»RSS Benchmark sajt

      Prikazani rezultati od 1 do 2 od ukupno 2

      Tema: Linpack Stress Test by Regeneration

      1. #1
        Member Avatar korisnika sam633
        Na forumu od
        Nov 2011
        Poruka
        4.703

        Linpack Stress Test by Regeneration



        Linpack Stress Test is a small console front-end with the latest build of Linpack (Intel Math Kernel Library Benchmarks 2018.3.011). Linpack is the ultimate CPU stress testing software available today - best used to test stability of overclocked CPUs. Unstable systems tend to crash in just a matter of minutes by running Linpack.

        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.

        I created this project because LinX, IntelBurnTest, and OCCT use outdated Linpack binaries. Make sure to keep an eye on the temperatures as Linpack generates excessive amount of stress.
        Linpack doesn't work well with problem sizes above 35000 (9500MB), and yet it is not documented in LinX or IBT.
        v0.3 - Added support for AMD CPUs.
        Poslednje verzije će biti ovde: (mada ima i na drugim forumima, trenutna verzija je 0.6)

        https://www.techpowerup.com/forums/t...leased.247335/
        Igoritza:"Necu se ja vala ovde vise nikad registrovati."

      2. #2
        Member Avatar korisnika danilo989
        Na forumu od
        Oct 2006
        Poruka
        1.261
      3. Moja mašina
        • CPU: Intel Core i7-6700k @4500MHz 1.36v; Noctua NH-D15
        • MoBo: ASUS Maximus VIII Hero
        • RAM: 2x8GB Corsair Vengeance LPX CMK16GX4M2B3200C16R DDR4 @3200MHz 1.35v
        • VGA: Gigabyte GeForce GTX 980 GV-N980G1 Gaming-4GD
        • MON: Philips 226V4LSB
        • HDD: Kingston HyperX Savage 240GB; WD Red WD20EFRX-68EUZN0; WD Red WD40EFRX-68N32N0 @LC-Power LC-35U3
        • Case: Fractal Design Define R5 Window
        • SND: Realtek ALC1150; Altec Lansing 220; HyperX Cloud KHX-H3CL/WR
        • OS: Windows 10 Education 1903 64 bit
        • LAP: Lenovo Ideapad 330S-14IKB 81F4014LYA; Rapoo MT750
        • INET: DSL
        • Ostalo: CoolerMaster CM Storm QuickFire TK Brown; Logitech G203 Prodigy; SanDisk Cruzer Extreme 64GB
      4. Odgonetnuo sam zackoljicu koja me kopkala mesecima... Sa novijim verzijama LinX-a (0.7+) bi mi se desavalo da residual vrednosti budu razlicite a test uspesno prolazio uprkos tome cak i na stock frekvencijama https://forum.benchmark.rs/showthrea...=1#post5025977
        Vidim da se i sa Linpack Xtreme 0.9.3 isto desava drugim ljudima https://www.techpowerup.com/forums/t...5/post-3899168 uz odgovor da je to normalno na novijim Intel CPUovima https://www.techpowerup.com/forums/t...5/post-3899247
        https://www.techpowerup.com/forums/t...5/post-3902308
        Spoiler:
        Linpack Xtreme v0.9.3 by Regeneration (64-bit)

        Current date/time: Mon Oct 15 11:40:04 2018

        CPU frequency: 4.498 GHz
        Number of CPUs: 1
        Number of cores: 4
        Number of threads: 8

        Parameters are set to:

        Number of tests: 1

        Number of equations to solve (problem size) : 35000
        Leading dimension of array : 35000
        Number of trials to run : 999999999
        Data alignment value (in Kbytes) : 4
        Maximum memory requested that can be used=9800704096, at the size=35000

        =================== Timing linear equation system solver ===================

        Size LDA Align. Time(s) GFlops Residual Residual(norm) Check
        35000 35000 4 124.094 230.3568 1.134067e-09 3.292025e-02 pass
        35000 35000 4 125.924 227.0085 1.134067e-09 3.292025e-02 pass
        35000 35000 4 126.182 226.5446 1.134067e-09 3.292025e-02 pass
        35000 35000 4 123.786 230.9284 1.134067e-09 3.292025e-02 pass
        35000 35000 4 126.791 225.4551 1.134067e-09 3.292025e-02 pass
        35000 35000 4 126.620 225.7608 1.134067e-09 3.292025e-02 pass
        35000 35000 4 128.556 222.3603 1.134067e-09 3.292025e-02 pass
        35000 35000 4 125.318 228.1066 1.134067e-09 3.292025e-02 pass
        35000 35000 4 123.279 231.8795 1.134067e-09 3.292025e-02 pass
        35000 35000 4 122.835 232.7169 1.134067e-09 3.292025e-02 pass
        35000 35000 4 122.904 232.5869 1.134067e-09 3.292025e-02 pass
        35000 35000 4 123.319 231.8032 1.134067e-09 3.292025e-02 pass
        35000 35000 4 125.617 227.5631 1.134067e-09 3.292025e-02 pass
        35000 35000 4 125.561 227.6639 1.134067e-09 3.292025e-02 pass
        35000 35000 4 129.370 220.9606 1.070718e-09 3.108133e-02 pass
        35000 35000 4 125.033 228.6262 1.134067e-09 3.292025e-02 pass
        35000 35000 4 132.295 216.0760 1.134067e-09 3.292025e-02 pass
        35000 35000 4 128.865 221.8269 1.134067e-09 3.292025e-02 pass
        35000 35000 4 126.597 225.8010 1.134067e-09 3.292025e-02 pass
        35000 35000 4 126.189 226.5321 1.134067e-09 3.292025e-02 pass
        35000 35000 4 124.685 229.2641 1.134067e-09 3.292025e-02 pass
        35000 35000 4 123.910 230.6981 1.134067e-09 3.292025e-02 pass
        35000 35000 4 123.665 231.1544 1.134067e-09 3.292025e-02 pass

        Konfa iz potpisa. GFlopsi variraju jer koristim komp tokom testa
        ( ゚Д゚)

      Tagovi ove teme

      Bookmarks

      Dozvoljeni tagovi

      • Ne možete pokrenuti nove teme
      • Ne možete odgovoriti u temi
      • Ne možete okačiti atačmente
      • Ne možete izmeniti svoje poruke
      •