mirror of
https://github.com/JayDDee/cpuminer-opt.git
synced 2025-09-17 23:44:27 +00:00
Initial upload v3.4.7
This commit is contained in:
88
RELEASE_ANNOUNCEMENT
Normal file
88
RELEASE_ANNOUNCEMENT
Normal file
@@ -0,0 +1,88 @@
|
||||
Release 3.4.7 of cpuminer-opt is available for download with support for
|
||||
Windows.
|
||||
|
||||
cpuminer-opt now supports 38 algorithms on CPUs with at least SSE2
|
||||
capabilities including Intel Core2, Nehalem and AMD equivalent. See the
|
||||
performance chart below for details.
|
||||
|
||||
In addition 19 algorithms have optimizations to take advantage of
|
||||
CPUs with AES_NI for even greater performance, including the Intel
|
||||
Westbridge and newer and AMD equivalent. See the performance
|
||||
comparison below.
|
||||
|
||||
New in 3.4.7
|
||||
|
||||
- fixed benchmark, except for x11evo
|
||||
- added CPU temperature to share submission report
|
||||
|
||||
Users with non-SSE2 CPUs or who want to mine algos not supported by
|
||||
cpuminer-opt may find cpuminer-multi by TPruvot useful.
|
||||
|
||||
Chart out of date, will be removed.
|
||||
|
||||
The performance chart below is for an Intel i7-6700K @ 4 GHz, 16 GB mem.
|
||||
|
||||
Normalization rates have been added to the chart to help with profit
|
||||
switching pools. Reference algo x11 = 1.
|
||||
|
||||
Due to the peculiarities of some algorithms their performance on other CPU
|
||||
architectures may not scale equally. Their normalizations rates will also
|
||||
differ from those listed below. YMMV.
|
||||
|
||||
Normalized profitability = algo profitability * norm rate
|
||||
|
||||
AES-AVX SSE2(1) norm rate(5)
|
||||
------- ------- ---------
|
||||
x11 780 K 525 K 1
|
||||
x13 392 298 0.50
|
||||
x14 370 271 0.48
|
||||
x15 341 270 0.45
|
||||
x17 317 248 0.43
|
||||
x11gost 562 392 0.72
|
||||
x11evo 590 387 0.78
|
||||
quark 1195 924 1.61
|
||||
qubit 1182 765 1.45
|
||||
nist5 2000 1592 3.37
|
||||
zr5 850 650 1.15
|
||||
c11 784 475 0.99
|
||||
myr-gr 1572 1560 2.12
|
||||
hmq1725 214 161 0.29
|
||||
m7m 121 77.4 0.155
|
||||
lyra2re 1380 900 1.76
|
||||
lyra2rev2 1350 980 1.73
|
||||
cryptonight 290 H 165 H 0.00039
|
||||
cryptolight 685 ? 0.00093
|
||||
hodl 600 200 0.00081
|
||||
lbry (4) 2620 3.53
|
||||
neoscrypt (4) 32 K 0.043
|
||||
argon2 (4) 33.7 0.045
|
||||
groestl (4) 931 1.26
|
||||
skein (4) 5747 7.77
|
||||
skein2 (4) 8675 11.7
|
||||
pentablake (4) 3960 5.35
|
||||
keccak (4) 7790 10.5
|
||||
scrypt (4) 113 0.153
|
||||
sha256d (4) 62.5 0.084
|
||||
veltor (4) 1017 1.30
|
||||
blake (4) 22.4 M 30.4
|
||||
blake2s (4) 19.0 25.7
|
||||
vanilla (4) 33.0 44.6
|
||||
blakecoin (4) 33.9 45.8
|
||||
decred (4) 22.6 30.5
|
||||
axiom (4) 72 H 0.000098
|
||||
yescrypt (4) 3760 0.0051
|
||||
scryptjane (4) 250 0.00034
|
||||
pluck(2) (4) 1925 0.0026
|
||||
drop(2) (4) 934 K 1.26
|
||||
fresh(2) (4) 528 0.71
|
||||
whirlpool(2) (4) 1290 1.74
|
||||
whirlpoolx(2) (4) 5110 6.9
|
||||
|
||||
Footnotes:
|
||||
(1) SSE2 rates are simulated in software (-march=core2) on an i7.
|
||||
(2) Benchmark tested only
|
||||
(3) CPU architecture not supported for algo. It won't work.
|
||||
(4) AES_NI Optimization not available for CPU artchitecture. Uses SSE2, slower.
|
||||
(5) Normalised profitability = algo profitability * norm rate, x11 = 1
|
||||
(6) Not supported on Windows
|
||||
|
Reference in New Issue
Block a user