Šestijádro s HT a škálování při kompilaci a OpenSSL

2.3.2012 23:15 | Přečteno: 2019× | Linux | | poslední úprava: 5.3.2012 19:27

Po roce se mi do ruky zase dostal zajímavý procesor, který myslím stojí za otestování. Oproti minulému měření se jedná o nárůst o dvě jádra a dva paměťové kanály a to by mohlo přinést nějaké zajímavé výsledky.

Testovací sestava

Intel i7-3930K (frekvence zafixována na 4,5 GHz aby Turbo při měření nemíchalo s výsledky), 32GB DDR3 1600MHz, X79 chipset)

Software

Aktuální stabilní Arch, gcc 4.6.2, jádro 3.2.8, x86_64, openSSL 1.0.0 d bn(64,64)

Metodika

Překládám vanilkové 3.2.9 s implicitním "Linusovým" .config (lze získat pomocí make defconfig). HT a jádra vypínám v BIOSu. Celý proces kompilace probíhá v ramdisku. Měřím pomocí time make -jN resp. openssl speed rsa2048 -multi N. Všechna měření dělám dvakrát a výsledky průměruju.

Výsledky

Kompilace Real time

Kompilace Sys + user

Signs/s pro 2048bit rsa, 1.0.0 d, bn(64,64)

Interpretace

Přínos HT je 19 % v případě kompilace a 12 % v případě RSA a jen málo záleží na tom, zda se jedná o dvou, čtyř či šestijádro.

RSA škáluje takřka ideálně - šestijádro je o 1,5x výkonnější než čtyřjádro a 3x výkonnější než dvoujádro a je jedno, zda poměřujeme HT či ne-HT varianty.

Škálování kompilace je o něco horší. Šestijádro je 1,43x výkonnější než čtyřjádro a 2,72x výkonnější nežli dvoujádro. Je opět jedno, zda poměřujeme HT či ne-HT varianty.

S nárůstem počtu jader (je jedno zda skutečných či "virtuálních") roste i režie systému. Jádro se s -j1 přeloží na dvoujádru bez HT o procento rychleji nežli na šestijádru s HT. V případe OpenSSL je rozdíl kolem půl procenta ale to už je pod hranicí chyby měření.

Očekávaný a i přes to pozoruhodný je graf "Sys+user". Je vidět, že množství spotřebovaného času procesoru velmi pomalu roste na procesorech bez HT. Podobně se chovají i procesory s HT až na rozsah počtu použitých threadů, kdy "zabírá" HT - v této části množství použitého procesorového času prudce roste.

Závěr

Tento kousek budu mít ještě nějakou chvíli doma, můžu tedy něco vyzkoušet, pokud o to bude zájem.

TrueCrypt

7zip

7-Zip [64] 9.20  Copyright (c) 1999-2010 Igor Pavlov  2010-11-18
p7zip Version 9.20 (locale=C,Utf16=off,HugeFiles=on,12 CPUs)

RAM size:   64391 MB,  # CPU hardware threads:  12
RAM usage:   2551 MB,  # Benchmark threads:     12

Dict        Compressing          |        Decompressing
      Speed Usage    R/U Rating  |    Speed Usage    R/U Rating
       KB/s     %   MIPS   MIPS  |     KB/s     %   MIPS   MIPS

22:   32306   957   3285  31428  |   427757  1149   3356  38575
23:   33450  1043   3268  34081  |   436059  1193   3344  39888
24:   32204  1076   3217  34625  |   420024  1175   3316  38963
25:   30450  1090   3188  34766  |   420977  1192   3321  39581
----------------------------------------------------------------
Avr:         1041   3240  33725              1177   3334  39252
Tot:         1109   3287  36488

John the Ripper

Benchmarking: Traditional DES [128/128 BS SSE2-16]... (12xOMP) DONE
Many salts:     37374K c/s real, 3143K c/s virtual
Only one salt:  27967K c/s real, 2356K c/s virtual

Benchmarking: BSDI DES (x725) [128/128 BS SSE2-16]... (12xOMP) DONE
Many salts:     1216K c/s real, 102828 c/s virtual
Only one salt:  1035K c/s real, 86200 c/s virtual

Benchmarking: FreeBSD MD5 [SSE2i 12x]... (12xOMP) DONE
Raw:    302976 c/s real, 25226 c/s virtual

Benchmarking: OpenBSD Blowfish (x32) [32/64 X2]... (12xOMP) DONE
Raw:    9481 c/s real, 794 c/s virtual

Benchmarking: Kerberos AFS DES [48/64 4K]... DONE
Short:  592384 c/s real, 592384 c/s virtual
Long:   1948K c/s real, 1948K c/s virtual

Benchmarking: LM DES [128/128 BS SSE2-16]... (12xOMP) DONE
Raw:    124305K c/s real, 10367K c/s virtual

Benchmarking: dynamic_0: md5($p)  (raw-md5)  [SSE2i 10x4x3]... DONE
Raw:    36744K c/s real, 36024K c/s virtual

Benchmarking: dynamic_1: md5($p.$s)  (joomla)  [SSE2i 10x4x3]... DONE
Many salts:     26896K c/s real, 26896K c/s virtual
Only one salt:  20065K c/s real, 20065K c/s virtual

Benchmarking: dynamic_2: md5(md5($p))  (e107)  [SSE2i 10x4x3]... DONE
Raw:    18854K c/s real, 18854K c/s virtual

Benchmarking: dynamic_3: md5(md5(md5($p)))  [SSE2i 10x4x3]... DONE
Raw:    12699K c/s real, 12699K c/s virtual

Benchmarking: dynamic_4: md5($s.$p)  (OSC)  [SSE2i 10x4x3]... DONE
Many salts:     27561K c/s real, 27561K c/s virtual
Only one salt:  20675K c/s real, 20675K c/s virtual

Benchmarking: dynamic_5: md5($s.$p.$s)  [SSE2i 10x4x3]... DONE
Many salts:     22904K c/s real, 22904K c/s virtual
Only one salt:  17820K c/s real, 17820K c/s virtual

Benchmarking: dynamic_6: md5(md5($p).$s)  [SSE2i 10x4x3]... DONE
Many salts:     38272K c/s real, 38272K c/s virtual
Only one salt:  13746K c/s real, 13746K c/s virtual

Benchmarking: dynamic_8: md5(md5($s).$p)  [SSE2i 10x4x3]... DONE
Many salts:     28533K c/s real, 28533K c/s virtual
Only one salt:  20774K c/s real, 20774K c/s virtual

Benchmarking: dynamic_9: md5($s.md5($p))  [SSE2i 10x4x3]... DONE
Many salts:     27228K c/s real, 27228K c/s virtual
Only one salt:  11825K c/s real, 11825K c/s virtual

Benchmarking: dynamic_10: md5($s.md5($s.$p))  [SSE2i 10x4x3]... DONE
Many salts:     14006K c/s real, 14006K c/s virtual
Only one salt:  11848K c/s real, 11848K c/s virtual

Benchmarking: dynamic_11: md5($s.md5($p.$s))  [SSE2i 10x4x3]... DONE
Many salts:     14265K c/s real, 14265K c/s virtual
Only one salt:  12033K c/s real, 12033K c/s virtual

Benchmarking: dynamic_12: md5(md5($s).md5($p))  (IPB)  [SSE2i 10x4x3]... DONE
Many salts:     5585K c/s real, 5585K c/s virtual
Only one salt:  4176K c/s real, 4176K c/s virtual

Benchmarking: dynamic_13: md5(md5($p).md5($s))  [SSE2i 10x4x3]... DONE
Many salts:     5580K c/s real, 5580K c/s virtual
Only one salt:  4195K c/s real, 4195K c/s virtual

Benchmarking: dynamic_14: md5($s.md5($p).$s)  [SSE2i 10x4x3]... DONE
Many salts:     22457K c/s real, 22457K c/s virtual
Only one salt:  11883K c/s real, 11883K c/s virtual

Benchmarking: dynamic_15: md5($u.md5($p).$s)  [64x2 (MD5_Body)]... DONE
Many salts:     12003K c/s real, 12003K c/s virtual
Only one salt:  5918K c/s real, 5918K c/s virtual

Benchmarking: dynamic_16: md5(md5(md5($p).$s).$s2)  [64x2 (MD5_Body)]... DONE
Many salts:     6961K c/s real, 6961K c/s virtual
Only one salt:  4066K c/s real, 4066K c/s virtual

Benchmarking: dynamic_17: phpass ($P$ or $H$)  [SSE2i 2x4x3]... DONE
Raw:    25704 c/s real, 25704 c/s virtual

Benchmarking: dynamic_18: md5($s.Y.$p.0xF7.$s) (Post.Office MD5)  [64x2 (MD5_Body)]... DONE
Many salts:     5509K c/s real, 5509K c/s virtual
Only one salt:  5100K c/s real, 5100K c/s virtual

Benchmarking: dynamic_19: Cisco PIX (MD5) [SSE2i 10x4x3]... DONE
Raw:    24277K c/s real, 24277K c/s virtual

Benchmarking: dynamic_20: Cisco PIX (MD5 salted) [SSE2i 10x4x3]... DONE
Many salts:     27964K c/s real, 27964K c/s virtual
Only one salt:  20584K c/s real, 20584K c/s virtual

Benchmarking: dynamic_21: HTTP Digest Access Auth [SSE2i 10x4x3]... DONE
Many salts:     3033K c/s real, 3033K c/s virtual
Only one salt:  2954K c/s real, 2954K c/s virtual

Benchmarking: dynamic_22: md5(sha1($p)) [SSE2i 10x4x3]... DONE
Raw:    9016K c/s real, 9016K c/s virtual

Benchmarking: dynamic_23: sha1(md5($p)) [SSE2i 10x4x3]... DONE
Raw:    8565K c/s real, 8565K c/s virtual

Benchmarking: dynamic_24: sha1($p.$s) [SSE2i 10x4x3]... DONE
Many salts:     14013K c/s real, 14013K c/s virtual
Only one salt:  11548K c/s real, 11548K c/s virtual

Benchmarking: dynamic_25: sha1($s.$p) [SSE2i 10x4x3]... DONE
Many salts:     13496K c/s real, 13496K c/s virtual
Only one salt:  11682K c/s real, 11682K c/s virtual

Benchmarking: dynamic_26: sha1($p) raw-sha1 [4x2]... DONE
Raw:    17893K c/s real, 17893K c/s virtual

Benchmarking: dynamic_27: FreeBSD MD5 [SSE2i 4x3]... DONE
Raw:    43476 c/s real, 43476 c/s virtual

Benchmarking: dynamic_28: Apache MD5 [SSE2i 4x3]... DONE
Raw:    42864 c/s real, 42864 c/s virtual

Benchmarking: dynamic_29: md5(unicode($p)) [SSE2i 10x4x3]... DONE
Raw:    24366K c/s real, 24366K c/s virtual

Benchmarking: dynamic_1001 md5(md5(md5(md5($p)))) [SSE2i 10x4x3]... DONE
Raw:    9475K c/s real, 9475K c/s virtual

Benchmarking: dynamic_1002 md5(md5(md5(md5(md5($p))))) [SSE2i 10x4x3]... DONE
Raw:    7617K c/s real, 7617K c/s virtual

Benchmarking: dynamic_1003 md5(md5($p).md5($p)) [64x2 (MD5_Body)]... DONE
Raw:    3643K c/s real, 3643K c/s virtual

Benchmarking: dynamic_1004 md5(md5(md5(md5(md5(md5($p)))))) [SSE2i 10x4x3]... DONE
Raw:    6366K c/s real, 6366K c/s virtual

Benchmarking: dynamic_1005 md5(md5(md5(md5(md5(md5(md5($p))))))) [SSE2i 10x4x3]... DONE
Raw:    5471K c/s real, 5471K c/s virtual

Benchmarking: dynamic_1006 md5(md5(md5(md5(md5(md5(md5(md5($p)))))))) [SSE2i 10x4x3]... DONE
Raw:    4794K c/s real, 4794K c/s virtual

Benchmarking: dynamic_1008 md5($p.$s) [RADIUS User-Password] [SSE2i 10x4x3]... DONE
Many salts:     26310K c/s real, 26310K c/s virtual
Only one salt:  19426K c/s real, 19426K c/s virtual

Benchmarking: dynamic_1009 md5($s.$p) [RADIUS Responses] [SSE2i 10x4x3]... DONE
Many salts:     31491K c/s real, 31491K c/s virtual
Only one salt:  22651K c/s real, 22651K c/s virtual

Benchmarking: Eggdrop [blowfish]... DONE
Raw:    35482 c/s real, 35482 c/s virtual

Benchmarking: DIGEST-MD5 [DIGEST-MD5 authentication]... DONE
Many salts:     2475K c/s real, 2475K c/s virtual
Only one salt:  2476K c/s real, 2476K c/s virtual

Benchmarking: More Secure Internet Password [RSA MD defined by BSAFE 1.x - Lotus v6]... DONE
Many salts:     123324 c/s real, 123324 c/s virtual
Only one salt:  73382 c/s real, 73382 c/s virtual

Benchmarking: EPiServer SID Hashes [SHA-1]... DONE
Many salts:     6791K c/s real, 6791K c/s virtual
Only one salt:  6362K c/s real, 6362K c/s virtual

Benchmarking: HTTP Digest access authentication [HDAA-MD5]... DONE
Many salts:     2585K c/s real, 2585K c/s virtual
Only one salt:  2463K c/s real, 2488K c/s virtual

Benchmarking: IPB2 MD5 [Invision Power Board 2.x salted MD5]... DONE
Many salts:     5250K c/s real, 5250K c/s virtual
Only one salt:  3223K c/s real, 3223K c/s virtual

Benchmarking: Kerberos v4 TGT [krb4 DES]... DONE
Raw:    4508K c/s real, 4508K c/s virtual

Benchmarking: Kerberos v5 TGT [krb5 3DES (des3-cbc-sha1)]... DONE
Raw:    79965 c/s real, 79965 c/s virtual

Benchmarking: MSCHAPv2 C/R MD4 DES [mschapv2]... (12xOMP) DONE
Many salts:     27131K c/s real, 2266K c/s virtual
Only one salt:  16814K c/s real, 1422K c/s virtual

Benchmarking: MYSQL_fast [mysql-fast]... DONE
Raw:    54579K c/s real, 54038K c/s virtual

Benchmarking: MYSQL [mysql]... DONE
Raw:    6684K c/s real, 6684K c/s virtual

Benchmarking: LM C/R DES [netlm]... (12xOMP) DONE
Many salts:     24772K c/s real, 2196K c/s virtual
Only one salt:  2099K c/s real, 965485 c/s virtual

Benchmarking: LMv2 C/R MD4 HMAC-MD5 [netlmv2]... (12xOMP) DONE
Many salts:     11225K c/s real, 933867 c/s virtual
Only one salt:  7999K c/s real, 672225 c/s virtual

Benchmarking: NTLMv1 C/R MD4 DES [ESS MD5] [netntlm]... (12xOMP) DONE
Many salts:     28053K c/s real, 2339K c/s virtual
Only one salt:  17141K c/s real, 1429K c/s virtual

Benchmarking: NTLMv2 C/R MD4 HMAC-MD5 [netntlmv2]... (12xOMP) DONE
Many salts:     9953K c/s real, 828059 c/s virtual
Only one salt:  7225K c/s real, 622874 c/s virtual

Benchmarking: HalfLM C/R DES [nethalflm]... (12xOMP) DONE
Many salts:     29163K c/s real, 2456K c/s virtual
Only one salt:  3919K c/s real, 1273K c/s virtual

Benchmarking: Netscreen MD5 [NS MD5]... DONE
Raw:    8258K c/s real, 8176K c/s virtual

Benchmarking: NT MD4 [128/128 X2 SSE2-16]... DONE
Raw:    59242K c/s real, 59242K c/s virtual

Benchmarking: PHPS -- md5(md5($pass).$salt) [SSE2i 10x4x3]... DONE
Many salts:     40279K c/s real, 40279K c/s virtual
Only one salt:  13476K c/s real, 13476K c/s virtual

Benchmarking: Post.Office MD5 [STD]... DONE
Many salts:     5548K c/s real, 5548K c/s virtual
Only one salt:  5151K c/s real, 5151K c/s virtual

Benchmarking: Mac OS X 10.4 - 10.6 salted SHA-1 [32/64]... (12xOMP) DONE
Many salts:     34329K c/s real, 2909K c/s virtual
Only one salt:  24093K c/s real, 2021K c/s virtual

Benchmarking: CRC-32 [32/64]... (4xOMP) DONE
Many salts:     568066K c/s real, 142372K c/s virtual
Only one salt:  132907K c/s real, 33226K c/s virtual

Benchmarking: HMAC MD5 [SSE2i 12x]... DONE
Raw:    12826K c/s real, 12699K c/s virtual

Benchmarking: Lotus5 [Lotus v5 Proprietary]... (12xOMP) DONE
Raw:    5550K c/s real, 462913 c/s virtual

Benchmarking: Generic salted MD4 [32/64]... DONE
Many salts:     12875K c/s real, 12748K c/s virtual
Only one salt:  12148K c/s real, 12148K c/s virtual

Benchmarking: MediaWiki -- md5($s.'-'.md5($p)) [SSE2i 10x4x3]... DONE
Many salts:     30049K c/s real, 30049K c/s virtual
Only one salt:  12109K c/s real, 12109K c/s virtual

Benchmarking: M$ Cache Hash [Generic 1x]... (12xOMP) DONE
Many salts:     169684K c/s real, 14152K c/s virtual
Only one salt:  46006K c/s real, 3837K c/s virtual

Benchmarking: M$ Cache Hash 2 (DCC2) [SSE2i 8x]... (12xOMP) DONE
Raw:    6368 c/s real, 536 c/s virtual

Benchmarking: MS Kerberos 5 AS-REQ Pre-Auth [mskrb5]... (12xOMP) DONE
Many salts:     11943K c/s real, 996158 c/s virtual
Only one salt:  6119K c/s real, 510377 c/s virtual

Benchmarking: MS-SQL [SSE2i 8x]... DONE
Many salts:     21459K c/s real, 21038K c/s virtual
Only one salt:  14424K c/s real, 14424K c/s virtual

Benchmarking: MS-SQL05 [SSE2i 8x]... DONE
Many salts:     21631K c/s real, 21631K c/s virtual
Only one salt:  19240K c/s real, 19240K c/s virtual

Benchmarking: MySQL 4.1 double-SHA-1 [SSE2i 8x]... DONE
Raw:    9679K c/s real, 9679K c/s virtual

Benchmarking: Netscape LDAP SHA [SSE2i 8x]... DONE
Raw:    20490K c/s real, 20490K c/s virtual

Benchmarking: NT v2 [SSE2i 12x]... DONE
Raw:    54429K c/s real, 54979K c/s virtual

Benchmarking: Oracle 11g [oracle11]... DONE
Many salts:     6604K c/s real, 6604K c/s virtual
Only one salt:  5631K c/s real, 5631K c/s virtual

Benchmarking: Oracle [oracle]... DONE
Raw:    1188K c/s real, 1188K c/s virtual

Benchmarking: PHPass MD5 [SSE2i 2x4x3]... DONE
Raw:    25704 c/s real, 25704 c/s virtual

Benchmarking: PIX MD5 [SSE2i 10x4x3]... DONE
Many salts:     25123K c/s real, 25123K c/s virtual
Only one salt:  25119K c/s real, 25119K c/s virtual

Benchmarking: pkzip [N/A]... (12xOMP) DONE
Many salts:     122717K c/s real, 10234K c/s virtual
Only one salt:  42811K c/s real, 3570K c/s virtual

Benchmarking: Raw MD4 [SSE2i 12x]... DONE
Raw:    56709K c/s real, 56147K c/s virtual

Benchmarking: Raw MD5 [SSE2i 12x]... DONE
Raw:    40151K c/s real, 40151K c/s virtual

Benchmarking: Raw MD5 [SSE2i 10x4x3]... DONE
Raw:    37516K c/s real, 37140K c/s virtual

Benchmarking: Raw SHA-1 [SSE2i 8x]... DONE
Raw:    17728K c/s real, 17728K c/s virtual

Benchmarking: Raw SHA [32/64]... DONE
Raw:    5526K c/s real, 5526K c/s virtual

Benchmarking: md5(unicode($p)) [SSE2i 10x4x3]... DONE
Raw:    25345K c/s real, 25345K c/s virtual

Benchmarking: Salted SHA-1 [SSE2i 8x]... DONE
Many salts:     19451K c/s real, 19451K c/s virtual
Only one salt:  17186K c/s real, 17186K c/s virtual

Benchmarking: SAP BCODE [sapb]... (12xOMP) DONE
Many salts:     24714K c/s real, 2059K c/s virtual
Only one salt:  19315K c/s real, 1614K c/s virtual

Benchmarking: SAP CODVN G (PASSCODE) [sapg]... (12xOMP) DONE
Many salts:     11928K c/s real, 994909 c/s virtual
Only one salt:  10819K c/s real, 903105 c/s virtual

Benchmarking: Generic salted SHA-1 [32/64]... DONE
Many salts:     6436K c/s real, 6310K c/s virtual
Only one salt:  6225K c/s real, 6225K c/s virtual

Benchmarking: Raw SHA-224 [32/64]... DONE
Raw:    3156K c/s real, 3156K c/s virtual

Benchmarking: Raw SHA-256 [32/64]... DONE
Raw:    3203K c/s real, 3203K c/s virtual

Benchmarking: Raw SHA-384 [64/64]... DONE
Raw:    2656K c/s real, 2656K c/s virtual

Benchmarking: Raw SHA-512 [64/64]... DONE
Raw:    2638K c/s real, 2638K c/s virtual

Benchmarking: Mac OS X 10.7+ salted SHA-512 [64/64]... (12xOMP) DONE
Many salts:     14793K c/s real, 1234K c/s virtual
Only one salt:  12627K c/s real, 1052K c/s virtual

Benchmarking: hmailserver [32/64]... DONE
Many salts:     3310K c/s real, 3277K c/s virtual
Only one salt:  3096K c/s real, 3096K c/s virtual

Benchmarking: sybasease [32/64]... (12xOMP) DONE
Many salts:     2447K c/s real, 204162 c/s virtual
Only one salt:  2278K c/s real, 190213 c/s virtual

Benchmarking: generic crypt(3) DES [?/64]... (12xOMP) DONE
Many salts:     3207K c/s real, 267710 c/s virtual
Only one salt:  3233K c/s real, 269945 c/s virtual

Benchmarking: Tripcode DES [48/64 4K]... DONE
Raw:    487808 c/s real, 487808 c/s virtual

Benchmarking: ssh [32/64]... (12xOMP) DONE
Raw:    1244K c/s real, 103766 c/s virtual

Benchmarking: pdf [32/64]... DONE
Many salts:     28192 c/s real, 27912 c/s virtual
Only one salt:  57886 c/s real, 57886 c/s virtual

Benchmarking: rar [32/64]... DONE
Raw:    56.0 c/s real, 56.0 c/s virtual

Benchmarking: zip [32/64]... (12xOMP) DONE
Raw:    5376 c/s real, 447 c/s virtual

Benchmarking: dummy [N/A]... DONE
Raw:    84986K c/s real, 84145K c/s virtual

fcrackzip

 cpmask: (skipped)
 zip1: cracks/s = 12298943
*zip2, USE_MULT_TAB: cracks/s = 12720251

Geekbench

http://browse.geekbench.ca/geekbench2/570586        

Obrázky

Komentáře

[A]

dam svoje skromne tipy na testovanie ktore by ma zaujimali. dopredu vdaka.

geekbench /odkaz na vysledok/
7zip benchmark (7z b)
fcrackzip (fcrackzip -B) 
john the ripper (john --test) 
truecrypt benchmark screenshot

[miho]

Vsechno doplneno.

[Michal Fecko]

Mozno o mesiac az dva budem mat moznost skusat i7 3960X Extreme Edition, 3.3GHz, 15MB a uz teraz sa tesim ako male decko

[xxxs]

pekne.

mala chybka: Šestijádro je 1,43x výkonnější než čtyřjádro a 2,72x výkonnější nežli čtyřjádro.

[__dark__]

Ja mam workstation s 2x6 cores (24 threads) s 24GB RAM, mozna bych neco mohl spustit, pokud to je v macports repozitari.

[miho]

Na otestovani kompilace bys musel nabootovat linux z live CD/USB.

Openssl bys vyzkouset mohl, v MacOS je implicitne nainstalovany; v konzoli spust openssl speed rsa2048 -multi X (za X dosad postupne 1,6,12,18,24) ale zrovna u toho testu moc prekvapeni cekat z hlediska skalovani nelze. Absolutni cislo nema smysl srovnavat protoze to je zkresleno rozdilem ve vykonu openssl na MacOS a v Linuxu (nevim jak ted, ale v Snow Leopardovi byl OpenSSL 2x pomalejsi nez v linuxu... soudruzi z Cupretina tam neco zvorali)

Geekbench je ke stazeni i pro MacOS ale jelikoz tam je malo variaci hardware, tak to ani nemusis tahat protoze vysledky jsou vicemene zname- pokud mas dvanactijadro s frekvenci 2.66, tak budes mit vysledek o cca 5 % horsi nez to moje sestijadro (novejsi architektura, podstatne vyssi frekvence), pokud 2.93, tak budes mit vysledek o cca 1 % lepsi.

[__dark__]

Tech variaci MacPro prave moc neni (mam 2.93 GHz), tak jsem se benchmarkovanim nikdy nezabyval, teda krome vlastnich veci :) Zitra muzu zkusit ten openssl, pokud bude trochu casu:)