Intel themselves said to disable HyperThreading before running the program.
Printable View
Intel themselves said to disable HyperThreading before running the program.
Probably because running 4x verse 8x instance of Linpack will cause Linpack to run twice as slow. Since all 8 threads are trying to use the exact same execution units hyperthreading can't improve upon the performance.
I have no plans to disable HT because I will be running virtual machines and loading all eight threads... so I don't care if Linx runs 2x as slow. I need to ensure the system is 100% stable with the exact settings and features I plan to be using. :)
And belatedly, thanks for the update AgentGod! :up:
And for those of you who still care...!
http://img6.imageshack.us/img6/8653/intelburntestv2.jpg
Sneak peak of v2.0 beta! It even has an animation to let you know if your system is frozen or not, AND it outputs results as Linpack runs while it checks for any errors in the results produced. Stress level selection is now also much simpler.
Thanks. This is my favorite stress testing program. Any ETA on the v2.0 release?
I worked on it some more today and found a major bug. Fortunately, I was able to fix it :)
So, which of the following do you guys think should be added?
1. Display total time taken to complete the run at the end.
2. Minimize to task tray.
3. Some sort of Validation system like CPU-Z?
4. Display CPU name and clock speed.
5. More warnings and confirmation dialogs.
6. Changing the name to "CPU CoreBurner" or "CPUBurnTest" to reflect the fact that it has been working on both Intel and AMD processors since v1.7.
or All of the Above/Combination of above items?
Your feedback will be greatly appreciated :)
Thanks for all of your support!
Or anything else not mentioned?
I love this program so much. Keep up the good work.
Added logging feature today in v2.0! It logs output results, date, time, processor name, clock speed, total memory, and more.
Sweet.
Anything else you guys would like to be added prior to release?
Keep the name IntelBurnTest.
There's a program called ATITool and they don't change the name although it works with Nvidia graphic cards too.
I like the new look. Keep up the good work and let's see what you've got so far.:up:
Get ready! Prepare for the release of v2.0 on May 15, 2009!
Nice, i know some people shoot this program down but i love it. Keep up good work :up:
Thanks again! :up:
The Day Is Here!
IntelBurnTest updated to v2.0!
- Completely re-coded from the ground up
- New, intuitive graphical user interface
- Windows Vista/7 Aero friendly
- Real-time result output
- Real-time error checking, auto-stops when error occurs
- Simplified stress level selection
- Easy copy of results to clipboard (right-click context menu)
- Built-in system freeze test (animated flame)
- Real-time free system memory display
- Uncluttered integrated title/status bar
- Added log results to file feature, with date and time, CPU/RAM info
- Displays total time taken to complete test at the end
- Correctly detects the maximum memory limit for x86 Windows
- Fixed a bug that caused testing memory to be limited to 6.1 GB
- New eye-catching icon
- Various bug fixes carried over from v1.93
Download it now via first post!
Enjoy!!!
beautiful work,
but I just to test the CPU?
has tried to open here this error :(
http://img196.imageshack.us/img196/2674/erroq.jpg
You need Visual C++ 2008 Runtime, as stated in the Readme, which can be found here:
http://www.microsoft.com/downloads/d...d-074b9f2bc1bf
New version does not run nearly as hot as 1.9. Not as stressful?
this is what i get:
http://i143.photobucket.com/albums/r...1/itberror.jpg
Great work AgentGod, like it more then first one. It doesn push temp to to high compare to first version......keep it up :up::yepp::cool:
jesus, i had to do a small vcore bump on something that was 36 hours prime small fft's stable to get 10 maximum runs here.
great feeling to have it pass though.
btw, agentGOD, you rock, and i think you made prime obsolete.
Very nice AgentGod as winter came i have 22C inside my room and i was able to do a quick test of your app as this batch is very low vid 1,1875 i have to be careful with temps.
http://img198.imageshack.us/img198/2...coreibtsti.jpg
Cheers :up:
Sergio
Nice chip, I want it :D
lol
Thanks :up:
I tried new IBT 2.0 with an Asus Rampage II Extreme and a Xeon W3520 with HT enabled under Win7 x64 and I was not able to load cpu cores at 100%.
Is there any button/command that I haven't seen or is it an issue of the new program?
Great job, Agent!
AgentGOD, i just want to say Thanks for this great program. I've been experiencing problems for some time and thought it was a hardware failure because my system was 24+ hours Prime stable.
But Linpack failed just after the first loop. So I corrected the voltages and now everything works great. :up:
Q9550 @ 4250mhz
MemTest86+ = OK!
IBT v2 Standard = Success!
IBT v2 Maximum = freeze :(
Updated IntelBurnTest to v2.1!
- Optimized error-checking code, reduces CPU overhead
- Now displays more accurate ending time
- Simplified graphical user interface further
- Improved color of captions in the GUI
- Added ability to override the number of threads for Linpack to use (as requested by users)
- Added display of progress in the title bar
- Now outputs current time for each iteration to log file (hh:mm:ss)
- Now auto-scrolls output list to the bottom as results are outputted
- Fixed a GUI quirk
Core i7 users, please confirm if the new thread-forcing feature is working :D
Thanks, and enjoy!
Seems to not be working? unless I did something wrong?
http://i18.photobucket.com/albums/b140/hoss3430/ibt.jpg
Same results?
http://i18.photobucket.com/albums/b1...s3430/ibt2.jpg
Hmm... try changing it to Auto/16.
Lol, I thought the same thing, tried both, still same. :(
Ok, I've updated the release version, please re-download via main link. Should work fine with threading now.
Also guys, if you could, please test using 64-bit OS and forced 8-threads on a quad-core (like a Q9xxx/Q8xxx/Q6xxx) and Custom memory size of 64 MB, with 10 runs. (Just a tiny experiment, I'd like to see)
Because according to my little experiment, my Q6600 isn't stable. Failed within 1.76 seconds of doing this test, but passed the Standard test fine @ Auto threading (4 threads).
See screenshot:
http://i43.tinypic.com/11lpf05.jpg
Agent, is it possible if you could add a module to support temperature display as well? I know we can run other temp monitoring program to do that, but it would really be nice if you could add a module so the correct temp of each core are displaydd somewhere also. Thanks...
i get that same error on my q9550 when i run 8 threads, 64mb test.
I just downloaded the newest version, awesome work, great program :up:
AG
I have an issue
I set to 8 threads, 64mb in 64bit mode and it errors everytime
if a run linx with whatever settings including those it goes right thru?
LMK if you need more infos
try 16 threads. :D worked on mine
http://i18.photobucket.com/albums/b1...s3430/ibt3.jpg
I will not pretend to know why :p but 16 threads fixed it, it would error out right out the door with 8 threads. all 8 threads maxed right now and no issue in 64bit mode.
was passing linx without issues in 64 bit with any problem size in linx so I was starting to wonder....
3.8ghz 19x200 HT ON 1.2vcore bios set (1.158 cpuz) 1.2qpi and 1.1 IOH 1.64 vdimm 7-7-7-24 1T is now IBT stable as linx stable
THANKS !
My post, with pic was at stock, I tried a quad 9450 overclocked and at stock, 64mb mem 16 threads, all worked great.
8 threads with beta failed 64mb, and maximum mem with both stock CPU or overclocked CPU. edit sorry 8 threads on standard does work. 128mb seems to be working fine on 8 threads also.
you need me to test i7 920 stock clocks with 8 threads 64bit mode 64mb? turbo off or on?
i7 stock 64mb ram 8 threads = fail
i7 stock standard ram 8 threads = pass
i7 stock maximum ram 8 threads = fail
i7 stock 128 mb ram 8 threads = pass
i7 stock tried all above 16 threads = pass
i7 OC same results as above
9450 quad stock same as above except failed on standard also
9450 OC same as above
tested 2 other i7's same results as first i7
Hope this makes it clearer.
Yes it does, thank you again for your contribution :)
Looks like the new Linpack binary (created by Intel in April 2009) doesn't like certain test size combinations with 8 threads.
So the conclusion is: 16 threads for single Core i7 w/ HT, 32 threads for dual Nehalem servers, and "Auto" for everything else, yields the most accurate results :)
Ah ok,
Because i get failure with 8 threads every time, at more than standard.
Before i was Testing with 2x 2.0 to get 100% load and every test was fine.
Why i can't burning with a shortcut of v2.1?
With v1.9 i got no problems with burning from a shortcut.
Thanks.
BTW, the software work great for me with E8400 CO at overclock.
Great software, i love here.. you only with her to burn my cpu :P
Great program for testing!
http://i21.photobucket.com/albums/b2...20tests1-2.jpg
Damn, first time I see my CPU temps reach 75C!!! LOL! I only reach around 65-68C on Prime95.
AgentGOD:up::up:
I downloaded the new version and I'm getting this error message:
http://img34.imageshack.us/img34/3583/23018935.th.jpg
Any help? I'm running Windows XP x64 here and I have a Q9650.
^^ Tried that, still didn't work. WTF? Any more suggestions?
guys, I'm a little confused here, how many threads and what settings do I test an i7 setup? Should I disable ht before testing? Things were so easy with my e8400 = )
Hello.
I got a weird problem with the program.The program version i'm currently using is 2.1. So, here's the situation: no matter what options i choose for number of test, threads or memory amount, as soon as i click the 'start' button i get the message that i have passed the test successfully. For example, i choose 100 tests, 2 threads and 2000mb or ram, the moment i hit 'start' i get 'Success! Your system has been able to maintain stability...'. And that's everytime, no matter the chosen settings. I'm using vista ultimate 64bit.
using v2.1, standard stress level with threads set to auto it will successfully complete any amount of runs I set it too. If I switch stress level to maximum and leave threads to auto (Haven't tried any other) I hit start and then immediately after it says it completed successfully, no matter how many runs I have it set to.
i've got the same problem with my e8400, i cant run the 2.1v but the other versions works well.
http://img29.imageshack.us/img29/7236/70514354.th.jpg
How many threads shall i set it too on my QX9650? 4?
Also how much does this test the stability of my RAM?? Is it still best to memtest or prime to check the memory?
Thanks great tool :)
Thanks agentGOD
I have installed Vista 64 on another partition so am using IBT with that rather than XP
By using Vista..would that highlight any RAM issues better than XP. Would you recommend still using memtest with RAM stability even while testing with vista or if I can get 10 clear passes in Vista I should be ok??
Does Linpack work properly on Win7?
can you update the original post about the i7 needing 16 threads to give a true result please :)
not that i didnt enjoy reading the thread, but it will help the average jo bloggs.
great little package you put together :D
Hmmm my pc shut down 15 seconds in. Temps were only around 65c.
Updated IntelBurnTest to v2.2!
- Now outputs result of test (success, failure, etc) in log
- Now outputs stress level tested in log
- Added "high", and "very high" stress levels
- Fixed a small bug that could've broken logging
- Added debug logging for troubleshooting (right-click logging checkbox)
As said in the readme, if you have any issues, please help me find the cause by enabling the debug logger (added in v2.2), and then sending me the log file, as well as a screenshot of the issue.
Thanks to all for your support!
Nice AgentGOD! keep up the good work!:up:
I can't run the program, it gives me this msg:
http://img268.imageshack.us/img268/3566/ibt.jpg
the fire thingy that animates ( freeze test) sometimes stops moving if you open a window and drag it on top so it covers the flame. To get it moving again I have to drag the IBT window "off screen" and then back on. Seems to just be a visual bug, the benching never gets affected, but since it looks like a freeze, not sure if this can be fixed. I have an nvidia 9600gt video card, latest drivers as of 6-20-09 and running vista x64 sp2 not sure if any of that matters.......