Can you guys run a benchmark test for me?

[matt271]

Hi guys

Can you guys run a benchmark for me? This is for a paper; similar to one I had asked for a year ago; but this one is optimized differently.

c:

#include <stdio.h> #include <stdlib.h> #include <math.h> #include <time.h> #define TESTS 10 float ofk_accumulator = 0.0; float InvSqrt(float x) {    float xhalf = 0.5f * x;    int i = *(int*) &x;    i = 0x5f3759df - (i >> 1);    x = *(float*) &i;    x = x * (1.5f - xhalf * x * x);    return x; } double testInvSqrt() {    time_t start = clock();    int i;    float f, g;    for (i = 0x00800000; i < 0x7f800000; i++) {        f = *(float *) &i;        g = InvSqrt(f);        ofk_accumulator += g;    }    return (double) (clock() - start) / CLOCKS_PER_SEC; } double testSqrtf() {    time_t start = clock();    int i;    float f, g;    for (i = 0x00800000; i < 0x7f800000; i++) {        f = *(float *) &i;        g = 1.0f / sqrtf(f);        ofk_accumulator += g;    }    return (double) (clock() - start) / CLOCKS_PER_SEC; } int main(int argc, char** argv) {    int i;    float a, b, d, sum = 0.0F;        FILE *f = fopen("benchmark.txt", "w");    for (i = 0; i < TESTS; i++) {        a = testInvSqrt();        printf("%f", a);        b = testSqrtf();        printf("\t%f", b);        d = b / a;        printf("\t%f\n", d);        sum += d;        fprintf(f, "%f\t%f\t%f\n", a, b, d);    }    printf("%f\n", sum / TESTS);    fprintf(f, "%f\n", sum / TESTS);    fclose(f);        // thanks kaser...    if (ofk_accumulator > 3.0)        printf(" ");    else        printf("   ");    return (EXIT_SUCCESS); }

Please compile with gcc -O3 -lm test.c or equivalent and post cpu details cat /proc/cpuinfo.

Thank you very much!

Last time, some people found it ran too long. If that is the case, just cancel it and post what it completed. Test only takes about 3 minutes on my Macbook Pro.

If you want to leave any name or other details, I can note you as a personal communication in my references.

[Sur_real]

disclaimer: I ran this test on windows. My linux machine is a VM so I dunno if that would have been the best idea

Benchmark.txt

code:

4.805000        37.835999       7.874298 4.735000        37.823002       7.987962 4.740000        37.825001       7.979959 4.713000        37.792000       8.018672 4.693000        37.918999       8.079906 4.761000        37.932999       7.967443 4.715000        37.912998       8.040933 4.725000        37.876999       8.016296 4.725000        37.937000       8.028995 4.706000        37.869999       8.047174 8.004164

CPU:

code:

CPU Identification utility v2.09                 (c) 1997-2012 Jan Steunebrink  ==============================================================================  CPU Vendor and Model: Intel Core i7 Quad M i7-2600QM/2700QM series D2-step  Internal CPU speed  : 1995.4 MHz  System CPU count    : 1 Physical CPU(s), 4 Core(s) per CPU, 8 Thread(s)  CPU-ID Vendor string: GenuineIntel  CPU-ID Name string  : Intel(R) Core(TM) i7-2630QM CPU @ 2.00GHz  CPU-ID Signature    : 0206A7  CPU Features        : Floating-Point Unit on chip  : Yes                        Time Stamp Counter           : Yes                        Enhanced SpeedStep Technology: Yes                        Hyper-Threading Technology   : Yes                        Execute Disable protection   : Yes                        64-bit support               : Yes                        Virtualization Technology    : Yes  Instr set extensions: MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2  Size of L1 cache    : 4 x 64 KB  Integrated L2 cache : 4 x 256 KB  Integrated L3 cache : 6144 KB

[matt271]

Thanks!

[crossley7]

CPU Info:

code:

Intel(R) Processor Identification Utility Version: 4.41.20111201 Time Stamp: 2012/03/26 01:57:33 Operating System: 6.1-7601-Service Pack 1 Number of processors in system: 1 Current processor: #1 Active cores per processor: 2 Disabled cores per processor: 0 Processor Name: Intel(R) Core(TM) i5-2410M CPU @ 2.30GHz Type: 0 Family: 6 Model: 2A Stepping: 7 Revision: 1A Maximum CPUID Level: D L1 Instruction Cache: 2 x 32 KB L1 Data Cache: 2 x 32 KB L2 Cache: 2 x 256 KB L3 Cache: 3 MB Packaging: LGA1155 Enhanced Intel SpeedStep(R) Technology: Yes MMX(TM): Yes Intel(R) SSE: Yes Intel(R) SSE2: Yes Intel(R) SSE3: Yes Intel(R) SSE4: Yes Intel(R) AES-NI: Yes Intel(R) AVX: Yes Enhanced Halt State: Yes Execute Disable Bit: Yes Intel(R) Hyper-Threading Technology: Yes Intel(R) 64 Architecture: Yes Intel(R) Virtualization Technology: Yes System Graphics: Intel(R) HD Graphics 3000 Expected Processor Frequency: 2.30 GHz Reported Processor Frequency: 2.69 GHz Expected System Bus Frequency: 100 MHz Reported System Bus Frequency: 100 MHz *************************************************************

benchmark.txt:

code:

24.809999       36.719002      1.480008 26.749001       36.891998      1.379192 24.524000       36.707001      1.496779 25.052000       37.408001      1.493214 25.207001       36.953999      1.466021 24.780001       37.258999      1.503592 25.486000       37.292000      1.463235 25.969000       37.479000      1.443221 26.422001       37.837002      1.432026 25.191999       37.539001      1.490116 1.464740

[Ultrahex]

proc info:

code:

processor       : 0 vendor_id       : GenuineIntel cpu family      : 6 model      : 42 model name      : Intel(R) Core(TM) i7-2640M CPU @ 2.80GHz stepping        : 7 microcode       : 0x25 cpu MHz  : 2790.629 cache size      : 4096 KB physical id     : 0 siblings        : 4 core id  : 0 cpu cores       : 2 apicid    : 0 initial apicid  : 0 fpu          : yes fpu_exception   : yes cpuid level     : 13 wp            : yes flags      : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts nopl xtopology nonstop_tsc aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes xsave avx lahf_lm ida arat epb xsaveopt pln pts dts tpr_shadow vnmi flexpriority ept vpid bogomips        : 5583.81 clflush size    : 64 cache_alignment : 64 address sizes   : 36 bits physical, 48 bits virtual power management: (there is 4 of these I don't care to repeat)

benchmark.txt

code:

5.860000        23.920000       4.081911 5.820000        23.900000       4.106529 5.810000        23.930000       4.118761 5.850000        23.889999       4.083761 5.850000        23.889999       4.083761 5.810000        23.889999       4.111876 5.810000        23.879999       4.110155 5.840000        23.900000       4.092465 5.810000        23.889999       4.111876 5.830000        23.889999       4.097770 4.099887

[matt271]

thank you guys!