もりねずみーのやつ

インターネット上でしか生きられません

Raspberry Pi 3 B+でベンチマーク

結果

819.0 ※ただし1.4GHzを維持出来た場合

https://twitter.com/morinezumiiii/status/1007884693046378496

ラズパイ2などの結果は以下参照。

Raspberry Pi 2での油冷オーバークロック - もりねずみーのやつ

Raspberry PiでNode.js(ミニラックサーバ構築編)

ログ

[morinezumiiii@morinezumiiii-pi unixbench]$ ./Run
sh: 3dinfo: command not found

   #    #  #    #  #  #    #          #####   ######  #    #   ####   #    #
   #    #  ##   #  #   #  #           #    #  #       ##   #  #    #  #    #
   #    #  # #  #  #    ##            #####   #####   # #  #  #       ######
   #    #  #  # #  #    ##            #    #  #       #  # #  #       #    #
   #    #  #   ##  #   #  #           #    #  #       #   ##  #    #  #    #
    ####   #    #  #  #    #          #####   ######  #    #   ####   #    #

   Version 5.1.3                      Based on the Byte Magazine Unix Benchmark

   Multi-CPU version                  Version 5 revisions by Ian Smith,
                                      Sunnyvale, CA, USA
   January 13, 2011                   johantheghost at yahoo period com

Use of uninitialized value in printf at ./Run line 1379.
Use of uninitialized value in printf at ./Run line 1380.
Use of uninitialized value in printf at ./Run line 1379.
Use of uninitialized value in printf at ./Run line 1380.
Use of uninitialized value in printf at ./Run line 1379.
Use of uninitialized value in printf at ./Run line 1380.
Use of uninitialized value in printf at ./Run line 1379.
Use of uninitialized value in printf at ./Run line 1380.
Use of uninitialized value in printf at ./Run line 1589.
Use of uninitialized value in printf at ./Run line 1590.
Use of uninitialized value in printf at ./Run line 1589.
Use of uninitialized value in printf at ./Run line 1590.
Use of uninitialized value in printf at ./Run line 1589.
Use of uninitialized value in printf at ./Run line 1590.
Use of uninitialized value in printf at ./Run line 1589.
Use of uninitialized value in printf at ./Run line 1590.

1 x Dhrystone 2 using register variables  1 2 3 4 5 6 7 8 9 10

1 x Double-Precision Whetstone  1 2 3 4 5 6 7 8 9 10

1 x Execl Throughput  1 2 3

1 x File Copy 1024 bufsize 2000 maxblocks  1 2 3

1 x File Copy 256 bufsize 500 maxblocks  1 2 3

1 x File Copy 4096 bufsize 8000 maxblocks  1 2 3

1 x Pipe Throughput  1 2 3 4 5 6 7 8 9 10

1 x Pipe-based Context Switching  1 2 3 4 5 6 7 8 9 10

1 x Process Creation  1 2 3

1 x System Call Overhead  1 2 3 4 5 6 7 8 9 10

1 x Shell Scripts (1 concurrent)  1 2 3

1 x Shell Scripts (8 concurrent)  1 2 3

4 x Dhrystone 2 using register variables  1 2 3 4 5 6 7 8 9 10

4 x Double-Precision Whetstone  1 2 3 4 5 6 7 8 9 10

4 x Execl Throughput  1 2 3

4 x File Copy 1024 bufsize 2000 maxblocks  1 2 3

4 x File Copy 256 bufsize 500 maxblocks  1 2 3

4 x File Copy 4096 bufsize 8000 maxblocks  1 2 3

4 x Pipe Throughput  1 2 3 4 5 6 7 8 9 10

4 x Pipe-based Context Switching  1 2 3 4 5 6 7 8 9 10

4 x Process Creation  1 2 3

4 x System Call Overhead  1 2 3 4 5 6 7 8 9 10

4 x Shell Scripts (1 concurrent)  1 2 3

4 x Shell Scripts (8 concurrent)  1 2 3

========================================================================
   BYTE UNIX Benchmarks (Version 5.1.3)

   System: morinezumiiii-pi: GNU/Linux
   OS: GNU/Linux -- 4.14.42-1-ARCH -- #1 SMP Tue May 22 00:34:20 UTC 2018
   Machine: armv7l (unknown)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: ARMv7 Processor rev 4 (v7l) (0.0 bogomips)

   CPU 1: ARMv7 Processor rev 4 (v7l) (0.0 bogomips)

   CPU 2: ARMv7 Processor rev 4 (v7l) (0.0 bogomips)

   CPU 3: ARMv7 Processor rev 4 (v7l) (0.0 bogomips)

   11:26:45 up  1:31,  1 user,  load average: 0.03, 0.07, 0.84; runlevel unknown

------------------------------------------------------------------------
Benchmark Run: Sat Jun 16 2018 11:26:45 - 11:54:59
4 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables        5710658.3 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     1148.6 MWIPS (9.9 s, 7 samples)
Execl Throughput                               1292.4 lps   (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        175847.5 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           59591.4 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        383032.1 KBps  (30.0 s, 2 samples)
Pipe Throughput                              409518.3 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  74782.5 lps   (10.0 s, 7 samples)
Process Creation                               2210.0 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                    882.9 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    329.4 lpm   (60.1 s, 2 samples)
System Call Overhead                         910125.1 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0    5710658.3    489.3
Double-Precision Whetstone                       55.0       1148.6    208.8
Execl Throughput                                 43.0       1292.4    300.6
File Copy 1024 bufsize 2000 maxblocks          3960.0     175847.5    444.1
File Copy 256 bufsize 500 maxblocks            1655.0      59591.4    360.1
File Copy 4096 bufsize 8000 maxblocks          5800.0     383032.1    660.4
Pipe Throughput                               12440.0     409518.3    329.2
Pipe-based Context Switching                   4000.0      74782.5    187.0
Process Creation                                126.0       2210.0    175.4
Shell Scripts (1 concurrent)                     42.4        882.9    208.2
Shell Scripts (8 concurrent)                      6.0        329.4    548.9
System Call Overhead                          15000.0     910125.1    606.8
                                                                   ========
System Benchmarks Index Score                                         340.5

------------------------------------------------------------------------
Benchmark Run: Sat Jun 16 2018 11:54:59 - 12:23:13
4 CPUs in system; running 4 parallel copies of tests

Dhrystone 2 using register variables       21401597.3 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     4504.5 MWIPS (9.9 s, 7 samples)
Execl Throughput                               2655.2 lps   (29.6 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        288190.0 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           89004.4 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        604423.9 KBps  (30.0 s, 2 samples)
Pipe Throughput                             1493791.7 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 251058.2 lps   (10.0 s, 7 samples)
Process Creation                               6257.7 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   2266.6 lpm   (60.1 s, 2 samples)
Shell Scripts (8 concurrent)                    330.4 lpm   (60.6 s, 2 samples)
System Call Overhead                        3276786.7 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   21401597.3   1833.9
Double-Precision Whetstone                       55.0       4504.5    819.0
Execl Throughput                                 43.0       2655.2    617.5
File Copy 1024 bufsize 2000 maxblocks          3960.0     288190.0    727.8
File Copy 256 bufsize 500 maxblocks            1655.0      89004.4    537.8
File Copy 4096 bufsize 8000 maxblocks          5800.0     604423.9   1042.1
Pipe Throughput                               12440.0    1493791.7   1200.8
Pipe-based Context Switching                   4000.0     251058.2    627.6
Process Creation                                126.0       6257.7    496.6
Shell Scripts (1 concurrent)                     42.4       2266.6    534.6
Shell Scripts (8 concurrent)                      6.0        330.4    550.7
System Call Overhead                          15000.0    3276786.7   2184.5
                                                                   ========
System Benchmarks Index Score                                         819.0

所感

ダメ。

ラズパイ3で熱暴走の問題があったけど、3B+ではCPU温度70℃超えると、サーマルスロットリングが発生して600MHz~1.2GHzまでクロックダウンする。

config.txtで定格1.4GHzにしたら何度もハングアップした。

そもそも3からクロックアップしてあんなちっこいゴミみたいなヒートスプレッダでどうにかなるという考え方がどうかしてる。

そして、よくケースとかに付属してるヒートシンク、あれ1個ではサーマルスロットリングを回避できなかった。

https://twitter.com/morinezumiiii/status/1007780629713309696

今回、定格1.4Gz保ったままUnixBench完走出来たのは、別途ヒートシンクを何個か重ねて追加してサーマルスロットリングを回避出来たからであって、標準の状態では、よっぽど寒い場所でない限り、定格1.4GHz稼働は絶対に無理。

https://twitter.com/morinezumiiii/status/1007801129067933697

3B+の1.4GHzという表記は、IntelでいうところのTurboBoost時のクロック数の表記に近いものだと思ったほうが良い。

よって、個人的な所感としては、3B+は無印3をマシにした程度である、と思う。

となると無印3はゴミですね、今更買う価値無しです。

今回はちょっとガッカリだったので辛口となりました。

おわり。