Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Find other programs to test with besides Prime95 #4

Open
sp00n opened this issue Mar 8, 2021 · 11 comments
Open

Find other programs to test with besides Prime95 #4

sp00n opened this issue Mar 8, 2021 · 11 comments
Labels
enhancement New feature or request help wanted Extra attention is needed under investigation Checking out the issue

Comments

@sp00n
Copy link
Owner

sp00n commented Mar 8, 2021

Other programs might be even better suited for stability testing on edge cases, however they need to be able to be called from the command line and they need to stop or write an error log when an error happens, similar to what Prime95 does.

Aida64 Engineering Portable Trial might be worth a shot.

@sp00n sp00n added enhancement New feature or request help wanted Extra attention is needed under investigation Checking out the issue labels Mar 8, 2021
@sp00n
Copy link
Owner Author

sp00n commented Mar 9, 2021

SuperPi?

@AtlAntA118
Copy link

y-cruncher can be called from command line. It does give stop errors. It has an option to write an output file. Not sure if errors are written to that file as well..

@sp00n
Copy link
Owner Author

sp00n commented Mar 17, 2021

What is the option for the output file?
The v8 branch has support for y-cruncher, but I couldn't find a way to generate a log file for it.

@AtlAntA118
Copy link

You are right. I must have misread some documentation but as far as i can see there is no option for an output file. It does give outputs in the console window. So you could try and write the console window stuff to a txt file with some commands?
https://helpdeskgeek.com/how-to/redirect-output-from-command-line-to-text-file/

@sp00n
Copy link
Owner Author

sp00n commented Mar 19, 2021

Unfortunately I can't. I tried to use > logfile.txt and > logfile.txt 2>&1, but the only thing that appears in there is the "^C" when I abort the now blank screen with CTRL+C.
So apparently it does capture the output, but it doesn't write it to a file. Maybe you could give it a try?

Also, there is actually an -o option for writing an output file, but it doesn't do anything at least for stress testing (and in fact it even gives me an "Invalid Parameter: -o" message when I try to use it).

@AtlAntA118
Copy link

Sorry for the late response. I see you got it working anyway. Good job!

@ChiefZigZag
Copy link

ChiefZigZag commented Jun 21, 2021

I used a program recently which was able to find instability way quicker than anything else in regards to boost clock instability.. check this out.. it doesn't give errors the machine literally reboots if it hits a unstable core. I would recommend running this and output it into a file Boosttester.exe > Boost.log before beginning a much lengthier and deeper test core cycler offers. https://github.com/jedi95/BoostTester

@sp00n
Copy link
Owner Author

sp00n commented Jun 21, 2021

I used a program recently which was able to find instability way quicker than anything else in regards to boost clock instability.. check this out.. it doesn't give errors the machine literally reboots if it hits a unstable core. https://github.com/jedi95/BoostTester

I've used BoostTester before as well, but only to get a glimpse at the maximum possible boost clocks.
The program has zero error checking algorithms (you can look at the source code), so it's only useful if your cores are still in a highly unstable state.
I could basically run it for hours with -30 on all cores without a problem.

@ChiefZigZag
Copy link

ChiefZigZag commented Jun 21, 2021

I used a program recently which was able to find instability way quicker than anything else in regards to boost clock instability.. check this out.. it doesn't give errors the machine literally reboots if it hits a unstable core. https://github.com/jedi95/BoostTester

I've used BoostTester before as well, but only to get a glimpse at the maximum possible boost clocks.
The program has zero error checking algorithms (you can look at the source code), so it's only useful if your cores are still in a highly unstable state.
I could basically run it for hours with -30 on all cores without a problem.

I think it depends on the Chip/Board.. I have a 5900x on a Crosshair Impact which passed 48 hours "All" Prime95 Corecycler but it would randomly reboot in certain games .. Running boosttester had it reboot on the first round on the weaker cores, there was more than one! Couldn't get instability on any other tests including AIDA64.. perhaps nothing pushes the boost clocks the same way? I have no idea but it was my experience. My other boards and CPUs didn't experience this at all and Corecycler alone did the trick.

@LucidLuxxx
Copy link

Linpack Extreme uses CLI to run benchmarks and stress tests and does print a results.txt file.

@sp00n
Copy link
Owner Author

sp00n commented Aug 20, 2024

Linpack is now part of the possible stress test programs. It's not necessarily Linpack Xtreme, but one of the binaries you can select (2018) matches the binary that is also used in Linpack Xtreme.

Linpack Xtreme itself is basically just a batch file wrapped in an .exe file and calling the binary with the settings you provide.

BoostTester is included in the /tools/ directory, I've actually created my own slight modification of it now, which includes the changes from Mann1x for slightly higher boost clocks and my own changes to support the new Intel architecture with P- and E-Cores.

Oh, and y-cruncher and Aida64 have been added a long time ago as well. 😅

And still on the lookout for other possible candidates.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request help wanted Extra attention is needed under investigation Checking out the issue
Projects
None yet
Development

No branches or pull requests

4 participants