MFLOPS rating

ôÅÓÔÉÒÏ×ÁÎÉÑ ÐÏ ÔÅÓÔÁÍ MFLOPS ÓeÒ×ÅÒÁ RM600 E-60 É ðë

PEAK MFLOPS performance for the FADD, FSUB, FMUL, and FDIV
operations based on specific 'instruction mixes'. The execution
modules (1-8) include various percent weightings of FDIV's (from 0% to
25% FDIV's). Look hire for details.
š
RM600E60, Reliant Unix 5.44, R10000 200MHz
ëÜÛ 4 íb
cc: SNI: CDS++ V1.0C3200
module\cc directives - -O -F X4 -F X4 -K lp64
1, MFLOPS 35.8543 147.0661 147.1264 147.1264
2, MFLOPS 43.6647 72.1998 70.137 70.6346
3, MFLOPS 54.7559 271.6604 260.2871 260.2871
4, MFLOPS 56.5371 286.5672 99.7921 215.2466
5, MFLOPS 50.3664 60.3038 107.2213 131.6079
6, MFLOPS 57.3193 75.2865 160.7274 211.2692
7, MFLOPS 30.2006 39.2839 36.3292 38.5107
8, MFLOPS 46.8064 122.9981 166.3418 199.688
MFLOPS(1) 46.7613 95.0043 92.1437 92.7213
MFLOPS(2) 41.7887 66.142 72.9741 82.8708
MFLOPS(3) 47.3066 89.8138 114.7489 139.0994
MFLOPS(4) 52.8254 122.308 157.8748 215.3846
š š š š š
FreeBSD 2.2.6-RELEASE ,
PentiumPro 200 MHz
ëÜÛ 512 Kb
gcc v. 2.7.2.1
module\cc directives - -O6 š š
1, MFLOPS 13.3983 35.2846 š š
2, MFLOPS 8.7535 98.3396 š š
3, MFLOPS 15.6315 40.7971 š š
4, MFLOPS 15.7222 48.7249 š š
5, MFLOPS 15.9989 27.1170 š š
6, MFLOPS 17.2808 30.8390 š š
7, MFLOPS 10.2866 25.6528 š š
8, MFLOPS 17.1564 30.9867 š š
MFLOPS(1) 10.2243 67.3047 š š
MFLOPS(2) 13.5189 30.3862 š š
MFLOPS(3) 15.3806 31.9680 š š
MFLOPS(4) 16.6411 34.5604 š š

24.02.99
Vejnovets Igor

š ôÅÓÔÉÒÏ×ÁÎÉÑ ÐÏ MFLOPS ÒÁÚÌÉÞÎÙÈ ðë:
š

Free BSD 2.2 6-REALASE,
PentiumPro200 MHz;
ëÜÛ 512 Kb;
(gcc v. 2.7 2.1)
Free BSD 2.2 6-REALASE,
PentiumPro200 MHz;
ëÜÛ 512 Kb;
(gcc v. 2.7 2.1)
Windows 2000 Server,
K7-750T,ËÜÛ 256 Kb;
(Borland Ctt Builder5)
Pentium III/500;
ËÜÛ 512 Kb;
PentiumIV 2,4Ggz, 533Mgz,
ËÜÛ 512Kb;
Module\cc directives
-
-0.6
-
-
-
1, MFLOPS 13.3983 35.2846 š81.7481 179.7392 š580.6399
2, MFLOPS š8.7535 98.3396 109.9117 109.2683 1056.9154
3, MFLOPS 15.6315 40.7971 š97.7011 199.5415 š394.5066
4, MFLOPS 15.7222 48.7249 109.2523 191.1399 š377.2102
5, MFLOPS 15.9982 27.1170 148.2665 146.1705 š516.4522
6, MFLOPS 17.2808 30.8390 165.4779 214.6285 š543.0076
7, MFLOPS 10.2866 25.6528 š84.3215 š53.9705 š217.9651
8, MFLOPS 17.1564 30.9867 155.0575 203.2284 š517.3808
MFLOPS (1) 10.2243 67.3047 105.5972 128.2343 š682.3517
MFLOPS (2) 13.5189 30.3862 106.8430 103.7930 š348.0283
MFLOPS (3) 15.3806 31.9680 122.7729 154.4910 š443.0063
MFLOPS (4) 16.6411 34.5604 133.8189 203.8502 š468.4732

ššš 21.01.2001. - 10.2002.
šš Vejnovets Igorš



FLOPS C Program (Double Precision), V2.0 18š (Itanium, GCC)
Moduleš
Errorš
RunTime (usec)
MFLOPS
1
4.0146e-13
0.0528
265.1746
2
-1.4166e-13
0.0451
155.1673
3
4.7184e-14
0.0358
475.4812
4
-1.2546e-13
0.0458
327.6800
5
-1.3800e-13
0.0869
333.8021
6
3.2374e-13
0.0498
582.4753
7
-8.4583e-11
0.1350
88.8925
8
3.4855e-13
0.0538
557.6345

šš Iterationsššššš =š 512000000
šš NullTime (usec) =šššš 0.0013
šš MFLOPS(1)šššššš =šš 198.9926
šš MFLOPS(2)šššššš =šš 187.4302
šš MFLOPS(3)šššššš =šš 317.5407
šš MFLOPS(4)šššššš =šš 491.5841


FLOPS C Program (Double Precision), V2.0 18 (Itanium, Intel C++)
Module
Error
RunTime (usec)
MFLOPS
1
4.0146e-13
0.0070
1992.4084
2
-1.4166e-13
0.0050
1395.9746
3
4.7184e-14
0.0010
16880.4848
4
-1.2546e-13
0.0030
4980.5706
5
-1.3800e-13
0.0130
2222.8583
6
3.2374e-13
0.0060
4814.5516
7
-6.5654e-11
0.0221
543.5852
8
3.4855e-13
0.0070
4271.7655

šš Iterationsššššš =š 512000000
šš NullTime (usec) =šššš 0.0040
šš MFLOPS(1)šššššš =š 1993.9285
šš MFLOPS(2)šššššš =š 1283.5914
šš MFLOPS(3)šššššš =š 2465.6482
šš MFLOPS(4)šššššš =š 5332.5374

šš š01.2003.
šš Vejnovets Igorš


NEWS OVERVIEW Getting Startedš Software Benchmarks Personal