|
DescriptionRandMem benchmark carries out eight tests at increasing data sizes to produce data transfer rates in MBytes Per Second from caches and memory. Serial and random address selections are employed, using the same program structure, with read and read/write tests for 32 bit integers and 64 bit floating point numbers. The C/C++ program structure is as follows with array xi indexing via sequential or random numbers stored in the array. Read - toti = toti & xi[xi[i+0]] | xi[xi[i+2] & xi[xi[i+4]] |& to i+30 Read/write - xi[xi[i+2]] = xi[xi[i+0]]; repeated to i+30 and i+28 The main purpose is to demonstrate performance differences between sequential and random access when using the same CPU instructions, particularly the impact of burst reading (and writing) over a bus. In this case, with random access, 32 bytes or more will be read when only four are requested. (see also BusSpd2K - link below) Random speeds are also affected by lower level cache sizes. A pre-compiled version of the benchmark can be found in RandMem.zip which also contains the source code, providing further explanatory comments. Information on maximum speeds when different processing is involved can be obtained from BusSpd2K results.htm and SSE3DNow results.htm . A version to test two CPUs is also available (See DualCore.htm, Win64.htm and Below). Latest version is for up to 8 cores or hyperthreads (see Quad Core 8 Thread.htm and later details. Then there is My Main Page for other PC benchmarks and results. Following is an example output for a 3.0 Ghz Pentium 4E. Variations in performance identify L1 and L2 cache sizes, but it should be noted that speed can be slower than normal when data size equals cache size. A second example probably represents the best performance that could be expected on a May 2014 desktop, assuming no overclocking. |
Integer....................... Double/Integer................
Serial........ Random........ Serial........ Random........
RAM Read Rd/Wrt Read Rd/Wrt Read Rd/Wrt Read Rd/Wrt
KB MB/Sec MB/Sec MB/Sec MB/Sec MB/Sec MB/Sec MB/Sec MB/Sec
L1 6 9620 5664 9712 3460 7991 7733 7992 4655
12 8592 5057 9840 3436 8015 7874 8894 4414
L2 24 6533 3017 5327 3043 7653 4443 7431 3941
48 6457 2986 3857 2707 7855 3969 6891 3505
96 6304 2944 3482 2541 7517 4821 6899 3250
192 6892 3073 3309 2626 7651 4131 6205 3354
384 6720 2925 2231 2156 7245 4114 5594 3296
768 5224 2631 1255 1314 7455 4196 2568 2262
RAM 1536 3517 1727 436 300 3454 2092 1394 1161
3072 3340 1681 235 171 3565 2090 440 312
6144 3517 1731 182 141 3533 2080 325 224
12288 3508 1688 169 131 3569 2000 284 207
24576 3523 1656 166 126 3523 2092 266 196
49152 3510 1683 162 124 3537 2035 258 188
98304 3442 1736 160 122 3557 2091 255 187
196608 3511 1731 155 118 3499 2083 249 185
393216 3304 1659 133 105 3554 2089 231 176
Core i7 4820K mainly ruuning at 3.9 GHz using Turbo Boost
32 GB 1600 MHz RAM over 4 channels, Windows 8.1
Integer....................... Double/Integer................
Serial........ Random........ Serial........ Random........
RAM Read Rd/Wrt Read Rd/Wrt Read Rd/Wrt Read Rd/Wrt
KB MB/Sec MB/Sec MB/Sec MB/Sec MB/Sec MB/Sec MB/Sec MB/Sec
6 24753 21240 24353 20950 27914 26690 27901 26866
12 24674 21377 24041 20986 28277 24369 28276 27232
24 24599 21373 24361 21586 28457 24246 28440 25932
48 22414 20560 18133 12948 28389 24984 28045 22632
96 22465 20538 13834 8952 28354 24827 22114 13686
192 22480 20579 11814 7779 28353 24880 18659 12085
384 21765 17461 7988 5917 26567 21036 14434 9949
768 21847 17211 6070 5018 26933 19937 10299 7930
1536 21853 17168 5439 4604 26452 20292 8886 7261
3072 21456 16651 3263 3165 26243 20120 8286 6868
6144 21383 16613 1607 1575 26209 20114 3338 3184
12288 13559 10997 1165 1137 18529 14306 2042 1965
24576 12429 10285 926 858 16547 12810 1575 1468
49152 12596 10358 758 702 16559 12756 1283 1192
98304 12572 10351 603 572 16509 12777 1059 1012
196608 12599 10363 510 492 16422 12752 834 818
393216 12573 10368 468 454 16403 12771 733 728
786432 12565 10383 442 429 16512 12775 687 685
Maximum speed 800 MHz x 2 DDR x 8 bus width x 4 channels = 51.2 GB/sec
Multiple cores need to be used for a higher throughput from RAM
|
|
The benchmark has also been ported to 32-Bit and 64-Bit Linux using the supplied GCC compiler (all free software) - see linux benchmarks.htm, and download benchmark execution files, source code, compile and run instructions in memory_benchmarks.tar.gz. Using Windows the file downloaded wrongly as .tar.tar but is fine when renamed .tar.gz. See Linux results below. There is now a version to run on Android based tablets and phones, using a subset of the standard program functions. See
Android Benchmarks.htm
and results below.
ResultsSeparate tables of speeds obtained via L1 cache, L2 cache and RAM are given below. Except when connected via the memory bus, performance via caches tends to be proportional to CPU MHz for a given type of processor. So, only a sample of results are provided. Details of cache sizes, speed and range of CPU MHz can be found in CPUSpeed.htm. L1 Cache Results - The results show that Pentium 4 CPUs can produce faster data transfer rates than AMD processors of the same MHz on serial and random integer reading tests, but they perform much worse with reading/writing. Core 2 Duo is the best overall. To clarify the results, the table includes instructions/byte ratios that can be used to convert MBytes/second to Millions of Instructions Per Second (MIPS). These have been used to produce the example %MIPS/MHz figures shown. Note that serial and random speeds are generally similar on L1 cache. L2 Cache Results - Results provided are fastest serial speeds and random speeds at 96 KB. The latter decreases with increased memory demands and depends on L1 cache size to some extent. Again example %MIPS/MHz calculations are given, showing that both Intel P4 and AMD CPUs can be faster on some of the tests but Core 2 Duo is better than both. RAM Results - This shows maximum serial speeds and random access performance at 6144 KB (6 MB). An indication of the type and speed of RAM is also shown. Best performance on sequential operation is shared between Pentium 4 and Pentium 4E, with Pentium M being best on random access. AMD CPUs show improvements over Integer Serial Read speeds on the other test but only the Athlon 64 provides reasonable comparative performance across the board. The latest results are for Core 2 Duo, showing some excellent and some poor speeds. The latter might be due to the nForce 570 chipset used. Later results via Intel 965 chipset confirmed this suspicion. RAM Random Access Results - As random access speeds tend to reduce with increased data size, results are given here for the first random reading test at 1.5 to 96 MB. L2 cache sizes are also included to show that larger caches generally produce better performance. The impact of reading a burst of data is reflected by the relative good performance of Pentium III CPUs. These have 32 byte bursts compared with 64 or more on the other CPUs. AMD Athlons have greater reductions in speed at the higher data sizes. Dual Core Results See below. Eight Thread Benchmarks See below. Linux Results See below. Android Results See below.
|
L1 Cache Results in MBytes/Second - sorted by CPU MHz
Integer Double/Integer
Serial Random Serial Random
Read Rd/Wrt Read Rd/Wrt Read Rd/Wrt Read Rd/Wrt
80486 DX2 66 63 80 69 87 47 65 51 80
Pentium 100 205 243 200 233 248 301 258 281
Pentium MMX 200 439 525 434 510 565 669 564 634
Pentium Pro 200 654 308 654 470 760 662 794 681
Pentium II 450 1471 1072 1508 1077 1745 1530 1801 1495
Celeron A 450 1496 1084 1511 1084 1757 1508 1761 1485
Pentium III 450 1500 1066 1482 1034 1702 1472 1719 1499
AMD K62 500 1114 1434 1131 1356 790 1575 841 1510
Celeron 2 566 1900 1375 1908 1357 2276 1928 2263 1882
Duron 700 1582 1730 1615 1727 2819 2320 2575 2253
Pentium IIIE 733 2460 1772 2462 1751 2909 2491 2928 2437
Athlon 800 1843 2025 1918 2017 2031 2401 1893 2444
Athlon Tbird 1000 2310 2514 2360 2471 4038 3310 3687 3256
Celeron M 1295 4620 3199 4511 3152 6404 4359 6666 4383
Atom M 1600 2639 3215 2722 3213 3398 3786 3437 3838
Pentium 4 1800 6361 3421 6559 2378 6139 5687 6138 3021
Ath4 Barton 1800 4068 4290 4077 4438 7377 5960 6654 5843
Core 2 Duo M 1830 4317 7669 6611 5123 8875 9348 9316 8444
Pentium M 1862 6586 4612 6701 4584 9793 6304 9771 6288
Pentium 4 1900 6553 3667 6788 2511 6361 6188 6443 3192
Turion 64 M 1900 4691 5222 4776 4965 7891 6653 7569 6660
Opteron 2000 4514 4909 4532 4922 8063 6609 7421 6464
Celeron C2 M 2000 6884 7227 7095 5034 10163 10333 6852 7987
Athlon XP 2080 4728 5215 4755 5158 8268 6830 7618 6800
Athlon 64 2210 5554 6072 5532 6129 9772 7799 9165 7724
Core i5 2467M @@@@ 7800 7822 8834 7978 10059 9427 10114 10698
Core 2 Duo 1 CP 2400 8821 9518 8806 7379 12415 12690 12405 12464
Pentium 4E HT 3000 9620 5664 9840 3460 8015 7874 8894 4655
Pentium 4 3000 10397 5781 10768 3830 10230 9448 10255 4938
Core i7 **** 10809 11713 10802 12145 14813 14343 14405 15544
Phenom II 3000 12252 8269 11570 8222 15567 10000 15514 10664
Core i7 #### 12122 7425 12505 6818 16279 9503 16598 10807
Pentium 4 3678 12630 7668 13268 4703 12561 11942 12478 6096
Core i7 4820K $$$$ 24674 21377 24041 20986 28277 24369 28276 27232
MIPS multiply by 0.55 0.37 0.55 0.37 0.28 0.31 0.28 0.31
$$$$ 3.7 GHz i7 4820K, running at up to 3.90 GHz using Turbo Boost
#### 2.8 GHz i7 860, running at up to 3.46 GHz using Turbo Boost
**** 2.8 GHz i7 930, running at up to 3.06 GHz using Turbo Boost
@@@@ 1.6 GHz i5 2467M, running at up to 2.30 GHz using Turbo Boost
%MIPS/MHz
Pentium II/III 186 89 186 88 112 106 112 104
Pentium M 196 91 200 90 148 106 148 106
Pentium 4 190 77 200 47 96 101 95 52
Pentium 4E HT 178 69 182 42 75 82 83 48
Atom M 91 74 94 74 59 73 60 74
Celeron C2 M 189 134 195 93 142 160 96 124
Core 2 Duo M 130 155 199 104 136 158 143 143
Core 2 Duo 1 CP 202 147 202 114 145 164 145 161
Core i5 @ assume 2.3 GHz 187 126 211 128 122 127 123 144
Core i7 * assume 3.06 GHz 196 140 195 145 136 146 132 158
Core i7 # assume 3.46 GHz 193 79 199 73 132 85 134 97
Core i7 $ assume 3.90 GHz 348 203 339 199 203 194 203 216
Athlon XP 126 92 127 91 112 103 103 102
Turion 64 136 102 138 97 116 109 112 109
Athlon 64 139 101 139 102 124 110 117 109
Phenom II 225 102 212 101 145 103 145 110
Core i5 @ assume 1.6 GHz 268 181 304 184 176 183 177 207
Core i7 * assume 2.80 GHz 214 154 214 159 149 160 145 173
Core i7 # assume 2.80 GHz 238 98 246 90 163 105 166 120
To Start To Index L2 Cache Results in MBytes/Second at 96 KB - same order as L1 speeds
Integer Double/Integer
Serial Random Serial Random
Read Rd/Wrt Read Rd/Wrt Read Rd/Wrt Read Rd/Wrt
80486 DX2 66 23 17 11 12 22 15 12 14
Pentium 100 96 73 32 40 89 69 35 46
Pentium MMX 200 195 135 86 93 183 132 94 110
Pentium Pro 200 487 269 208 132 613 310 357 207
Pentium II 450 700 325 313 136 559 398 323 177
Celeron A 450 994 769 287 233 912 813 319 309
Pentium III 450 801 335 303 141 794 406 526 230
AMD K62 500 400 182 72 55 425 221 111 77
Celeron 2 566 1505 1222 373 356 1186 1376 388 426
Duron 700 1143 1073 678 718 1210 1313 1203 1390
Pentium IIIE 733 2060 1674 1531 993 2593 1827 2275 1479
Athlon 800 840 576 610 320 1169 864 1193 1048
Athlon Tbird 1000 1636 1543 976 1028 2520 1906 2454 2066
Celeron M 1295 3386 2678 1930 1009 4447 3183 3115 1649
Atom M 1600 2160 2306 718 944 2775 2584 1208 1455
Pentium 4 1800 4143 2129 2621 1901 6541 5023 4903 2313
Ath4 Barton 1800 2968 2819 1571 1814 4525 3220 4378 3733
Core 2 Duo M 1830 5793 6735 3061 2717 7520 6418 5412 4198
Pentium M 1862 4833 4132 2807 1458 6733 4965 4393 2371
Pentium 4 1900 5115 2215 2745 1965 6786 3036 4713 2437
Turion 64 M 1900 2804 2671 2486 2393 4426 3994 4797 4140
Opteron 2000 3128 3198 2881 2731 5222 3671 5249 4402
Celeron C2 M 2000 6213 7155 3319 3006 8788 7702 6050 4428
Athlon XP 2080 3458 3311 2054 2112 5232 3931 5083 4419
Athlon 64 2210 4070 3734 3322 3257 6140 4420 6124 5218
Core i5 2467M @@@@ 8593 7538 5300 3390 11588 8536 7796 5175
Core 2 Duo 1 CP 2400 7752 8989 4112 3655 10739 9632 7335 5771
Pentium 4E HT 3000 6892 3073 3482 2541 7855 4821 6899 3250
Pentium 4 3000 8104 3238 4291 3117 9936 6036 8324 3856
Core i7 **** 10156 10801 5895 5623 13359 12881 9894 9110
Phenom II 3000 10549 7860 6381 5215 15308 9662 14830 9879
Core i7 #### 11111 6666 5911 5429 13574 8977 10187 8073
Pentium 4 3678 9894 4533 5166 3785 12423 9155 9174 4396
Core i7 4820K $$$$ 22465 20538 13834 8952 28354 24827 22114 13686
$$$$ 3.7 GHz i7 4820K, running at up to 3.90 GHz using Turbo Boost
#### 2.8 GHz i7 860, running at up to 3.46 GHz using Turbo Boost
**** 2.8 GHz i7 930, running at up to 3.06 GHz using Turbo Boost
@@@@ 1.6 GHz i5 2467M, running at up to 2.30 GHz using Turbo Boost
L3 Cache at 3072 KB i5 1536 KB
Phenom II 3000 7874 6680 1077 1017 9428 8358 2048 2045
Core i5 2467M @@@@ 7064 5632 2243 1904 10357 7834 3927 2977
Core i7 **** 9718 9846 2364 2312 12661 11345 5207 4408
Core i7 #### 9762 6331 2378 2620 14411 9396 5608 4601
Core i7 4820K $$$$ 21853 17168 5439 4604 26452 20292 8886 7261
MIPS multiply by 0.55 0.37 0.55 0.37 0.28 0.31 0.28 0.31
%MIPS/MHz
Pentium IIIE 156 84 116 50 99 78 87 63
Pentium M 144 81 84 29 102 83 66 40
Pentium 4 149 45 78 38 95 78 70 37
Pentium 4E HT 127 38 64 31 74 50 65 34
Atom M 74 53 25 22 49 50 21 28
Celeron C2 M 171 132 91 56 123 119 85 69
Core 2 Duo M 174 136 92 55 115 109 83 71
Core 2 Duo 1 CP 178 139 94 56 125 124 86 75
Core i5 @ assume 2.3 GHz 205 121 127 55 141 115 95 70
Core i7 * assume 3.06 GHz 184 129 107 67 123 131 91 93
Core i7 # assume 3.46 GHz 177 71 94 58 110 80 82 72
Core i7 $ assume 3.90 GHz 317 195 195 85 204 197 159 109
Athlon XP 92 58 55 37 71 59 69 66
Turion 64 M 81 52 72 47 65 65 71 68
Athlon 64 102 62 83 54 78 63 78 74
Phenom II 193 97 117 64 143 100 138 102
Core i5 @ assume 1.6 GHz 295 174 182 78 203 165 136 100
Core i7 * assume 2.80 GHz 201 142 117 74 134 144 99 102
Core i7 # assume 2.80 GHz 218 88 116 72 136 99 102 89
To Start To Index RAM Results in MBytes/Second at 6144 KB - sorted by serial read speeds
Integer Double/Integer
Serial Random Serial Random
MHz Read Rd/Wrt Read Rd/Wrt Read Rd/Wrt Read Rd/Wrt
80486 DX2 66 21 10 6 7 17 10 8 9
Pentium 100 55 35 11 14 52 40 19 22
Pentium MMX P0 200 121 83 23 26 111 78 31 36
Pentium Pro P0 200 129 75 32 21 142 84 53 16
AMD K62 P1 500 132 87 18 15 130 94 26 20
Celeron A P0 300 232 115 62 41 138 151 77 57
Duron P2 700 247 193 36 30 393 343 49 46
Athlon Tbird P2 1000 249 207 38 33 488 358 62 53
Celeron 2 P0 566 276 167 80 53 253 192 93 70
Athlon P2 800 250 191 38 33 371 323 54 51
Pentium II P1 450 300 152 89 61 196 194 107 79
Pentium III P1 450 329 167 98 68 350 233 169 112
Ath4 Barton D1 1800 383 265 69 48 559 343 115 77
Athlon 4 D1 1667 453 426 129 93 699 573 222 149
Pentium IIIEB P2 1000 469 257 142 97 513 344 215 156
Pentium IIIEB P2 733 474 204 96 66 391 251 123 88
Athlon XP D2 2080 884 727 183 116 1224 880 311 187
Pentium 4 P2 1900 940 387 48 42 914 483 76 64
Celeron M 1295 1029 456 89 55 1467 632 144 93
Pentium 4 R1 1400 1324 689 107 84 1123 912 159 118
Pentium 4 D1 1800 1394 630 98 80 1658 803 168 123
Opteron D3 2000 1536 1377 121 111 2297 1822 235 217
Pentium 4 D1 2533 1561 599 75 58 1623 738 118 90
Pentium 4 D1 3066 1737 655 70 51 1718 785 125 81
Turion 64 M DC3 1900 1758 1392 247 191 2222 1704 430 304
Pentium 4 R2 2533 1968 1019 172 145 2919 1352 297 220
Atom M DD2 1600 2058 1072 52 81 2283 1392 84 127
Pentium M DC1 1862 2073 787 340 213 2442 1238 616 376
Athlon 64 D3 1995 2100 965 156 122 2520 1432 291 225
Athlon 64 DC2 2210 2145 1451 248 159 3008 1785 402 254
Pentium 4 DC1 2533 2335 847 98 72 2303 978 166 114
Celeron C2 M DC3m 2000 3000 1212 302 183 3027 1455 514 311
Pentium 4 DC2 3678 3150 1850 181 124 4115 2103 294 196
Core 2 Duo M DC3M 1830 3384 1524 459 296 3349 1864 849 534
Pentium 4E HT DC2 3000 3523 1736 182 141 3569 2092 325 224
Core 2 Duo 1CP DC3b 2400 4854 2605 789 597 5532 3799 1486 1309
Core 2 Duo 1CP DC3a 2400 4947 770 349 208 1685 1052 932 557
Core 2 Duo 1CP DC3c 2400 5136 2775 878 657 6086 4041 1637 1396
Phenom II $C DC33 3000 6120 6079 747 654 9065 7991 1395 1220
Core i5 2467M DC33 @@@@ 6127 5396 484 458 7722 6141 825 786
Core i7 $C DC32 **** 7261 5273 953 854 7008 5650 1665 1483
Core i7 $C DC33 #### 7811 5110 1071 870 8036 5998 1652 1742
Core i7 $C QC34 $$$$ 21383 16613 1607 1575 26209 20114 3338 3184
Core i7 12.3MB QC34 $$$$ 13559 10997 1165 1137 18529 14306 2042 1965
Maximum 13559 10997 1165 1137 18529 14306 2042 1965
$$$$ 3.7 GHz i7 4820K, running at up to 3.90 GHz using Turbo Boost
#### 2.8 GHz i7 860, running at up to 3.46 GHz using Turbo Boost
**** 2.8 GHz i7 930, running at up to 3.06 GHz using Turbo Boost
@@@@ 1.6 GHz i5 2467M, running at up to 2.30 GHz using Turbo Boost
Key P0 66 MHz P1 100 MHz
P2 133 MHz D1 DDR 133 MHz
D2 DDR 166 MHz D3 DDR 200 MHz
DC1 Dual Channel DDR 133 MHz DC2 Dual Channel DDR 200 MHz
DC3a DDR2 533 MHz nForce 570 chipset DC3b DDR2 533 MHz Intel 965 chipset
DC3c DDR2 800 MHz Intel 965 chipset DC3M DDR2 666 MHz Mobile CPU
DC3m DDR2 533 MHz Mobile CPU DC33 DDR3 1333 MHz
DC32 DDR3 1066 MHz QC34 DDR3 1600 MHz 4 Channels
R1/R2 RDRAM 400/533 MHz $C 12.3 MB Mainly or all L3 cache
To Start To Index RAM Random Reading MBytes/Second 1.5 MB to 96 MB - sorted by speed at 1.5 MB
---------------------- MBytes ------------------------
MHz Cache 1.5 3 6 12 24 48 96
Pentium 4 P2 1900 0.25 56 51 48 47 46 45 39
Pentium 4 P2 2400 0.25 69 55 50 48 47 47 43
Atom M DD2 1600 0.50 69 57 52 50 47 46 41
Pentium 4 D1 3066 0.50 70 70 70 73 72 70 69
Athlon Tbird P2 1000 0.25 75 47 38 35 32 30 26
Athlon Tbird P2 900 0.25 78 46 38 34 31 29 26
Athlon Tbird D1 1200 0.25 97 64 53 49 44 40 34
Pentium 4 D1 2533 0.50 102 80 75 70 70 69 67
Athlon 4 D1 1394 0.25 107 99 95 93 54 39 29
Pentium 4 D1 1800 0.25 115 104 98 99 98 93 75
Pentium 4 DC1 2533 0.50 140 110 98 98 96 95 91
Athlon 4 D1 1667 0.25 148 134 129 121 65 43 32
Pentium IIIEB P2 933 0.25 160 136 136 140 139 119 91
Pentium IIIEB P2 1000 0.25 160 147 142 141 139 130 99
Athlon 4 D2 1768 0.25 164 152 143 139 92 64 49
Pentium 4 R1 2400 0.25 170 142 131 123 122 121 116
Celeron M 1295 1.00 199 107 89 80 77 73 65
Athlon XP D1 1733 0.25 200 177 169 163 100 66 50
Athlon XP D2 2080 0.25 213 191 183 174 101 70 53
Pentium 4 R2 2533 0.50 220 182 172 160 156 154 152
Athlon XP D2 2167 0.25 229 206 196 187 116 81 60
Pentium 4 DC2 3678 0.50 240 197 181 171 168 167 158
Turion 64 M DC3 1900 0.50 344 281 247 242 219 168 116
Athlon 64 DC2 2210 0.50 362 272 248 238 230 218 164
Pentium 4E DC2 3000 1.00 436 235 182 169 166 162 160
Celeron C2 M DC3m 2000 1.00 607 260 302 251 262 247 191
Athlon 64 D3 2000 1.00 614 274 156 124 111 105 102
Opteron D3 2000 1.00 636 255 121 91 80 75 73
Phenom II DC33 3000 0.5/6.0 1640 1077 747 587 507 441 320
Pentium M DC1 1862 2.00 1850 551 340 281 255 239 231
Core 2 Duo M DC3M 1830 2.00 1866 708 459 388 348 330 329
Core i5 2467M DC33 @@@@ 0.25/3.0 2243 1239 655 484 406 344 278
Core 2 Duo 1CP DC3c 2400 4.00 2623 2088 878 567 473 440 421
Core 2 Duo 1CP DC3b 2400 4.00 2650 2176 789 497 411 374 359
Core 2 Duo 1CP DC3a 2400 4.00 2664 2174 349 183 140 131 122
Core i7 DC32 **** 0.25/8.0 3481 2364 1273 953 774 617 464
Core i7 DC33 #### 0.25/8.0 3561 2378 1312 1071 897 710 532
Core i7 QC34 $$$$ 0.25/10 5439 3263 1607 1165 926 758 603
$$$$ 3.7 GHz i7 4820K, running at up to 3.90 GHz using Turbo Boost
#### 2.8 GHz i7 860, running at up to 3.46 GHz using Turbo Boost
**** 2.8 GHz i7 930, running at up to 3.06 GHz using Turbo Boost
@@@@ 1.6 GHz i5 2467M, running at up to 2.30 GHz using Turbo Boost
|
L1 Cache Results in MBytes/Second
Integer Double/Integer
Serial Random Serial Random
MHz Read Rd/Wrt Read Rd/Wrt Read Rd/Wrt Read Rd/Wrt
Phenom II 32 bit 3000 14594 11443 13012 11353 16990 20481 17007 20246
Phenom II 64 bit 3000 12613 9165 12676 9137 17022 13705 17013 13673
Core 2 Duo 32 bit 2400 8554 12116 8562 10824 13691 15021 13691 18108
Core 2 Duo 64 bit 2400 9113 12124 9113 5262 13733 16118 13734 15867
Atom N455 32 bit 1667 3949 5146 3959 5107 5898 3910 5913 3773
Atom N455 64 bit 1667 3334 4260 3413 4213 5078 6379 5077 6377
L2 Cache Results in MBytes/Second at 96 KB
Integer Double/Integer
Serial Random Serial Random
MHz Read Rd/Wrt Read Rd/Wrt Read Rd/Wrt Read Rd/Wrt
Phenom II 32 bit 3000 11088 9912 6861 5520 16960 16197 16554 14645
Phenom II 64 bit 3000 11989 8464 6866 5221 16934 11776 16496 11888
Core 2 Duo 32 bit 2400 7629 9436 4023 3737 11362 12106 7380 6450
Core 2 Duo 64 bit 2400 8019 9473 4113 3701 11381 11971 7382 6419
Atom N455 32 bit 1667 3010 3318 811 1097 3920 3415 1502 1912
Atom N455 64 bit 1667 2539 2943 792 1070 3536 3254 1470 1918
L2/L3 Cache Results in MBytes/Second at 3072 KB
Integer Double/Integer
Serial Random Serial Random
MHz Read Rd/Wrt Read Rd/Wrt Read Rd/Wrt Read Rd/Wrt
Phenom II 32 bit 3000 7795 7357 1103 1082 9492 9093 2051 2046
Phenom II 64 bit 3000 4016 6914 1078 1045 9531 9200 2048 2043
Core 2 Duo 32 bit 2400 7609 8830 2305 2064 11277 10226 3984 3494
Core 2 Duo 64 bit 2400 7452 8113 2020 1820 10233 9482 3306 3163
RAM Results in MBytes/Second at 12288 KB
Integer Double/Integer
Serial Random Serial Random
MHz Read Rd/Wrt Read Rd/Wrt Read Rd/Wrt Read Rd/Wrt
Phenom II 32 bit 3000 4300 3595 664 561 6788 5531 1143 952
Phenom II 64 bit 3000 2771 3636 590 502 6688 5495 1012 848
Core 2 Duo 32 bit 2400 4967 2383 596 422 5829 3190 1107 738
Core 2 Duo 64 bit 2400 3257 2360 592 411 5448 3136 1090 760
Atom N455 32 bit 1667 2698 1298 54 83 3225 1685 86 131
Atom N455 64 bit 1667 2531 1249 54 83 3091 1632 86 132
RAM Results in MBytes/Second at 786432 KB
Integer Double/Integer
Serial Random Serial Random
MHz Read Rd/Wrt Read Rd/Wrt Read Rd/Wrt Read Rd/Wrt
Phenom II 32 bit 3000 4296 3562 238 212 6841 5492 396 361
Phenom II 64 bit 3000 2870 3547 192 167 6938 5499 327 283
Core 2 Duo 32 bit 2400 4922 2289 253 220 5726 3024 531 376
Core 2 Duo 64 bit 2400 3297 2277 149 152 5637 3001 297 281
On 64-Bit Ubuntu
Core 2 Duo 32 bit 2400 4879 2280 151 156 5713 3016 300 277
|
Android ResultsSome processors do not appear to run at the specified MHz and, in all cases, results from RAM are slow compared with PCs. See Android Benchmarks32.htm Later benchmarks were produced that can use ARM or Intel Native code, particularly to measure performance of Android devices using Intel Atom CPUs. The original benchmarks will run on the Atom systems but, via a code conversion layer. Old and new results are provided below and further details can be found in New Results. Note Atom improved performance using Native Intel code. The program format was also compiled to run on a Windows based tablet, using Intel processors (later compiler release than earlier PC versions). This benchmark was also run on a desktop PC. Results for these are included below, with details in New Results. |
MBytes/Second transferring 4 Byte Words
Memory Serial....... Random.......
CPU MHz KBytes Read Rd/Wrt Read Rd/Wrt
ARM 926EJ 800 16 841 1119 666 955 L1
32 222 147 83 62 L2?
65536 186 166 19 22 RAM
ARM Cortex-A9 800 16 1777 1879 1669 1809 L1
64 799 861 621 755 L2
65536 121 115 30 32 RAM
ARM Cortex-A9 1500 16 2777 3014 2671 3021 L1
64 872 1109 875 998 L2
65536 335 228 41 38 RAM
ARM Cortex-A9 1600 16 3925 4257 3881 4275 L1
64 1276 1284 1042 1182 L2
65536 369 317 75 71 RAM
ARM Cortex-A15 1700 16 2881 2478 3388 3650 L1
64 3669 2511 2201 2249 L2
65536 1973 1117 106 104 RAM
ARM Cortex-A53 1300 16 2807 3606 2753 3595 L1
32 Bit 64 2615 3266 914 1166 L2
65535 2104 1652 72 64 RAM
ARM Cortex-A53 1300 16 3865 3033 3798 3027 L1
64 Bit 64 3094 2803 1011 1077 L2
65535 1856 1609 57 68 RAM
Atom Z3745 1860 16 3434 5064 3462 5113 L1
Original 64 2837 4058 2068 2561 L2
65536 2457 2041 140 170 RAM
Atom Z3745 1860 16 4291 5626 4584 5630 L1
ARM/Intel 64 3677 4253 2629 2644 L2
Version 65536 2989 2068 141 186 RAM
Atom Z8300 1840 16 4081 5576 4641 5577 L1
Windows 32 64 3413 3923 2703 2738 L2
Bit Version 65536 2812 2735 146 160 RAM
Atom Z8300 1840 16 3075 3919 3494 4406 L1
Windows 64 64 2952 3174 2312 2405 L2
Bit Version 65536 2670 2646 142 173 RAM
Core i7-4820K 3900 16 25947 25477 25277 26063 L1
Windows 64 64 22187 22667 15256 9902 L2
Bit Version 65536 12410 10451 676 643 RAM
|
Dual Core ResultsPrograms RandMP64 and RandMP32 in DualCore.zip are compiled to run via Win64 and Win32, both manipulating 32 bit integers via C. The tests are run firstly as a single thread and secondly using two threads at the same priority level. There are four tests, Serial Read (RD), Serial Read/Write (RW), Random Read and Random Read/Write. The following is an example of log file results. Results below also include 64 bit scores on a Core 2 Duo via 64 bit Windows Vista and a Turion 64 X2 via 32 bit Vista, then 64-Bit Windows 7 on a Phenom II X4 and a Core i7. |
Athlon 64 X2 Dual Core 4200+ 2.21 GHz, DCDDR RAM, WinXP Pro X64
------------------ MBytes Per Second At --------------------
6 KB 24 KB 96 KB 384 KB 768 KB 1536 KB 12 MB 96 MB
1 Thread
Serial RD 8343 8454 5089 5092 2358 2330 2317 2299
Serial RW 3956 3962 2423 2368 1333 1328 1357 1341
Random RD 8164 8239 3732 1642 861 387 255 174
Random RW 3913 3956 2527 1403 560 239 160 134
2 Threads
Serial RD1 8434 8310 5022 5079 2032 2032 2041 2082
Serial RD2 8486 8479 5113 4957 1943 1941 1969 1945
Serial RW1 1049 1160 1110 1113 1044 858 854 855
Serial RW2 1047 1128 1082 1080 1044 862 830 815
Random RD1 8145 8063 3642 1608 480 194 128 100
Random RD2 8217 8203 3682 1625 480 196 126 101
Random RW1 423 573 458 407 353 157 85 69
Random RW2 426 588 456 407 345 154 85 69
|
|
With two threads, each has its own code and use the same data but the second thread starts at the half way point. Each has the same number of repeat passes where variations in the time taken are reflected in the relative speeds of the two threads. Running two threads on a single CPU produces the same sort of total speed as the single thread. With two CPUs, the speed of read only is mainly around double that of a single thread but speed via caches with read/write can be worse than for a single thread (or a single CPU). This can be put down to Windows flushing caches to maintain data coherency when sharing the same data array. Note that Hyper-Threading (HT) can produce performance gains with two threads. Below are available single and dual processor results for data in L1 Cache L2 Cache and RAM.
|
L1 Cache Results in MBytes/Second - 6 KB Serial
CPUs MHz RD RD1 RD2 RW RW1 RW2
Celeron 1 450 1604 1141 864 535 413 410
Pentium III 1 731 2724 1411 1426 1430 730 712
Celeron M 1 1295 4936 2409 2460 2644 1347 1319
Atom M No HT 1 1600 3594 1833 1810 1821 1181 1029
Atom M HT 1 1600 3718 2444 2425 1946 1823 1762
Pentium 4 1 1900 6776 4697 3646 2496 1631 1505
Celeron C2M DC3m 1 2000 7516 4054 3924 7855 3758 3690
Athlon XP 1 2088 7773 4510 3911 3655 1890 1832
Pentium 4E HT 1 3000 11217 5134 5095 2255 1522 1500
Turion 64 M 2 1900 5996 6865 6886 3252 901 884
Opteron 2 1992 7580 7630 7603 3586 840 843
Athlon MP 2 2000 7471 7570 7468 3542 374 368
Athlon 64 32b 2 2210 8343 8434 8486 3956 1049 1047
Athlon 64 64b 2 2210 8552 8374 8532 4346 1090 1083
Phenom II 32b 4 3000 15215 15045 15014 6524 2596 2332
Phenom II 64b 4 3000 15853 15416 15402 5572 2695 2348
Xeon P4 2 3065 11979 12009 11921 5392 924 872
Pentium 4D 2 3000 11614 11472 11533 2279 715 710
Core 2 Duo DC3M 2 1830 6972 6193 6133 6565 2002 2052
Core 2 Duo DC3a 2 2400 9477 9372 9238 9513 1090 1094
Core 2 Duo DC3b 2 2400 9524 9479 9513 9541 1490 1487
C2 Duo 32b DC3c 2 2400 8483 8699 8662 8775 2061 2049
C2 Duo 64b DC3c 2 2400 8742 8435 8460 8428 2007 2010
Corei7 64b DC32 4 **** 11183 11337 11326 10534 5892 5513
Corei7 64b DC33 4 #### 11797 11758 11751 4769 3043 2848
Cor i5 32b DC33 2 @@@@ 14891 12932 13611 11374 4416 4275
Cor i7 64b QC33 4 $$$$ 28193 28209 28385 20840 9306 8071
L1 Cache Results in MBytes/Second - 6 KB Random
CPUs MHz RD RD1 RD2 RW RW1 RW2
Celeron 1 450 1762 1014 795 804 464 447
Pentium III 1 731 2807 1406 1374 1426 641 611
Celeron M 1 1295 5017 2536 2511 2647 1351 1320
Atom M No HT 1 1600 3176 1944 1790 1818 1163 1021
Atom M HT 1 1600 3412 2631 2602 1840 1797 1790
Pentium 4 1 1900 6555 4669 3590 2373 1267 1222
Celeron C2M DC3m 1 2000 7844 4002 3963 7745 3948 3840
Athlon XP 1 2088 7527 4429 3781 3686 1928 1837
Pentium 4E HT 1 3000 11410 5125 5070 3593 1709 1677
Turion 64 M 2 1900 6592 6266 6323 3206 349 347
Opteron 2 1992 7352 7363 7378 3514 390 382
Athlon MP 2 2000 7289 7346 7274 3504 212 209
Athlon 64 32b 2 2210 8164 8145 8217 3913 423 426
Athlon 64 64b 2 2210 8176 8147 8154 4384 494 495
Phenom II 32b 4 3000 13897 14120 14142 6190 689 686
Phenom II 64b 4 3000 15116 15146 15078 5666 690 692
Xeon P4 2 3065 11694 11935 12026 4188 133 132
Pentium 4D 2 3000 11621 11467 11547 3630 191 193
Core 2 Duo DC3M 2 1830 6312 6417 6183 6189 969 997
Core 2 Duo DC3a 2 2400 9490 9450 9513 9264 460 462
Core 2 Duo DC3b 2 2400 9511 9484 9514 9386 532 534
C2 Duo 32b DC3c 2 2400 8663 8660 8617 8812 738 735
C2 Duo 64b DC3c 2 2400 8918 8576 8598 8014 730 733
Corei7 64b DC32 4 **** 11309 11090 11092 11105 1502 1501
Corei7 64b DC33 4 #### 11759 11785 11803 4777 1534 1527
Cor i5 32b DC33 2 @@@@ 14958 10758 11904 11506 1268 1267
Cor i7 64b QC33 4 $$$$ 7159 27156 27178 21208 1944 1927
DC3 DDR2 a 533 MHz nForce 570 chipset, b/c 533/800 MHz Intel 965 chipset
DC3M Mobile CPU 666 MHz DDR2, DC32 DDR3 1066 MHz, DC33 DDR3 1333 MHz
QC33 DDR3 1333 MH 4 Channels
#### 2.8 GHz i7 860, running up to 3.46 GHz using Turbo Boost - detuned
**** 2.8 GHz i7 930, running up to 3.06 GHz using Turbo Boost
@@@@ 1.6 GHz i5 2467M, running at up to 2.30 GHz using Turbo Boost
$$$$ 3.7 GHz i7 4820K, running at up to 3.90 GHz using Turbo Boost
To Start To Dual Core Index L2 Cache Results in MBytes/Second - 96 KB Serial
CPUs MHz RD RD1 RD2 RW RW1 RW2
Celeron 1 450 1153 722 576 537 290 254
Pentium III 1 731 1744 1017 1004 928 506 468
Celeron M 1 1295 3832 1960 1923 2007 1026 1002
Atom M No HT 1 1600 2761 1403 1570 1886 950 1077
Atom M HT 1 1600 2802 2248 2204 1902 1722 1706
Pentium 4 1 1900 5149 3460 2592 1609 838 814
Celeron C2M DC3m 1 2000 5432 3359 3191 6297 3163 3056
Athlon XP 1 2088 3616 2043 1813 2193 1153 1097
Pentium 4E HT 1 3000 7030 4035 4009 1429 1148 1133
Turion 64 M 2 1900 3937 3914 3939 1950 942 909
Opteron 2 1992 4522 4636 4607 2151 869 851
Athlon MP 2 2000 3571 3557 3528 2100 425 400
Athlon 64 32b 2 2210 5089 5022 5113 2423 1110 1082
Athlon 64 64b 2 2210 5115 5064 5176 2702 1110 1089
Phenom II 32b 4 3000 12532 12432 12494 6374 3105 2833
Phenom II 64b 4 3000 12645 12690 12877 5543 3041 2755
Xeon P4 2 3065 8858 8865 8455 2528 1182 1021
Pentium 4D 2 3000 7192 7191 6948 1377 762 769
Core 2 Duo DC3M 2 1830 5885 5803 5330 5688 4808 5288
Core 2 Duo DC3a 2 2400 7882 7793 7440 7573 6910 6919
Core 2 Duo DC3b 2 2400 7954 7839 7527 7601 6936 6939
C2 Duo 32b DC3c 2 2400 7859 7623 7422 7552 6754 6881
C2 Duo 64b DC3c 2 2400 7498 7334 7183 7665 6931 6789
Corei7 64b DC32 4 **** 10459 10480 10254 9884 7083 7070
Corei7 64b DC33 4 #### 10645 10609 10401 4333 3317 3304
Cor i5 32b DC33 2 @@@@ 13274 10425 9257 9181 4812 4852
Cor i7 64b QC33 4 $$$$ 18762 18746 19021 17190 9914 9913
|
Eight Thread BenchmarkRand8Thread32 and Rand8Thread64, compiled for 32 bit and 64 bit Windows, use the same 32 bit integer addressing tests as RandMP, but run via 1, 2, 4, 6 and 8 threads. The benchmarks were produced particularly for Core i7 processors that have 4 cores but properties show 8 CPUs due to the availability of Hyperthreading. For further information on these and other eight thread benchmarks see Quad Core 8 Thread.htm. There are four tests, Serial Read (RD), Serial Read/Write (RW), Random Read and Random Read/Write and each test is run using memory demands between 6 KB and 96 MB to cover all cache sizes and RAM. Below is are detailed results on a Core i7 930 processor. For these versions, a simple sum of data transfer speed is provided at each thread size to indicate total throughput. Later summaries for Quad Core 32 bits, Quad Core 64 bits, Dual Core 32 bits and Dual Core 64 bits provide just overall totals derived using harmonic mean, more suitable when considering tests that might finish at different times. The RD and RW program loops manipulate 512 words, with 32 bit and 64 bit compilations using the same type of machine code instructions. RD uses 3 MOV and 1 AND, loading data, plus 1 OR using registers, per four words. RW uses 3 MOV, loading words and 1 MOV saving results. This means that RD speed in Millions of Instructions Per Second (MIPS) can be calculate as results in MBytes per second divided by 3.2 and RW MB/second by 4.0. Serial and random tests use the same code functions, the variety depending on the data used. This benchmark uses data from the same array for all threads, but starting at different points. As with the dual core version, with RW and particularly random, flushing dedicated caches, to maintain data coherency, leads to reduced performance using more than one thread.
Here, speed using shared L2 or L3 cache can be faster than using L1 cache.
With a large L3 cache and multiple threads, some might be sharing cached data at 96 MB (like 8 each accessing 12 MB). Then measured performance can be faster than theoretical maximum memory speed.
|
Detailed Results Core i7 with 4 CPUs or 8 CPUs with Hyperthreading
RandMP 8 Thread Write/Read Test 32 bit Ver. 2.0 Sat Aug 07 13:44:54 2010
Via Microsoft 32-bit C/C++ Optimizing Compiler Version 13.10.3077 for 80x86
------------------ MBytes Per Second At --------------------
6 KB 24 KB 96 KB 384 KB 768 KB 1536 KB 12 MB 96 MB
1 Thread
Serial RD 11458 11623 10380 8832 8809 8828 4382 4266
Serial RW 15282 15637 12223 10243 10216 10234 5493 4533
Random RD 11266 11679 6233 4391 3778 3499 993 459
Random RW 14375 14739 5887 4079 3401 3104 888 428
2 Threads
Serial RD1 11326 11592 10401 8711 8705 8702 5599 4356
Serial RD2 11335 11620 10431 9059 9033 9048 5599 4356
Total 22661 23211 20831 17771 17738 17751 11198 8713
Serial RW1 6782 7401 9109 9096 9646 9636 4281 3828
Serial RW2 6944 7412 9107 9082 9654 9630 4281 3828
Total 13726 14814 18216 18178 19299 19266 8562 7656
Random RD1 11273 11674 6236 4397 3757 3462 993 455
Random RD2 11275 11659 6227 4362 3742 3453 993 454
Total 22547 23334 12462 8759 7499 6915 1986 909
Random RW1 1513 1894 2277 2596 2503 2303 901 430
Random RW2 1514 1893 2278 2593 2504 2301 901 430
Total 3027 3787 4555 5188 5006 4604 1801 859
4 Threads
Serial RD1 8293 8062 7719 7331 6918 7523 5446 4389
Serial RD2 8839 8959 9108 7089 8470 6692 5557 4187
Serial RD3 10604 9296 8711 6691 6647 7035 5330 4389
Serial RD4 9620 9713 7526 8162 7613 8631 5268 4389
Total 37356 36030 33064 29273 29649 29880 21601 17354
Serial RW1 4097 5298 6530 5974 5773 5427 3824 3469
Serial RW2 4152 5215 6310 5858 5740 5509 3889 3469
Serial RW3 4032 5236 6188 6110 5753 5435 3873 3469
Serial RW4 3964 5804 6308 6584 5817 5562 3889 3469
Total 16245 21554 25336 24527 23084 21933 15476 13877
Random RD1 9402 8834 4923 3307 2714 2356 752 315
Random RD2 8786 9787 4787 3227 2629 2081 817 303
Random RD3 9795 9349 5057 3850 2669 2090 713 309
Random RD4 10324 11676 5323 3950 2810 2828 875 314
Total 38308 39647 20091 14333 10822 9355 3157 1241
Random RW1 699 1160 1496 1516 1449 1147 481 235
Random RW2 691 1197 1578 1515 1338 1218 513 221
Random RW3 692 1162 1514 1490 1355 1172 525 222
Random RW4 698 1169 1531 1716 1342 1186 545 222
Total 2780 4688 6118 6237 5484 4723 2064 900
6 Threads
Serial RD1 7228 9425 7500 6138 6074 6308 4625 4063
Serial RD2 7435 8034 6273 6162 5822 7160 4740 4257
Serial RD3 6854 7596 7057 7605 6072 5563 4950 4257
Serial RD4 7096 6144 7299 6109 6395 6708 5791 4063
Serial RD5 8082 10203 6774 5891 6349 5738 4338 4063
Serial RD6 7139 7402 6944 7957 6210 7361 5543 4257
Total 43834 48803 41847 39862 36922 38837 29988 24959
Serial RW1 2911 3875 5080 4635 4441 4071 3614 2403
Serial RW2 2682 4091 4536 4563 3761 4084 3578 2412
Serial RW3 2627 4055 4677 5034 3957 4070 3537 2338
Serial RW4 2778 3628 4818 4768 4132 4077 3640 2286
Serial RW5 2634 3880 4536 4592 4116 4073 3652 2648
Serial RW6 2602 3697 4508 5291 3850 4075 3568 2488
Total 16234 23227 28154 28883 24256 24450 21590 14574
Random RD1 6738 9375 4375 3114 2046 1717 539 257
Random RD2 8171 10302 3623 2616 1950 1504 540 247
Random RD3 9455 7173 3655 3113 2087 2094 580 202
Random RD4 8217 7245 4984 3054 2556 1702 526 215
Random RD5 6579 7257 4406 2855 1894 1594 522 257
Random RD6 7176 7586 4255 3815 2122 1516 522 232
Total 46337 48938 25298 18566 12655 10126 3228 1410
Random RW1 471 846 1068 1184 968 822 380 152
Random RW2 471 848 1135 1142 1047 870 419 164
Random RW3 474 872 1157 1101 965 810 415 148
Random RW4 504 877 1118 1072 1131 798 382 150
Random RW5 482 903 1063 1065 1075 842 394 183
Random RW6 501 846 1160 1194 924 916 390 156
Total 2903 5193 6701 6757 6110 5058 2380 953
8 Threads
Serial RD1 5604 7709 5506 5176 6162 5130 4596 3240
Serial RD2 6231 6084 5566 5201 4873 5023 5003 3240
Serial RD3 6602 5767 5635 5077 5420 5328 4745 3240
Serial RD4 5561 5630 5031 5106 4916 4802 4877 3240
Serial RD5 5375 5496 5269 4939 4647 4946 4270 3132
Serial RD6 5395 5461 4808 4966 4453 4659 4366 3132
Serial RD7 6190 5644 5902 5128 5846 5024 4383 3240
Serial RD8 5653 6412 5284 5229 5191 5522 3989 3132
Total 46611 48204 43002 40823 41507 40435 36230 25594
Serial RW1 2973 3628 3348 3007 2935 3216 3115 1782
Serial RW2 2253 2896 3521 3119 3252 3282 2791 1764
Serial RW3 2240 2906 3450 2994 3014 3440 2807 1640
Serial RW4 2217 2876 3496 2967 2958 3198 2877 1616
Serial RW5 2120 2719 3264 2997 2917 3145 2651 1616
Serial RW6 2123 2687 3234 2894 2901 3169 2825 1640
Serial RW7 2408 3460 3276 3002 3523 3468 2674 1782
Serial RW8 2240 2886 3486 3030 3074 3474 2744 1579
Total 18574 24058 27074 24010 24573 26394 22484 13419
Random RD1 6902 6734 3526 2528 1968 1332 442 192
Random RD2 6073 6120 3179 2434 2071 1239 448 164
Random RD3 6298 6153 3097 2571 1801 1287 436 178
Random RD4 5666 5665 3116 2317 1967 1154 432 166
Random RD5 5299 5705 3174 2263 2068 1132 441 159
Random RD6 5383 5349 2948 2122 1731 1168 434 161
Random RD7 6171 5552 3296 2526 1795 1162 440 190
Random RD8 5673 6012 3403 2267 1764 1095 441 161
Total 47465 47291 25740 19028 15166 9568 3514 1369
Random RW1 403 636 877 851 885 637 374 172
Random RW2 402 643 940 1048 877 611 345 128
Random RW3 404 641 896 861 802 637 405 133
Random RW4 410 637 914 860 781 606 365 116
Random RW5 403 647 882 860 767 594 315 113
Random RW6 411 649 957 883 765 609 324 114
Random RW7 439 647 895 1134 858 611 419 155
Random RW8 428 646 926 867 822 631 327 117
Total 3298 5146 7286 7364 6558 4935 2874 1049
|
Results in MBytes/Second 32 Bit Version - Quad Core ProcessorsFor these and later results, data sizes are 6 KB for L1 cache, 96 KB for L2 cache, 1536 KB for L3 cache (where appropriate) and 96 MB for RAM. Also, in all cases, performance gains on using 2, 4, 6 and 8 threads are also shown. Hyperthreading does not appear to improve throughput much on the faster reading speeds but the one shown might be running at 3060 MHz on using one or two cores and 2800 MHz when four CPUs are in use. The exception is reading from RAM which is slow and CPU speed limited using one thread. The four core Phenom comes out slightly faster on reading from caches, but is much slower on other tests.
|
CPUs Number Of Threads Gain At Threads
/HTs 1 2 4 6 8 2 4 6 8
Serial RD
Core i7 4/8 L1 11458 22661 37039 43717 46374 2.0 3.2 3.8 4.0
930 L2 10380 20832 32853 41711 42839 2.0 3.2 4.0 4.1
#### MHz L3 8828 17743 29610 38414 40330 2.0 3.4 4.4 4.6
Win 764 RAM 4266 8712 17347 24946 25589 2.0 4.1 5.8 6.0
Serial RW
Core i7 4/8 L1 15282 13724 16240 16209 18379 0.9 1.1 1.1 1.2
930 L2 12223 18216 25326 28104 27047 1.5 2.1 2.3 2.2
#### MHz L3 10234 19266 21931 24450 26351 1.9 2.1 2.4 2.6
Win 764 RAM 4533 7656 13876 14543 13390 1.7 3.1 3.2 3.0
Random RD
Core i7 4/8 L1 11266 22548 38174 45592 47141 2.0 3.4 4.0 4.2
930 L2 6233 12463 20059 24986 25667 2.0 3.2 4.0 4.1
#### MHz L3 3499 6915 9211 10002 9531 2.0 2.6 2.9 2.7
Win 764 RAM 459 909 1241 1398 1364 2.0 2.7 3.0 3.0
Random RW
Core i7 4/8 L1 14375 3027 2780 2901 3297 0.2 0.2 0.2 0.2
930 L2 5887 4555 6117 6693 7281 0.8 1.0 1.1 1.2
#### MHz L3 3104 4604 4721 5047 4933 1.5 1.5 1.6 1.6
Win 764 RAM 428 860 899 948 1026 2.0 2.1 2.2 2.4
#### 2.8 GHz running at up to 3.06 GHz via Turbo Boost, dual channel 1066 MHz DDR3 RAM
##################################################################################
CPUs Number Of Threads Gain At Threads
/HTs 1 2 4 6 8 2 4 6 8
Serial RD
Core i7 4/8 L1 28442 57130 114198 114435 107457 2.0 4.0 4.0 3.8
4820K L2 20531 41075 82142 87468 92156 2.0 4.0 4.3 4.5
$$$$ MHz L3 17015 34734 69551 77040 81525 2.0 4.1 4.5 4.8
Win 8.1 RAM 6004 12438 25044 38420 42316 2.1 4.2 6.4 7.0
Serial RW
Core i7 4/8 L1 30091 21439 20928 24068 28856 0.7 0.7 0.8 1.0
4820K L2 22100 20942 38196 48821 53497 0.9 1.7 2.2 2.4
$$$$ MHz L3 17341 33271 65558 60361 73659 1.9 3.8 3.5 4.2
Win 8.1 RAM 10680 21454 42836 50906 53162 2.0 4.0 4.8 5.0
Random RD
Core i7 4/8 L1 27862 55813 111471 111534 104011 2.0 4.0 4.0 3.7
4820K L2 13514 27231 54374 54880 59899 2.0 4.0 4.1 4.4
$$$$ MHz L3 5557 11141 20900 21977 14510 2.0 3.8 4.0 2.6
Win 8.1 RAM 627 1238 2472 2533 2479 2.0 3.9 4.0 4.0
Random RW
Core i7 4/8 L1 29930 3734 3215 4134 5002 0.1 0.1 0.1 0.2
4820K L2 9374 5108 8194 8510 9159 0.5 0.9 0.9 1.0
$$$$ MHz L3 4759 7101 12497 13962 13291 1.5 2.6 2.9 2.8
Win 8.1 RAM 588 1256 2496 2526 2521 2.1 4.2 4.3 4.3
$$$$ 3.7 GHz running at up to 3.9 GHz via Turbo Boost, quad channel 1600 MHz DDR3 RAM
RAM max throughput 51.2 GB/second
##################################################################################
CPUs Number Of Threads Gain At Threads
/HTs 1 2 4 6 8 2 4 6 8
Serial RD
Phenom II 4/0 L1 15212 29350 58904 58896 54909 1.9 3.9 3.9 3.6
3000 MHz L2 12236 24767 49039 50798 47318 2.0 4.0 4.2 3.9
Win 764 L3 8148 16402 30391 33436 32457 2.0 3.7 4.1 4.0
1333 MHz DDR3 RAM 3917 6983 11299 12484 12002 1.8 2.9 3.2 3.1
Serial RW
Phenom II 4/0 L1 7741 5100 5750 6598 6517 0.7 0.7 0.9 0.8
3000 MHz L2 7998 5906 7479 8466 8345 0.7 0.9 1.1 1.0
Win 764 L3 7132 13142 7489 8566 8582 1.8 1.1 1.2 1.2
1333 MHz DDR3 RAM 3589 5981 8576 7913 7802 1.7 2.4 2.2 2.2
Random RD
Phenom II 4/0 L1 14367 27877 56817 55300 54129 1.9 4.0 3.8 3.8
3000 MHz L2 7250 14355 28436 29723 27962 2.0 3.9 4.1 3.9
Win 764 L3 1560 3419 6641 7403 7410 2.2 4.3 4.7 4.8
1333 MHz DDR3 RAM 339 679 1140 1336 1242 2.0 3.4 3.9 3.7
Random RW
Phenom II 4/0 L1 7585 1381 752 833 757 0.2 0.1 0.1 0.1
3000 MHz L2 5985 1624 1230 1387 1245 0.3 0.2 0.2 0.2
Win 764 L3 1505 1724 1377 1545 1572 1.1 0.9 1.0 1.0
1333 MHz DDR3 RAM 313 634 1113 1157 1153 2.0 3.6 3.7 3.7
|
Results in MBytes/Second 64 Bit Version - Quad Core ProcessorsResults of the program compiled for 64 bit working are mainly similar to those at 32 bits but the additional 64 bit optimisation facilities appear to produce some improvements.
|
CPUs Number Of Threads Gain At Threads
/HTs 1 2 4 6 8 2 4 6 8
Serial RD
Core i7 4/8 L1 11343 22851 38716 47028 46794 2.0 3.4 4.1 4.1
930 L2 10431 20843 33916 43626 41990 2.0 3.3 4.2 4.0
#### MHz L3 10091 20142 33559 40724 40403 2.0 3.3 4.0 4.0
Win 764 RAM 7346 13768 16172 18725 26284 1.9 2.2 2.5 3.6
Serial RW
Core i7 4/8 L1 15324 13602 16083 16944 17407 0.9 1.0 1.1 1.1
930 L2 12269 18093 24448 28394 25193 1.5 2.0 2.3 2.1
#### MHz L3 10196 19278 21970 24410 25561 1.9 2.2 2.4 2.5
Win 764 RAM 4537 7670 13752 15016 15210 1.7 3.0 3.3 3.4
Random RD
Core i7 4/8 L1 11339 22930 38319 44619 47599 2.0 3.4 3.9 4.2
930 L2 6203 12407 21751 25326 26476 2.0 3.5 4.1 4.3
#### MHz L3 3508 6938 9323 9628 9612 2.0 2.7 2.7 2.7
Win 764 RAM 457 909 1277 1449 1332 2.0 2.8 3.2 2.9
Random RW
Core i7 4/8 L1 14560 3040 2907 2986 3392 0.2 0.2 0.2 0.2
930 L2 5887 4560 5788 6368 6860 0.8 1.0 1.1 1.2
#### MHz L3 3108 4622 4741 5038 5091 1.5 1.5 1.6 1.6
Win 764 RAM 428 860 905 910 1035 2.0 2.1 2.1 2.4
#### 2.8 GHz running at up to 3.06 GHz via Turbo Boost, dual channel 1066 MHz DDR3 RAM
##################################################################################
CPUs Number Of Threads Gain At Threads
/HTs 1 2 4 6 8 2 4 6 8
Serial RD
Core i7 4/8 L1 30475 61153 116516 113013 110888 2.0 3.8 3.7 3.6
4820K L2 18350 36843 73499 84322 95238 2.0 4.0 4.6 5.2
$$$$ MHz L3 17595 35297 70897 79454 87379 2.0 4.0 4.5 5.0
Win 8.1 RAM 11243 21158 41568 49434 63741 1.9 3.7 4.4 5.7
Serial RW
Core i7 4/8 L1 30195 21994 20774 24621 28160 0.7 0.7 0.8 0.9
4820K L2 22013 20967 38023 49331 54800 1.0 1.7 2.2 2.5
$$$$ MHz L3 17482 33428 65635 69907 72721 1.9 3.8 4.0 4.2
Win 8.1 RAM 10642 21508 42876 52409 49282 2.0 4.0 4.9 4.6
Random RD
Core i7 4/8 L1 28916 57863 110282 109650 107971 2.0 3.8 3.8 3.7
4820K L2 13124 26154 52010 59322 59694 2.0 4.0 4.5 4.5
$$$$ MHz L3 5669 11248 22295 22289 14663 2.0 3.9 3.9 2.6
Win 8.1 RAM 655 1302 2540 2652 2895 2.0 3.9 4.0 4.4
Random RW
Core i7 4/8 L1 30726 3761 3324 4318 5143 0.1 0.1 0.1 0.2
4820K L2 9498 5132 8197 8345 9152 0.5 0.9 0.9 1.0
$$$$ MHz L3 4813 7115 12507 14545 13838 1.5 2.6 3.0 2.9
Win 8.1 RAM 608 1282 2492 2718 2526 2.1 4.1 4.5 4.2
$$$$ 3.7 GHz running at up to 3.9 GHz via Turbo Boost, quad channel 1600 MHz DDR3 RAM
RAM max throughput 51.2 GB/second
##################################################################################
CPUs Number Of Threads Gain At Threads
/HTs 1 2 4 6 8 2 4 6 8
Serial RD
Phenom II 4/0 L1 16053 31565 62181 62063 62100 2.0 3.9 3.9 3.9
3000 MHz L2 13037 26141 51228 50205 46731 2.0 3.9 3.9 3.6
Win 764 L3 8700 17107 31347 32866 32859 2.0 3.6 3.8 3.8
1333 MHz DDR3 RAM 4421 7598 12673 13618 12287 1.7 2.9 3.1 2.8
Serial RW
Phenom II 4/0 L1 8791 5197 5850 6682 7155 0.6 0.7 0.8 0.8
3000 MHz L2 8762 5760 7418 8389 8791 0.7 0.8 1.0 1.0
Win 764 L3 7403 13461 7520 8457 7539 1.8 1.0 1.1 1.0
1333 MHz DDR3 RAM 3678 6072 8260 7244 8353 1.7 2.2 2.0 2.3
Random RD
Phenom II 4/0 L1 14830 30151 59339 61903 57432 2.0 4.0 4.2 3.9
3000 MHz L2 7462 14913 29571 29882 30293 2.0 4.0 4.0 4.1
Win 764 L3 1751 3445 6655 7482 7752 2.0 3.8 4.3 4.4
1333 MHz DDR3 RAM 326 574 1072 1337 1244 1.8 3.3 4.1 3.8
Random RW
Phenom II 4/0 L1 8760 1328 728 812 859 0.2 0.1 0.1 0.1
3000 MHz L2 6095 1597 1215 1368 1454 0.3 0.2 0.2 0.2
Win 764 L3 1492 1703 1406 1561 1635 1.1 0.9 1.0 1.1
1333 MHz DDR3 RAM 296 596 1046 1150 1236 2.0 3.5 3.9 4.2
|
Results in MBytes/Second 32 Bit Version - Dual Core ProcessorsOverall throughput is similar using between two and eight threads.
|
CPUs Number Of Threads Gain At Threads
/HTs 1 2 4 6 8 2 4 6 8
Serial RD
Athlon 64 2/0 L1 8318 16774 16913 19391 16262 2.0 2.0 2.3 2.0
2210 MHz L2 5054 10102 10161 10045 9939 2.0 2.0 2.0 2.0
XP x64 RAM 2026 3706 4441 3746 3723 1.8 2.2 1.8 1.8
Serial RW
Athlon 64 2/0 L1 5966 2517 2469 2478 2500 0.4 0.4 0.4 0.4
2210 MHz L2 3160 2346 2289 2676 2285 0.7 0.7 0.8 0.7
XP x64 RAM 1347 1680 1677 1728 1731 1.2 1.2 1.3 1.3
Random RD
Athlon 64 2/0 L1 8105 16209 16427 18271 17636 2.0 2.0 2.3 2.2
2210 MHz L2 3653 7249 8785 7196 7174 2.0 2.4 2.0 2.0
XP x64 RAM 158 203 201 204 229 1.3 1.3 1.3 1.4
Random RW
Athlon 64 2/0 L1 5753 949 973 968 974 0.2 0.2 0.2 0.2
2210 MHz L2 2966 939 938 946 944 0.3 0.3 0.3 0.3
XP x64 RAM 129 137 142 157 152 1.1 1.1 1.2 1.2
##################################################################################
CPUs Number Of Threads Gain At Threads
/HTs 1 2 4 6 8 2 4 6 8
Serial RD
Core 2 Duo 2/0 L1 9385 18492 21596 18560 20219 2.0 2.3 2.0 2.2
2400 MHz L2 7524 14404 17247 14423 14551 1.9 2.3 1.9 1.9
Vista 64 RAM 3545 5325 5613 6239 6067 1.5 1.6 1.8 1.7
Serial RW
Core 2 Duo 2/0 L1 12597 4855 6590 6640 6513 0.4 0.5 0.5 0.5
2400 MHz L2 8410 14668 14655 17085 17142 1.7 1.7 2.0 2.0
Vista 64 RAM 2322 2751 2932 2956 2644 1.2 1.3 1.3 1.1
Random RD
Core 2 Duo 2/0 L1 9442 18079 22175 20940 18294 1.9 2.3 2.2 1.9
2400 MHz L2 4198 6863 8371 8068 7914 1.6 2.0 1.9 1.9
Vista 64 RAM 452 704 709 762 761 1.6 1.6 1.7 1.7
Random RW
Core 2 Duo L1 12386 2094 2962 2556 2684 0.2 0.2 0.2 0.2
2400 MHz 2/0 L2 3660 2805 3250 3178 3107 0.8 0.9 0.9 0.8
Vista 64 RAM 285 500 482 501 496 1.8 1.7 1.8 1.7
##################################################################################
CPUs Number Of Threads Gain At Threads
/HTs 1 2 4 6 8 2 4 6 8
Serial RD
Core 2 Duo 2/0 L1 5872 12130 13588 13538 13451 2.1 2.3 2.3 2.3
1830 MHz L2 5345 10235 12592 12827 11169 1.9 2.4 2.4 2.1
Vista 32 RAM 2624 3922 4127 4120 4061 1.5 1.6 1.6 1.5
Serial RW
Core 2 Duo 2/0 L1 8380 4482 5292 5671 5979 0.5 0.6 0.7 0.7
1830 MHz L2 6338 10893 11364 12136 12625 1.7 1.8 1.9 2.0
Vista 32 RAM 1587 1750 1798 2101 1916 1.1 1.1 1.3 1.2
Random RD
Core 2 Duo 2/0 L1 6107 12516 13395 14362 13570 2.0 2.2 2.4 2.2
1830 MHz L2 3174 5165 5262 6230 5329 1.6 1.7 2.0 1.7
Vista 32 RAM 361 600 540 604 532 1.7 1.5 1.7 1.5
Random RW
Core 2 Duo L1 8448 2278 2543 2887 2419 0.3 0.3 0.3 0.3
1830 MHz 2/0 L2 2741 2879 2857 2988 3044 1.1 1.0 1.1 1.1
Vista 32 RAM 227 318 334 399 353 1.4 1.5 1.8 1.6
|
Results in MBytes/Second 64 Bit Version - Dual Core ProcessorsResults of the program compiled for 64 bit working are again mainly similar to those at 32 bits.
|
CPUs Number Of Threads Gain At Threads
/HTs 1 2 4 6 8 2 4 6 8
Serial RD
Athlon 64 2/0 L1 8465 17083 17086 19488 18481 2.0 2.0 2.3 2.2
2210 MHz L2 5151 10287 10291 10180 10019 2.0 2.0 2.0 1.9
XP x64 RAM 2311 4069 4043 4054 4032 1.8 1.7 1.8 1.7
Serial RW
Athlon 64 2/0 L1 6976 2447 2454 2848 2488 0.4 0.4 0.4 0.4
2210 MHz L2 3757 2257 2270 2271 2397 0.6 0.6 0.6 0.6
XP x64 RAM 1485 1672 1686 1681 1717 1.1 1.1 1.1 1.2
Random RD
Athlon 64 2/0 L1 8163 16258 16340 16155 15659 2.0 2.0 2.0 1.9
2210 MHz L2 3722 7407 8846 7381 7304 2.0 2.4 2.0 2.0
XP x64 RAM 160 194 193 226 193 1.2 1.2 1.4 1.2
Random RW
Athlon 64 2/0 L1 6972 963 970 981 990 0.1 0.1 0.1 0.1
2210 MHz L2 3348 960 931 1126 953 0.3 0.3 0.3 0.3
XP x64 RAM 130 142 159 135 144 1.1 1.2 1.0 1.1
##################################################################################
CPUs Number Of Threads Gain At Threads
/HTs 1 2 4 6 8 2 4 6 8
Serial RD
Core 2 Duo 2/0 L1 9446 18858 18786 18540 18261 2.0 2.0 2.0 1.9
2400 MHz L2 7594 14450 14615 14825 14681 1.9 1.9 2.0 1.9
Vista 64 RAM 4989 6584 6217 6172 6467 1.3 1.2 1.2 1.3
Serial RW
Core 2 Duo 2/0 L1 12633 4701 4944 6003 5170 0.4 0.4 0.5 0.4
2400 MHz L2 8840 14387 14909 14908 15024 1.6 1.7 1.7 1.7
Vista 64 RAM 2369 2470 2532 2621 2675 1.0 1.1 1.1 1.1
Random RD
Core 2 Duo 2/0 L1 9523 18930 18480 18569 18277 2.0 1.9 1.9 1.9
2400 MHz L2 4238 6943 7080 7081 7066 1.6 1.7 1.7 1.7
Vista 64 RAM 458 658 692 684 681 1.4 1.5 1.5 1.5
Random RW
Core 2 Duo L1 11723 1967 1988 2071 2514 0.2 0.2 0.2 0.2
2400 MHz 2/0 L2 3734 2939 2991 3163 3018 0.8 0.8 0.8 0.8
Vista 64 RAM 305 451 436 455 456 1.5 1.4 1.5 1.5
##################################################################################
CPUs Number Of Threads Gain At Threads
/HTs 1 2 4 6 8 2 4 6 8
Serial RD
i5 2467M 2/2 L1 9380 23789 24118 29026 29218 2.5 2.6 3.1 3.1
@@@@ MHz L2 6201 14373 19259 23131 22407 2.3 3.1 3.7 3.6
Win 764 L3 6064 14018 17637 20402 22137 2.3 2.9 3.4 3.7
RAM 5000 9773 13658 16524 16732 2.0 2.7 3.3 3.3
Serial RW
i5 2467M 2/2 L1 11689 9174 11952 15014 16080 0.8 1.0 1.3 1.4
@@@@ MHz L2 8786 8650 12880 15862 18424 1.0 1.5 1.8 2.1
Win 764 L3 6875 11868 16812 20080 17831 1.7 2.4 2.9 2.6
RAM 5507 8444 11121 10860 11210 1.5 2.0 2.0 2.0
Random RD
i5 2467M 2/2 L1 11395 21719 23203 27865 29838 1.9 2.0 2.4 2.6
@@@@ MHz L2 5282 9947 12625 15696 15497 1.9 2.4 3.0 2.9
Win 764 L3 2555 4988 3167 3531 3687 2.0 1.2 1.4 1.4
RAM 291 575 558 677 644 2.0 1.9 2.3 2.2
Random RW
i5 2467M 2/2 L1 11888 1995 2265 2517 2637 0.2 0.2 0.2 0.2
@@@@ MHz L2 3754 2369 3839 4742 4602 0.6 1.0 1.3 1.2
Win 764 L3 1995 3282 3117 3498 3759 1.6 1.6 1.8 1.9
RAM 282 526 543 666 645 1.9 1.9 2.4 2.3
@@@@ 1.6 GHz i5 2467M, running at up to 2.30 GHz using Turbo Boost
|
Roy Longbottom January 2016