LinX - a nicer friendier Linpack

[nyker96]

Just ran across a new GUI that's sprout up recently that gives the IntelBurnTest aka Linpack a nicer friendlier GUI interface. It even tells ya which run it's currently on etc and stops as soon as there's a discrepancy. Also allows you to choose the problem size. Overall has a nice progress bar/info panel that makes you feel more 'connected' to the testing process.

In short, it does everything except making coffee.

Link: http://www.xtremesystems.org/f...howthread.php?t=201670

 

[VirtualLarry]

Does it fix the problem with IntelBurnTest's "maximum stress" setting giving errors due to out of memory?

 

[nyker96]

yes, it has a safe limit which it won't allow you to exceed. That's that max 32bit limit, I think it's under 2000MB.

 

[nerp]

Thanks for the heads up, nyker. I'll be grabbing this for sure.

 

[Cheex]

I'll definitely check it out. Thanks!!

 

[aigomorla]

Originally posted by: VirtualLarry
Does it fix the problem with IntelBurnTest's "maximum stress" setting giving errors due to out of memory?

why would you want to test max memory?

the whole point in linpack was to load up your cpu and have it do cpu intensive tasks, not both.

The good thing about this program is its fast, i shifted all my stress program to this.

I set it on custom settings, use 512megs of ram, and let it run 10 passes.

If it passes that your very good to go, i have yet for my system to fail crunching on WCG for 2 weeks now without a reboot.

Oh just a note, this program WILL PISS YOU OFF. It WONT pass where OCCT and Prime would pass 24/7. And YOU will need to compromise.

:X

Gillbot is scared silly to use it, and it PISSES me off from time to time because i'll get 1/10 thats not correct and need to retune.

Lastly, expect a 10-15C higher temp mark then OCCT and Prime using this program. So make sure you got your CPU temps in check or you'll throttle VERY fast.

 

[Rubycon]

Originally posted by: aigomorla

Oh just a note, this program WILL PISS YOU OFF. It WONT pass where OCCT and Prime would pass 24/7. And YOU will need to compromise.

And a system that passes this test for a week can produce artifacts in output of encoded video!

System testing is getting political in that there are apparently different levels of "claimed stability". Rock solid? Well there's a Mohs scale of hardness that goes from 1 to 10. Most are probably calcite stable. :laugh:

 

[supastar1568]

I will now use LinPack to test my future systems stability.

From what I had passed in Prime, I failed in Linpack. I was finally able to figure out that I needed +.05V on the FSB, and +.025V on the northbridge, the cpu voltage remained the same and I got 100 passes completed on Linpack. Obviously, I will never stress my system equal to 100 passes of Linpack.

Also, I have learned to turn error checking OFF. If it is on, the program will run the full amount of tests and give you a "stable" or "not stable". Turning it off, shows the results dynamically, and you are able to quit out of the program when you see an error, no point in having to wait till the end if you have even one error.

 

[LOUISSSSS]

Originally posted by: Rubycon

Originally posted by: aigomorla

Oh just a note, this program WILL PISS YOU OFF. It WONT pass where OCCT and Prime would pass 24/7. And YOU will need to compromise.

And a system that passes this test for a week can produce artifacts in output of encoded video!

System testing is getting political in that there are apparently different levels of "claimed stability". Rock solid? Well there's a Mohs scale of hardness that goes from 1 to 10. Most are probably calcite stable. :laugh:

QFT, stability is passed around left and right with people knowing that nothing is 100% stable in the CPU world. there are just too many factors. even if you prove your cpu to be 99.99999999999999% stable, the programs you are running might only be 99.9% stable and you'd be limited to that.

i trust OCCT thus far. passing that means my OS has been stable for everything i do. and nothing i do reaches higher than those temps. theres no point in further heating up the CPU.

 

[nyker96]

Just a little bit of my personal experience concerning the fact that this program will FAIL even know other stress testers will pass your CPU.

I recently bought one of those $86 a pop E7200. I tested the thing 3.6@1.3525v with 4 hours or ortho + 4hrs OCCT passed. Even did a 4hr x264 job, absolutely no problem.

Then tried this baby on, failed within an hour. lowered the chip to 3.52Ghz. Passed about 5 hrs of it. temperature side: at full load OCCT/ortho it's about 55-60C. At full load Linpack, it's about 65C. So conclusion? This thing will stress your CPU more than OCCT/ortho, it just means it require a whole new level of stability to satisfy it.

I don't think anything I normally run on a daily basis (Photoshop CS3//x264//occasional compiler) will come near this level of stress but it's up to you to decide what you want out of your system not Linpack. I think you can just say passing 2hrs of this thing is good enough and tune your OC accordingly. As for me I just left it at 3.52Ghz since I do some work stuff on this machine and an extra degree of safety is appreciated.

 

[Avalon]

I gave it a run of 10 passes on my E8400 @ 4Ghz, and passed. Going to try 100 overnight tonight. This definitely heats the bejeesus out of your CPU. I was reaching 73C in coretemp and 53C in speedfan with my Ultra-120 on it.

 

[aigomorla]

Originally posted by: Rubycon

Originally posted by: aigomorla

Oh just a note, this program WILL PISS YOU OFF. It WONT pass where OCCT and Prime would pass 24/7. And YOU will need to compromise.

And a system that passes this test for a week can produce artifacts in output of encoded video!

System testing is getting political in that there are apparently different levels of "claimed stability". Rock solid? Well there's a Mohs scale of hardness that goes from 1 to 10. Most are probably calcite stable. :laugh:

yeah, but my systems are stable enough even for you to hack away at it.

and you know it too.

 

[supastar1568]

Yes, I have had no failires after 50 runs, but have got errors going to 100. I finally passed 100 tests in Linpack.

Then I failed Prime95 after 6 hours. My suggestion to all would be to get it linpack stable, then finish it off with Prime testing. 5 years later, you will have a conclusion whether you are really stable or not.

 

[aigomorla]

Originally posted by: supastar1568
Yes, I have had no failires after 50 runs, but have got errors going to 100. I finally passed 100 tests in Linpack.

Then I failed Prime95 after 6 hours. My suggestion to all would be to get it linpack stable, then finish it off with Prime testing. 5 years later, you will have a conclusion whether you are really stable or not.

Excuse me?

But your saying the oposite of what everyone is saying.

And i know for a fact, that linX is more intensive then prime95.

So can you explain what you mean?

 

[bryanW1995]

it's possible that he actually damaged the cpu and/or system by stressing it with linpack that much. that's the scary part of this. I agree that aiming for 100% stability is critical, but at some point we have to realize that it's entirely possible to damage your system if you stress-test it too much.

 

[VirtualLarry]

Originally posted by: bryanW1995
it's possible that he actually damaged the cpu and/or system by stressing it with linpack that much. that's the scary part of this. I agree that aiming for 100% stability is critical, but at some point we have to realize that it's entirely possible to damage your system if you stress-test it too much.

I disagree. Unless you are pushing too much vcore, or too high temps, then stress-testing is all good, not bad.

Edit: I don't know why everyone says that Linpack increases temps over prime95 or other programs. I just don't see it, not on any of my machines. Then again, I run SeventeenorBust in the background on my machines, 100% of CPU time is used, so temps are normally fairly high. Linpack is certainly no worse on temps.

Edit: Oh, btw, my friend's E5200 @ 3.75Ghz is also LinX-stable. I qualified it using Prime95 originally.

 

[bludragon]

Originally posted by: VirtualLarry
Edit: I don't know why everyone says that Linpack increases temps over prime95 or other programs. I just don't see it, not on any of my machines. Then again, I run SeventeenorBust in the background on my machines, 100% of CPU time is used, so temps are normally fairly high. Linpack is certainly no worse on temps.

Well there may be a difference with your cpu, but I get similar results to most others of temps being 8-10 degC higher with linx than orthos. That is dependent on nothing else running though, the temps can drop very fast if something else grabs the cpu. Actually, I was surprised at how fast the temps could change - makes me wonder if linpack stresses the part of the core that is right next to the temp diode.

 

[Dadofamunky]

Originally posted by: Rubycon

Originally posted by: aigomorla

Oh just a note, this program WILL PISS YOU OFF. It WONT pass where OCCT and Prime would pass 24/7. And YOU will need to compromise.

And a system that passes this test for a week can produce artifacts in output of encoded video!

System testing is getting political in that there are apparently different levels of "claimed stability". Rock solid? Well there's a Mohs scale of hardness that goes from 1 to 10. Most are probably calcite stable. :laugh:

That's exactly what's happening to me. PISSES me off.

Originally posted by: nyker96
Just a little bit of my personal experience concerning the fact that this program will FAIL even know other stress testers will pass your CPU.

I recently bought one of those $86 a pop E7200. I tested the thing 3.6@1.3525v with 4 hours or ortho + 4hrs OCCT passed. Even did a 4hr x264 job, absolutely no problem.

Then tried this baby on, failed within an hour. lowered the chip to 3.52Ghz. Passed about 5 hrs of it. temperature side: at full load OCCT/ortho it's about 55-60C. At full load Linpack, it's about 65C. So conclusion? This thing will stress your CPU more than OCCT/ortho, it just means it require a whole new level of stability to satisfy it.

I don't think anything I normally run on a daily basis (Photoshop CS3//x264//occasional compiler) will come near this level of stress but it's up to you to decide what you want out of your system not Linpack. I think you can just say passing 2hrs of this thing is good enough and tune your OC accordingly. As for me I just left it at 3.52Ghz since I do some work stuff on this machine and an extra degree of safety is appreciated.

Heck, looks like Linpack won't accept anything beyond stock!

 

[Gillbot]

Originally posted by: aigomorla
Gillbot is scared silly to use it, and it PISSES me off from time to time because i'll get 1/10 thats not correct and need to retune.

I'm not scared silly, I just KNOW it'll fail. There's a difference. I prefer to remain in the "dark".

 

[aigomorla]

Originally posted by: Gillbot

Originally posted by: aigomorla
Gillbot is scared silly to use it, and it PISSES me off from time to time because i'll get 1/10 thats not correct and need to retune.

I'm not scared silly, I just KNOW it'll fail. There's a difference. I prefer to remain in the "dark".

its warn in the light tho.

 

[Rubycon]

Originally posted by: aigomorla


yeah, but my systems are stable enough even for you to hack away at it.

and you know it too.

Ruby is 9 on the scale.

Originally posted by: Gillbot

Originally posted by: aigomorla
Gillbot is scared silly to use it, and it PISSES me off from time to time because i'll get 1/10 thats not correct and need to retune.

I'm not scared silly, I just KNOW it'll fail. There's a difference. I prefer to remain in the "dark".

Why?

I'd rather run STOCK than o/c not knowing if the system is not producing faithful results.

 

[Gillbot]

Originally posted by: Rubycon

Originally posted by: aigomorla


yeah, but my systems are stable enough even for you to hack away at it.

and you know it too.

Ruby is 9 on the scale.

Originally posted by: Gillbot

Originally posted by: aigomorla
Gillbot is scared silly to use it, and it PISSES me off from time to time because i'll get 1/10 thats not correct and need to retune.

I'm not scared silly, I just KNOW it'll fail. There's a difference. I prefer to remain in the "dark".

Why?

I'd rather run STOCK than o/c not knowing if the system is not producing faithful results.

I don't need it to be "linpack" stable when it runs all my apps and plays all my games 100% fine without a single crash. It passes OCCT fine and is stable enough for me to do everything I need it to do.

 

[VirtualLarry]

Does the temp increase in an inverse relationship to the size of memory tested? I set it to test max mem, which on 32-bit windows is something under 2048MB. If I selected to only test, say, 512MB or 256MB, would my temps run higher? Which is preferable during testing? More mem or less mem?


I only ask this because on my rigs, Prime95 "blend" gives me a lower temp than "small FFT".

 

[supastar1568]

Originally posted by: aigomorla

Originally posted by: supastar1568
Yes, I have had no failires after 50 runs, but have got errors going to 100. I finally passed 100 tests in Linpack.

Then I failed Prime95 after 6 hours. My suggestion to all would be to get it linpack stable, then finish it off with Prime testing. 5 years later, you will have a conclusion whether you are really stable or not.

Excuse me?

But your saying the oposite of what everyone is saying.

And i know for a fact, that linX is more intensive then prime95.

So can you explain what you mean?

Oh, dont't get me wrong, I like linpack much more than Prime95. But since I was trying to zone in my voltage to the minimal amount needed, I think I was in a grey area. I passed Linkpack with 100 runs, then failed Prime95 after 6 hours. This same test, with the same settings I passed in Prime95 for 24 hours a month or so ago. Therefore, I came to the conclusion that I was on the fringe of stability/instability. I bumped my voltage up, about 60mV and passed prime for 24 hours.

Regardless, I get failures much quicker in Linkpack than I do in Prime. Passed 100 passes Linpack, which takes about 5 hours. Failed Prime95 after 6 and a half, not exactley apples to apples.

 

[supastar1568]

My recommendation to all.

Do NOT turn on error checking. When it askes for yes or no, say no, which is "n".

This is because, if it is on, you do not get your results until the test is passed. Which is fine if you are going to bed and won't look at your computer till the morning.

Leaving it off gives you dyamic results, so if you have an inconsistant number, you can end the test right there, since no point in going further with the test. With it on, you have to wait to see if you failed or not.

Example:

With error checking on, you could do 50 passes. You can fail your second calculation but the test would finish the remaining 48 and tell you after.

With it off, you could see that your second calcuation failed and end the test right there.