How do you disable AVX in Prime95

[xilman]

Originally Posted by henryzz

Yes but NFS post-processing needs rock stable memory.

Indeed. What I like and what I get isn't always the same.

[wingman99]

Thanks for all the replies. I was also wondering what the round off checking does? I have more errors when it is enabled.

[VBCurtis]

Originally Posted by wingman99

Thanks for all the replies. I was also wondering what the round off checking does? I have more errors when it is enabled.

Without putting too fine a point on it, it checks for roundoff errors- if you don't check for them, you might miss some miscalculations (fewer error messages).

The FFT is a bit of an approximation- imagine if multiplying two integers produced a decimal output, such as 20.31. Now, since the inputs were integers, you know the output should be an integer, so you round off the decimals and say the output is 20. But what if the output is 20.49? Is the error 0.49 and the correct answer 20, or is the error 0.51 and the correct answer is 21?

Prime95 generally chooses an FFT such that errors stay under 0.40, to 'eliminate' this sort of uncertainty. If a roundoff error is over 0.40, either the program should use a slower but more accurate calculation (a larger FFT), or the computer made a calculation error. Prime95 checks this by repeating the step that produced the roundoff error- if repeatable, use larger FFT. If not repeatable, hardware error!

[wingman99]

Thanks for the information.

[LaurV]

To add to it, P95 checks anyhow for rounding errors, every xxx (1000? 10000?) iterations (quantity is not important, but the idea), to keep the calculus "on track", no matter if you check that box or not. Checking that box will make it look for rounding errors more often (every 100? every 10?) iterations. As this is a timely process, it is not recommended to do it at every iteration, and your test will run slower (like 5%, 10%? depending on the system/CPU/memory) when you check the box, but it will be safer. On a system which you know it is good (i.e. produces successful DC residues every now and then) you better don't check the box. On a "new" (unproved) system, check the box. For safety, many of us here always run with the box checked. Shit happens even with the best computers...

For a more mathematical explanation, see the chapter on the Gimps' math page.

[wingman99]

Thanks I started a new post because I did not ask if the round off check enabled works in the torture test. I'm working with overclocking PCs.