• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

A new CPU Benchmark: Multi Threaded PI @ LINPACK

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Pi benchmark - 8350 @ 4.7ghz.

Xr2NOuz.png
 
What relevance to the real LINPACK benchmark does this LINPACK have? Its it simply a name copy or is there something else? Its obviously not as heavy on the CPU as the real one.
 
OP already got all your bitcoins!

But seriously, it would have been better if OP had shown the source and given instructions on how it was compiled too.
 
Sable said:
one assumes that's why the thread was closed.

one also assumes that the reason the thread has been reopened is because the OP has presented his credentials and shown the mods that there is nothing nefarious about the software.

If this is this is the case it would have been nice of the mods to let us know. 😉
Click to expand...
That wasn't the case. The discussion was solely over a concern of a newly registered member's (OP's) first post being self promotional.

We're not really a software vetting service. In a case like this we like to assume you guys are smart enough to do that on your own.
 
no worries, I assumed incorrectly.

(although I think you also assume incorrectly with your last point 😉 )
 
ViRGE said:
That wasn't the case. The discussion was solely over a concern of a newly registered member's (OP's) first post being self promotional.

We're not really a software vetting service. In a case like this we like to assume you guys are smart enough to do that on your own.
Click to expand...

Sire, YES and NO. This software is not commercial. Is free software. And the idea was to make this benchmark popular among overclockers. The software is virus free, Trojans free, keylogers free.

The development begun one year ago and the birth of the benchmark was first pictured by LAB501.ro [forums].

Here:
http://forum.lab501.ro/showthread.php?t=9210

The benchmark has gone gold couple of monts later [the core algorithm that reside in two DLL's written in Visual C++]. Since then, only de user interface has changed several times. From that point of view, all the submissions to HWBOT remains valid.

We have almost 1000 submission for PI and 600 for LINPACK on HWBOT. I want to make this benchmark popular and relevant for overclockers. Is not a stability test or a heat test, is a benchmark that is relevant for multi core systems and can differentiate systems by result. The lower, the better. Also, the benchmark features perfect parallelization among any number of CPU's or cores.

LINPACK make use of BLAS to perform basic linear algebra operations with vectors and matrix. PI use BBP formula to compute up to 80000 decimals. All that, making user of every core or CPU available.
 
DEVELOPING MULTI THREADED PI @ LINPACK

The development of Multi Threaded PI @ LINPACK core algorithm:

1410768_269929783158368_1866372942_o.png


The development of Multi Threaded PI @ LINPACK front end:

966102_269930906491589_64650200_o.png
 
Last edited:
tertiaryphase said:
DEVELOPING MULTI THREADED PI @ LINPACK

The development of Multi Threaded PI @ LINPACK core algorithm:

1410768_269929783158368_1866372942_o.png
Click to expand...

I can see one memory leak in there already just by glancing at it. And that's just a screenshot converter, not a LinPACK algorithm!

EDIT: And mismatched memory allocators/deleters. If you allocate using malloc, then you free the memory using free! Not delete[]!
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top