Hello guys,
I’ve made some benchmarks. Earlier (2018) Rock Pi was only with small heatsink (1cm x 1cm) and tiny fan (25 mm). Today’s test (2021) is with dedicated heatsink (the big one!) and a 50 mm fan.
See results below. There is performance increase! 7-Zip has changed so it may be not the same code.
2018:
2018-12-18
$ stress-ng --cpu 1 --cpu-method matrixprod --metrics-brief -t 60
stress-ng: info: [1125] dispatching hogs: 1 cpu
stress-ng: info: [1125] cache allocate: using built-in defaults as unable to determine cache details
stress-ng: info: [1125] successful run completed in 60.22s (1 min, 0.22 secs)
stress-ng: info: [1125] stressor bogo ops real time usr time sys time bogo ops/s bogo ops/s
stress-ng: info: [1125] (secs) (secs) (secs) (real time) (usr+sys time)
stress-ng: info: [1125] cpu 253 60.22 60.19 0.00 4.20 4.20
$ stress-ng --cpu 6 --cpu-method matrixprod --metrics-brief -t 60
stress-ng: info: [1147] dispatching hogs: 6 cpu
stress-ng: info: [1147] cache allocate: using built-in defaults as unable to determine cache details
stress-ng: info: [1147] successful run completed in 60.30s (1 min, 0.30 secs)
stress-ng: info: [1147] stressor bogo ops real time usr time sys time bogo ops/s bogo ops/s
stress-ng: info: [1147] (secs) (secs) (secs) (real time) (usr+sys time)
stress-ng: info: [1147] cpu 793 60.20 359.21 0.08 13.17 2.21
$ sysbench --test=cpu --cpu-max-prime=20000 --num-threads=1 --max-time=60 run
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
WARNING: --max-time is deprecated, use --time instead
sysbench 1.0.11 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
Prime numbers limit: 20000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 701.24
General statistics:
total time: 60.0014s
total number of events: 42084
Latency (ms):
min: 1.42
avg: 1.42
max: 6.72
95th percentile: 1.42
sum: 59964.36
Threads fairness:
events (avg/stddev): 42084.0000/0.00
execution time (avg/stddev): 59.9644/0.00
$ sysbench --test=cpu --cpu-max-prime=20000 --num-threads=6 --max-time=60 run
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
WARNING: --num-threads is deprecated, use --threads instead
WARNING: --max-time is deprecated, use --time instead
sysbench 1.0.11 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 6
Initializing random number generator from current time
Prime numbers limit: 20000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 2237.70
General statistics:
total time: 60.0024s
total number of events: 134280
Latency (ms):
min: 1.42
avg: 2.68
max: 22.67
95th percentile: 3.62
sum: 359842.66
Threads fairness:
events (avg/stddev): 22380.0000/8091.23
execution time (avg/stddev): 59.9738/0.01
$ 7z b
7-Zip [64] 16.02 : Copyright (c) 1999-2016 Igor Pavlov : 2016-05-21
p7zip Version 16.02 (locale=pl_PL.UTF-8,Utf16=on,HugeFiles=on,64 bits,6 CPUs LE)
LE
CPU Freq: 1064 979 894 1260 1651 1792 1792 1792 1792
RAM size: 3812 MB, # CPU hardware threads: 6
RAM usage: 1323 MB, # Benchmark threads: 6
Compressing | Decompressing
Dict Speed Usage R/U Rating | Speed Usage R/U Rating
KiB/s % MIPS MIPS | KiB/s % MIPS MIPS
22: 4218 478 858 4104 | 92601 523 1509 7897
23: 4035 494 833 4112 | 87611 544 1394 7581
24: 3793 516 791 4079 | 84930 564 1323 7455
25: 3592 530 774 4102 | 81479 586 1237 7251
---------------------------------- | ------------------------------
Avr: 504 814 4099 | 554 1365 7546
Tot: 529 1090 5823
2021:
2021-11-25
$ stress-ng --cpu 1 --cpu-method matrixprod --metrics-brief -t 60
stress-ng: info: [2154] dispatching hogs: 1 cpu
stress-ng: info: [2154] successful run completed in 60.09s (1 min, 0.09 secs)
stress-ng: info: [2154] stressor bogo ops real time usr time sys time bogo ops/s bogo ops/s
stress-ng: info: [2154] (secs) (secs) (secs) (real time) (usr+sys time)
stress-ng: info: [2154] cpu 264 60.09 60.07 0.00 4.39 4.39
$ stress-ng --cpu 6 --cpu-method matrixprod --metrics-brief -t 60
stress-ng: info: [2105] dispatching hogs: 6 cpu
stress-ng: info: [2105] successful run completed in 60.23s (1 min, 0.23 secs)
stress-ng: info: [2105] stressor bogo ops real time usr time sys time bogo ops/s bogo ops/s
stress-ng: info: [2105] (secs) (secs) (secs) (real time) (usr+sys time)
stress-ng: info: [2105] cpu 935 60.14 354.24 0.03 15.55 2.64
$ sysbench --test=cpu --cpu-max-prime=20000 --num-threads=1 --max-time=60 run
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
WARNING: --max-time is deprecated, use --time instead
sysbench 1.0.18 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
Prime numbers limit: 20000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 704.18
General statistics:
total time: 60.0011s
total number of events: 42254
Latency (ms):
min: 1.42
avg: 1.42
max: 3.18
95th percentile: 1.42
sum: 59965.61
Threads fairness:
events (avg/stddev): 42254.0000/0.00
execution time (avg/stddev): 59.9656/0.00
$ sysbench --test=cpu --cpu-max-prime=20000 --num-threads=6 --max-time=60 run
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
WARNING: --num-threads is deprecated, use --threads instead
WARNING: --max-time is deprecated, use --time instead
sysbench 1.0.18 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 6
Initializing random number generator from current time
Prime numbers limit: 20000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 2470.14
General statistics:
total time: 60.0017s
total number of events: 148219
Latency (ms):
min: 1.41
avg: 2.43
max: 34.67
95th percentile: 3.55
sum: 359831.26
Threads fairness:
events (avg/stddev): 24703.1667/7799.87
execution time (avg/stddev): 59.9719/0.01
$ 7z b
7-Zip (z) 21.03 beta (arm64) : Copyright (c) 1999-2021 Igor Pavlov : 2021-07-20
64-bit arm_v:8 locale=pl_PL.UTF-8 Threads:6, ASM
Compiler: 9.2.1 20191025 GCC 9.2.1
Linux : 4.4.154-113-rockchip-gdb9dfc2cdd25 : #1 SMP Sat Aug 14 09:26:12 UTC 2021 : aarch64
PageSize:4KB hwcap:FF:CRC32:SHA1:SHA2:AES:ASIMD
LE
1T CPU Freq (MHz): 935 1301 1699 1797 1797 1797 1797
3T CPU Freq (MHz): 259% 1425 258% 1438
RAM size: 3811 MB, # CPU hardware threads: 6
RAM usage: 1334 MB, # Benchmark threads: 6
Compressing | Decompressing
Dict Speed Usage R/U Rating | Speed Usage R/U Rating
KiB/s % MIPS MIPS | KiB/s % MIPS MIPS
22: 5437 531 995 5290 | 130294 492 2256 11109
23: 5187 540 978 5286 | 129758 509 2205 11225
24: 4604 509 973 4951 | 125405 506 2174 11004
25: 4749 555 977 5422 | 121866 514 2108 10845
---------------------------------- | ------------------------------
Avr: 4994 534 980 5237 | 126831 505 2185 11046
Tot: 520 1583 8142