New Linpack Stress Test Released

Regeneration · 5 Sep 2018 at 10:07

Linpack Xtreme is a console front-end with the latest build of Linpack (Intel Math Kernel Library Benchmarks 2018.3.011) developed and maintained by ngohq.com. Linpack is a benchmark and the most aggressive stress testing software available today. Best used to test stability of overclocked PCs. Linpack tends to crash unstable PCs in a shorter period of time compared to other stress testing applications.

Linpack solves a dense (real*8) system of linear equations (Ax=b), measures the amount of time it takes to factor and solve the system, converts that time into a performance rate, and tests the results for accuracy. The generalization is in the number of equations (N) it can solve, which is not limited to 1000. Linpack uses partial pivoting to assure the accuracy of the results.

Linpack Xtreme was created because Prime95 is no longer effective like it used to be. LinX, IntelBurnTest, OCCT use outdated Linpack binaries from 2012. Modern hardware requires modern stress testing methodology with support for the latest instructions sets.

Linpack Xtreme is available for Windows, Linux, and as a bootable media. The bootable version is considered to be the most superior as the Linux SMP kernel is a lot more sensitive to hardware instabilities than Microsoft Windows. Watch this video for a short comparison of Prime95 vs. Linpack Xtreme.

Make sure to keep an eye on the temperatures as Linpack generates excessive amount of stress like never seen before.

Changes (v1.1.4):
* Fixed a crash on AMD Ryzen processors.
* Updated CPUID HWMonitor to version 1.43.

Downloads:
Linpack Xtreme for Windows | Mirror #1 | Mirror #2
Linpack Xtreme for Linux | Mirror #1 | Mirror #2
Linpack Xtreme Bootable Media

FlyingScotsman · 7 Sep 2018 at 10:40

Interesting, is the time it runs for user definable?
How long would you run it to determine stability.
I still prefer running realbench and playing games etc to test stability.

Jamin280672 · 7 Sep 2018 at 13:35

FlyingScotsman said:
Interesting, is the time it runs for user definable?
How long would you run it to determine stability.
I still prefer running realbench and playing games etc to test stability.

Took a change to download it, you can define how many times it runs, or unlimited and also define amount of ram to use upto 9.5gb, id like to be able to use more though, like upto 32gb maybe.

Very buggy, doesn't log how many times it run, ie does the normal warm up, cool down thing, but whereas IBT logs each pass, this doesn't, then halfway through the 6th test, just came up with a load of data across the window.

FlyingScotsman · 8 Sep 2018 at 14:16

Jamin280672 said:
Took a change to download it, you can define how many times it runs, or unlimited and also define amount of ram to use upto 9.5gb, id like to be able to use more though, like upto 32gb maybe.

Very buggy, doesn't log how many times it run, ie does the normal warm up, cool down thing, but whereas IBT logs each pass, this doesn't, then halfway through the 6th test, just came up with a load of data across the window.

Yeah think I will stick with my usual stress testing methods, cheers for trying it though Jamin.

Regeneration · 10 Sep 2018 at 11:53

FlyingScotsman said:
Interesting, is the time it runs for user definable?
How long would you run it to determine stability.
I still prefer running realbench and playing games etc to test stability.

Personally, I stick with long runs (12 hours), I like my systems 100% stable.

Jamin280672 said:
Took a change to download it, you can define how many times it runs, or unlimited and also define amount of ram to use upto 9.5gb, id like to be able to use more though, like upto 32gb maybe.

Very buggy, doesn't log how many times it run, ie does the normal warm up, cool down thing, but whereas IBT logs each pass, this doesn't, then halfway through the 6th test, just came up with a load of data across the window.

This bug was solved in the last version (v0.8).

The new Linpack is more demanding than IBT's 2012 Linpack. I had to add 20mV to my vcore for this.

TaKeN · 18 Sep 2018 at 21:01

This program is crazy!

I'm hitting 90 degrees with a Cryorig R1 Ultimate flat out and had to stop the stress, OCCT used to make it hit around 70, and video encoding is around 65..

Regeneration · 19 Sep 2018 at 18:19

TaKeN said:
This program is crazy!

I'm hitting 90 degrees with a Cryorig R1 Ultimate flat out and had to stop the stress, OCCT used to make it hit around 70, and video encoding is around 65..

ROFL

TaKeN · 19 Sep 2018 at 22:03

Regeneration said:
ROFL

Is this normal?

Regeneration · 20 Sep 2018 at 05:45

20c above OCCT is a little too much.

Are you using LLC or automatic voltage? make sure the motherboard is not feeding too much vcore. If it does, try to adjust it to a level where both idle and load voltages are close as possible.

TaKeN · 21 Sep 2018 at 14:18

I'm using LLC Level 2 and the Vcore was definitely not excessive for this cooler.. 2500k is the chip btw.

Regeneration · 21 Sep 2018 at 20:38

TaKeN said:
I'm using LLC Level 2 and the Vcore was definitely not excessive for this cooler.. 2500k is the chip btw.

Linpack Xtreme is a lot more stressful (AVX). Video encoding isn't a 'big deal' anymore and OCCT uses outdated binary (and sometimes crash without hardware instability).

With version 0.9.3, AMD Ryzen users can now bake their CPUs too.

Jamin280672 · 23 Sep 2018 at 12:48

0.9.3 is much better than the original one I tried, guys hitting extreme temps, you need a new cooler lol, Prime95 got my CPU and MOS hotter, but I did only just run a quick 5 passes of 0.9.3 to give it a try out on a Ryzen 2700X @ 4.2ghz 1.36v.

Regeneration · 28 Oct 2018 at 00:23

The video above demonstrates the superiority of Linpack Xtreme versus the latest version of Prime95. The overclocked PC passed nearly 2 hours of Prime95's small FFTs torture test and yet completely crashed within less than a minute and a half (77 seconds to be precise) with the bootable version of Linpack Xtreme integrated on Porteus Linux.

Computer specifications: Intel Xeon W3680 @ 4.17 GHz (144x29), ASUS P6T (vanilla), 3x 4GB G.Skill Ares F3 @ 2304 MHz (10-12-12-31 CR1), and Nvidia GeForce GTX 970.

PiKe · 28 Oct 2018 at 00:34

Think I will add the Portius Linux edition to my PXE server, will come in handy.

Parsley · 28 Oct 2018 at 00:40

But.... why?

PiKe · 28 Oct 2018 at 00:43

Parsley said:
But.... why?

Testing for less time before proving a system isn't stable is preferable over running prime95 for hours and then getting random crashes when gaming.

Hotwired · 28 Oct 2018 at 01:53

TaKeN said:
This program is crazy!

I'm hitting 90 degrees with a Cryorig R1 Ultimate flat out and had to stop the stress, OCCT used to make it hit around 70, and video encoding is around 65..

It wouldn't be the first time for a stress test to be described as a power virus.

I believe Furmark causing AMD and Nvidia cards to destructively pull more power than they were designed to handle is why all modern cards protect themselves with throttling.

Oh well, it needs someone to be a smart *** and create a program that can destroy hardware so the protection can be designed against it.

Regeneration · 28 Oct 2018 at 01:33

Hotwired said:
It wouldn't be the first time for a stress test to be described as a power virus.

I believe Furmark causing AMD and Nvidia cards to destructively pull more power than they were designed to handle is why all modern cards protect themselves with throttling.

Oh well, it needs someone to be a smart *** and create a program that can destroy hardware so the protection can be designed against it.

Furmark is a bad example for stress test. It won't load new textures, won't stress the VRAM or the core.

Linpack and Prime95 just do what computers originally designed to do... to solve equations.

Parsley · 28 Oct 2018 at 12:13

PiKe said:
Testing for less time before proving a system isn't stable is preferable over running prime95 for hours and then getting random crashes when gaming.

But testing for 5 minutes at extreme loads like this isn't anything like day to day stability even for something like rendering and such. It's like taking a middle of the road sports car and stress testing it at F1 rev ranges. Why? Just dial in an overclock and make it stable for your day to day, correct it as and when it needs it.

I'm entirely with hotwired on this. People said the same thing about Furmark back when it started killing cards. AVX stress testing was already OTT as it was

. People reliant on avx applications for productivity aren't likely to be overclocking them anyway.

Regeneration · 28 Oct 2018 at 18:52

If the overclock isn't 100% stable, you'll run into trouble over time. App and game crashes, data corruption, and more.

It's better to spend a day or two to be certain the system is stable and to forget about it.

I learned it in the hard way. My 1st i7-965 build kept corrupting the OS files every few months.