LinX - A simple Linpack interface

What is this difference between LinX and IntelBurnTest?

Which is better program?

Pretty much same stresswise, different user interface ;)

Guys, have a quick question. Want to move my testing from prime to LinX as its most likely a bit quicker at it, however having a few probs. When I select 'max memory' it either stops instantly, with a 'out of memory' error, or before I even start it tells me I dont have enough memory:

http://img149.imageshack.us/img149/1058/linx.jpg

The most I can run it at is about 3.9gb of data (I have 6gb as im on i7) but not sure if that gives me a high enough problem size to be 100% sure of stability.

In addition to this, the program doesnt seem to max my cores to 100% when Im running it. I do have HT on with my i7 - not sure if this is the problem, although prime doesnt have an issue with it?

Quote:

---

Originally Posted by Xerran

What is this difference between LinX and IntelBurnTest?

Which is better program?

---

LinX can load all 8 threads on the i7, IBT can't.

I also have 6GB but I run only 3072 problem size and 40 loops. Then windows memtest for full 6GB. That's it :)

MikeMK,
I'm not sure if I can do anything more to that «not enough memory» problem. LinX uses an approximated formula to calculate how much memory Linpack is going to consume at the given Problem Size. If the calculated amount of memory is bigger than what OS reports to be free memory then it shows that warning message saying that the values were corrected. When the «all mem» button is pressed LinX chooses the Problem Size value for Linpack to consume almost all free memory. But either that formula isn't correct enough for bigger numbers or Linpack doesn't like big numbers too. (Hopefully 2nd is the case as a 600 MB error doesn't make me a good mathematician and a programmer either)

I would be grateful if you could post or send me via PM a screenshot with LinX running at those 3.9 GB and a Task Manager window with a linpack_xeon64.exe process and its memory.

Quote:

---

Originally Posted by MikeMK

the program doesnt seem to max my cores to 100% when Im running it.

---

Linpack doesn't keep the cores always at 100%, its load is less consistent than that of Prime, especially at the beginning/end of each test. If that is what you experience then it's ok.

And yes, 3.9 GB is still quite enough for Linpack to check stability. :)

where did all of the previous posts go?

Wasn't me. :) That's probably due to the continuing XS server manipulations.

Quote:

---

Originally Posted by Dua|ist

Wasn't me. :) That's probably due to the continuing XS server manipulations.

---

can you give me some instructions on how to use your program. I unzipped the file with 7z and installed it on my C drive and now everytime I get the error that was initially reported in the "lost posts" write error or something.

http://img401.imageshack.us/img401/6452/lintest2.th.jpg

Like the interface, nice and simple.

This may be a dumb question but how do I know if the oc is unstable while using LinX? Does the testing stop if a error is detected, or does it just keep running until it finishes the amount set to loop for then show if there's any error?

Quote:

---

Originally Posted by koei

This may be a dumb question but how do I know if the oc is unstable while using LinX? Does the testing stop if a error is detected, or does it just keep running until it finishes the amount set to loop for then show if there's any error?

---

If there is any variation in the residual(norm) value for each pass that is completed, the OC is unstable.

Quote:

---

Originally Posted by rge

Dualist, Linx 0.5.5.1 worked perfect..that was a fast fix:up:

Here is pic with max memory 1748 on XP 32 bit, loading 99.9% avg all cores.

I also tried it at 2 notches of vcore below stable at 4.2Ghz, where prime reboots around the 2 to 3 hour mark, linx rebooted at 4th run on 1748 mem, then 3rd run on second try, so linx is back in business with 1 instance:up:.

Unfortunately on core i7's, I cant take screen pics of prime errors or linx errors as almost all OC failures on my core i7 are reboots or crashes.

---

Dear rge,
Where can I get the i7 load tester?

Is this both a stress test and a benchmark?

What is a good GFlops number?

http://chuckbam.com/i7_P6T-D/LinX0.5.8a.PNG

http://chuckbam.com/i7_P6T-D/LinX0.5.8b.PNG

chuckbam - GFlops depends on CPU.

Quote:

---

Originally Posted by kuebk

chuckbam - GFlops depends on CPU.

---

I know, but, how are my numbers for 3.6 GHz?

Quote:

---

Originally Posted by chuckbam

I know, but, how are my numbers for 3.6 GHz?

---

I'm surprised that you have such low GFlops at 3.6GHz.

With my Q9450 at 3.6GHz and DDR 1199MHz I've got higher GFlops as you can see in the screen shot in my sig.

Quote:

---

Originally Posted by A-Grey

I'm surprised that you have such low GFlops at 3.6GHz.

With my Q9450 at 3.6GHz and DDR 1199MHz I've got higher GFlops as you can see in the screen shot in my sig.

---

I will run it again with Power Management off

Quote:

---

Originally Posted by chuckbam

I will run it again with Power Management off

---

Yes disable all these things in the BIOS and use 3450Mb Memory that's what I used, than you will see the actual difference between your i7 and my C2Q.

Quote:

---

Originally Posted by A-Grey

I'm surprised that you have such low GFlops at 3.6GHz.

With my Q9450 at 3.6GHz and DDR 1199MHz I've got higher GFlops as you can see in the screen shot in my sig.

---

I7 with HT on and 12Gb of ram, if he uses only 4Gb of ram he will surpass ya for sure

Quote:

---

Originally Posted by Leeghoofd

I7 with HT on and 12Gb of ram, if he uses only 4Gb of ram he will surpass ya for sure

---

I don't doubt that. If he gets lower GFlops outcomes than I have with my Q9450 than there is something seriously wrong.;)

I get around 49.5 with all the power management turned off and some TSRs disabled. Where the i7 excels is with multi thread processing. In some reviews, they have some benchmarks where the Yorkfield does beet the i7.

But, trust me, I have a very capable computer. How do you do with CINEBENCH R10?
Or, this one? http://chuckbam.com/i7_P6T-D/Bios2/SAN-ProcArith.PNG


Tom's Hardware
Compare Core i7 I7-920 Quad Core Processor, Core 2 Quad Q9450 Processor

http://www.tomshardware.com/charts/d...d%5B2182%5D=on

Quote:

---

Originally Posted by chuckbam

I get around 49.5 with all the power management turned off and some TSRs disabled. Where the i7 excels is with multi thread processing. In some reviews, they have some benchmarks where the Yorkfield does beet the i7.

But, trust me, I have a very capable computer. How do you do with CINEBENCH R10?
Or, this one? http://chuckbam.com/i7_P6T-D/Bios2/SAN-ProcArith.PNG


Tom's Hardware
Compare Core i7 I7-920 Quad Core Processor, Core 2 Quad Q9450 Processor

http://www.tomshardware.com/charts/d...d%5B2182%5D=on

---

If I see all these scores and the difference between them I can only say that the Core i7 only benefits from the higher bandwidth with the triple channel memory.

For me it isn't worth the upgrade from my Core2Quad to a Core i7. That's what I think of it but everyone has to decide this for his own.

I don't doubt that you have a very capable computer.;)

Shall we make a chart of top Linpack scores here at XS then? ;)

I just realized that Linpack could be quite a nice benchmark. Not only it shows CPU's performance but also guarantees that the result achieved is almost 99% stable. But the competition is too tough I guess. :)

FWIW, I get 53.6 GFlops with my setup with HT disabled.