
Server information: dc47c1f7-ce84-4018-bdda-2d10af6c91a4



I'm turning the memory down to 1:1 which is 667mhz, I'll let you know how that goes now.
June 18, 2008 10:08:29 PM

Turned it down to 1:1 ratio and it did it again.... i'm beginning to think it's really not the memory....

Is there any way it may be that the mobo is bad or perhaps some driver issues?
June 18, 2008 10:40:50 PM

realypk said:
Turned it down to 1:1 ratio and it did it again.... i'm beginning to think it's really not the memory....

Is there any way it may be that the mobo is bad or perhaps some driver issues?


It can be anything else, but have you checked with boot-time memtest, first of all?

It might be an unstable power supply,I haven't heard of "Ultra Pro" myself. Yes it can be motherboard. I don't think it's a software issue since Vista X64 is pretty damn solid with its "Approved" drivers. HOWEVER, Vista is more fussy about having stable hardware than was XP.
June 18, 2008 11:32:07 PM

After almost 5 days of trying to get intouch with Abit i finally got to talk to somone(wonder how big that phone bill will be)

I talked to him he asked me about my setup and said oh well thats why your getting BSOD issues.... our current bios drivers only support the q9300 and below. Their drivers dont yet support the q9450, he suggested i try to find the beta 13 drivers which supposedly will support the chip, he said that if i go beta it voids the warranty though....

Basically he gave me 2 options install the beta 13 version or wait till they officially release the version 13....


Ummnnm WTF???

Never going to buy a Abit mobo again... next time i'm sticking with a good company like Asus or EVGA....

I guess it's off to newegg for an rma....

What mobo would you guys suggest the EVGA 780i SLI or the Asus Striker II? Or would you suggest somthing else? SLI is not a must, but bieng able to run 2 vid cards at pci-e 2.0 x16 and 8 gigs and a q9450 is.



Edit: Decided to say ahh wtf, seeing as newegg has autorized the return of the mobo/memory/cpu I'm going to F arround with it. I installed bios 13 from abit and it did not resolve the issue still crashing.

So at this point I'd really like to pinpoint what the error is so i dont just send everything back to newegg. Is there any way it could be the videocards, i dont see anything wrong with them tbh...? I pretty sure it isnt the PS seeing as everything is running at assigned volatages any other way to check if your PS is good? If i return the CPU, mobo, ram and exchange it for a different set up accept for the cpu will it likley fix the issue i'm having? I guess if the same thing happens with the new set up it will pretty much be pointing to only the vid cards, power supply or me bieng VERY unlucky....
June 19, 2008 3:32:31 AM

First and foremost, BIOS is not a "Driver," it is a bit of firmware that tells the motherboard how to handle things connected to it. In the case of CPUs an updated BIOS may add additional voltages, or support a new type of on-die cache, or add FSB speeds, and/or allow proper labelling of the CPU at boot up. The 9300 and 9450 are both 45nm Penryns with 1333 fsb. I don't see why one would be supported and not the other. Maybe he meant the older Q6600 and other 60 / 90 nm chips are supported but not Penryns?? Anyway your motherboard DOES handle Penryns and should support the ubiquitous Q9450
http://www.uabit.com/index.php?option=com_content&task=view&id=32&Itemid=48&page=1&model=402

If the motherboard was incompatible it would crash immediately after bootup, or during the first activity, or even just not boot. The failure wouldn't happen after 7 hours. Again, check your power supply and memory.
June 19, 2008 3:40:42 AM

Added - Concur with BustedSony, His post came in as I was typing. End Added

Agh - I should have noted that, One of the first things I do BEFORE buying a MB & proc is to verify that the Processor is supported by the MB.

If voltages are on auto there is a good chance that your processor voltage is incorrect. If the processor Id is not in bios the Bios may select incorrect values.

You could try to manually setting CPU voltage to the correct value (You need to verify, but I think around 1.1 to 1.25 might work).
Manually set Bus to 333 and CPU multiplier to 8.
Leave your memory voltage at Manf spec. Set MCH to FSB voltage to 0.1 greater than normal ( Not sure what normal is as I have a gigabyte board and my bios doesn't show the value, I can only increase in steps - sorry). As Busted said set memory to 667 with a 1::1 ratio - until/If you get this to work

Added - Yuk - I looked at the link BustedSony put up, They do not list supported memory/Proc like gigabye does - OR I MISSED IT. End added
June 19, 2008 3:54:19 AM

Than i really dont know what that dork at abit was talking about :( 


I'm just soooo stumped as to what it could be.... I know that the PS is ok.... and memory seems fine i've tested it for hours and hours with no errors...... Also i checked my voltages and the cpu is getting the correct voltages core temp says its at 1.2375v which is within the accepted norms for a q9450 i think


I did mention that it happens when i'm watching TV at the same time while gaming in windowed mode. It seems to happen allot more when i have the tv program running, is that just because it works harder or do you think the usb tv tuner/program could be causing this? Or could it be a sound issue because i'm gaming/using voice comms and watching tv at the same time?

Edit: And yes i meant bios version 13 not drivers woops....
June 19, 2008 4:16:00 AM

realypk said:



I did mention that it happens when i'm watching TV at the same time while gaming in windowed mode. It seems to happen allot more when i have the tv program running, is that just because it works harder or do you think the usb tv tuner/program could be causing this? Or could it be a sound issue because i'm gaming/using voice comms and watching tv at the same time?

Edit: And yes i meant bios version 13 not drivers woops....


Erm, Vista 64 Hauppuage drivers are known to be buggy, is that what's being used? Also the ATI 650 can cause crashes. Does the system crash, ever, when the capture card is NOT being used?!
June 19, 2008 4:19:10 AM

Yea i am using hauppauge drivers, i think it crashed once when i was not using the tv. But i'll let it sit overnight without the tv on full load and see how it does.


If it is the haupuage drivers is there anything i can do to fix it so i can still watch tv? ( i have a very small room and only have enough room for tv if watching it through my computer :(  )


OR if there is no fix for the haupauge usb tv tuner are there any other usb tv tuner cards you would suggest that are not buggy with vista 64(and can still get reception through coax cable from the wall)?
June 19, 2008 9:07:05 PM

Well i left the thing running on full load without the tv running and it froze up once. I think my internet whent out and all 10 progs i had running were connected to the internet when it froze. Now i'm wondering wether it is my computer or the tv tuner :(  and i've only got a few days to figure it out before i have to make a choice (rma or not)

Anything you guys would suggest here? Just keep running the apps and see if it crashes again?
June 19, 2008 9:18:10 PM

Just me or are you running SLI on a X38 board?
June 19, 2008 9:30:02 PM

Nope not running SLI i'm running 3 monitors 2 on one vid card 1 on the other (thats actually why i whent with the x38 seeing as sli only supports up to 2 monitors, i had no need for it and thought the x38 would be a more stable platform)

basically 2 vid cards running seperatly

Anyways i managed to replicate the crash i had yesterday and it was not a BSOD it was where all my monitors just froze up and the computer locked up. And this is while i was not watching tv, just doing multiple game clients(in windowed mode) to bring the load up to 100%

If the computer freezes up like that what does it point to? Still mobo/cpu/ram?
June 20, 2008 12:15:31 AM

At this point it looks like the issue could really lie anywhere. It still seems like it should be a memory error, but everything passed in memtest. I would simply RMA this system and get a new one with different parts (Somedifferent ram for sure, just to make sure)
June 20, 2008 3:32:12 AM

aye i'm rma'ing the CPU the mobo and the ram for completley different set of parts and see how that goes....

the two vid cards i'm keeping cause they were a steal, if it still crashes though than it could only be the Power supply or vid cards and if thats the case i have a long time to rma those
!