Hmm....I get higher temps in XP x86 then in Vista x64. Kindda weird i might say.
Printable View
Hmm....I get higher temps in XP x86 then in Vista x64. Kindda weird i might say.
I cannot pass IntelBurnTest 2.3 on my new Core i7:
Motherboard: Asus P6T7 WS Supercomputer
CPU: Intel Core i7 965
Memory: Corsair Dominator 6GB 1866MHz
OS: Windows 7 64bit
Everything is set to 'Auto' in the BIOS. I've tried all sorts of test (Maximum memory, standard memory, etc)... but it always fails after two test. Can anyone point me in the right direction on passing IntelBurnTest. I'm not looking to overclock (just yet).
Thanks in advance!
Set vDIMM to 1.65. Maybe MB is setting it too low. You can also try raising CPU VTT (QPI voltage, or something on Asus, I don't know).Quote:
Originally Posted by vannguyen0
I hope you solve your problem quickly, it's a shame not to OC that rig. :)
You could run Octo-SLI-Hexa-Fire-X setup and play Crysis in 25600x16000 :rofl:
2author
fix your [...] code to terminate spawn process when closing the main window. make different priorities for spawn process not just your hack w/ RMB click on start. fix task memory sizing for spawn process. 933M instead 1024M - isn't it ugly? Have you tested your algos for this poor shell?
I did it to leave room for Windows in order to prevent paging due to background processes trying to run (which you don't want).Quote:
Originally Posted by asmfan
Try upping the QPI/DRAM voltage to 1.5v or higher... the ASUS boards require an insane amount of power to run high speed RAM.Quote:
Originally Posted by vannguyen0
Hi AgentGod,
First, I've used this program pretty much since its inception and it has been very useful. Thank you for creating it and offering it for free; it's a great tool for overclockers! :)
Secondly, I was using it last night to test a new i7 920 D0 and I ran into some odd behavior. I'm not at the computer right now, and it was very late last night while I was doing it, so the details are kind of fuzzy and I will not be able to provide a debug report until later.
I'm using i7 920 on a EVGA X58 board with 6 GB RAM; however, my OC tools are in Windows XP so the behavior I experienced was in 32 bit XP. I booted into Windows with my overclock. I selected 16 threads and first did Standard test, 5 runs. 4 runs showed up stable, then a linpack.exe error popped up right before the 5th one finished, saying the program had an error and quit. The 5th run did not show but the box popped up saying Success!
I then selected 16 threads, Maximum test, 5 runs. It did not start priming the CPU for a VERY long time. It took the memory down to almost nothing, then it seemed like it started to prime, but not 100%. It did not show the first run ever completing. I then received an error saying something like "linpack memory has been relocated" (it was longer than this, but I was very tired).
I tried it once more, only this time it never took memory or burned - instantaneously the Success! message popped up out of nowhere.
I tried it again, the memory went down (over a long time), it took forever, never showed first run completing, and did some priming but never seemed to fully get going. Then I received a cryptic message about a DLL.
I rebooted the computer and tried it again with 16 threads, Maximum stress, 5 runs. It took the memory down, did some priming but again, never seemed to fully prime. It never showed any runs completing. I did not get any error messages or anything - it just kept going forever without completing any runs.
I'm hoping you can help me diagnose what's going on here. I've not experienced this behavior from the program while testing my overclocks before. Normally it will just reboot the computer, or it will fail the test. I'm thinking maybe the memory is not running stable and that's causing the issue here?
Can anyone here pass successfully:
1. the newest version (v2.3)
2. stress level set to "Very High"
3. Threads set to 4 or higher?
System seems to fail no matter with voltage changes... I wonder if it's me, or the program itself. System passes all other stress tests.
The weird part is, if I customize the stress level to something above "Very High" it passes. (Maximum passes, Very High doesn't)
One other person has tested the same settings, and his system also can't pass this test. But 2/2 isn't exactly a good sampling, which is why I need someone else to try it.
Yes I can. Intel i7 920, OC'd to 3.78GHz, Asus P6T deluxe v2, hyperthreading ON.Quote:
Originally Posted by intx
64-bit Windows 7 RTM, stress level 'very high', 32 threads, times to run 5.
I just tested, my system also passes 32 threads... (but it seems to be a lot easier to pass 32, temperatures reflect accordingly)Quote:
Originally Posted by YngDiego
Someone try 4-8 threads, and see if your system passes.
Can anyone here successfully pass the following settings:
1. the newest version (v2.3)
2. stress level set to "Very High" (Not Maximum)
3. Threads set to 4 or 8
1. v2.3Quote:
Originally Posted by intx
2. Very High
3. 8 threads
NO GO!
Not even @ 4GHz
Passed LinX @ same settings (both times)
So far,
IntelBurnTest v2.3 proves to be defective on the following settings: Standard/High/Very High, 4-8 Threads, 64-bit version.
It seems it will always fail this test.
4/4 people have reported failures (including donmarkoni).
Make that 5 out of 5:(Quote:
Originally Posted by intx
I even tried all stock settings, and still failed..
2.3
2.66 GHz
1066 MHz
8 threads
very high/maximum
Now question is , is it the Shell ( so IBT as the program ) or is it the Core (Linpacks from Intel), is it only related to I7 users or other CPu's too ? I can't pass 4 nor 8 threads either on very high
After first fail whacked the volts to 1.25 (from 1.12 which is 24/7 fold stable) , QPI from 1.25 up to 1.45, this doesn't work at all.
Other thing that seems strange is that Prime 20K custom heats up the CPU as much as IBT, that can't be right can it ?
IBT Failed test :
http://users.telenet.be/OAP2/IBTTEST2.jpg
Prime Custom 20K
http://users.telenet.be/OAP2/Primetest.jpg
Look at them temps, they are the same ???? Not lookign for Prime is better or worse here... IBT and LinX always heated up the CPU's far more than Prime, what's going on ? Or is it on I7 the same now ?
use the new LinX 0.6.2 which has the new linpack 10.2, it will stress your CPU harder and it will give higher tempsQuote:
Originally Posted by Leeghoofd
K did a first test with the latest Linx6.02: took max mem, it passed. Then launched IBT : very high memory and only 4 threads and it fails again ( like always at the 2nd iteneration ) Time to get ya gloves on AgentGod, this ain't working !!!
Temps on Linx are indeed 4-5°C higher, definetaly something wrong here...
http://users.telenet.be/OAP2/IBTTEST.jpg
AgentGOD, For your convenience, I have linked a debug file of failed test that should've passed (Standard, 4 threads).
debug file
have some table with the errors and their causes due ?
thanks
I tried to get INtel Burn to run on an 920 X58 rig last night...won't work in XP 32 or V 64. Something about parallel configuration incorrect.
Tried different DL sites and nothing.
Any ideas? Does the user account need a password to run IntelBurn? That's the only thing different about this rig than the other's I've tried.
Have you installed the necesarry software (look in the readme file mate), too bad he doesn't included them in one package
Hey guys. Is 46C highest (36C average) on burn test a good figure, for a passive water setup at stock for Q6600? :shrug:
Thanks :)
for me it is ok, but what is your room temp ?
Room temp is 21CQuote:
Originally Posted by Marvin
@all & agentGOD
is it a confirmed bug already??
i meant :
IBT v2.3 (8 tread) max/ very high stress + i7 + vista 64 bit & win7 64bit...i never pass more than 2x.
i have tried on 3 different mobos T-POWER X58, P6T DELUXE and P6TD DELUXE
all show the same behaviour <--always error :shrug:
or maybe i7 user should stick at 16 tread or more ?
will it fixed in the next release? when?
thx b4