Author

Topic: [ANN] dstm's ZCash / Equihash Nvidia Miner v0.6.2 (Linux / Windows) - page 103. (Read 224961 times)

full member
Activity: 434
Merit: 107
Average results seen in miner (after running for 45 min.):
1060 1st card
v0.5.4 = 299.9
v0.5.6= 310.5
% increase 3.53

1060 2nd card
v0.5.4= 298.6
v0.5.6= 306.6*
% increase 2.67

*I had to increase TDP from 70 > 75% due to card failure rate 0.97. With updated tdp failure is 0.98 and hopefully improving to 0.99 and above.

1070
v0.5.4= 492.4
v0.5.6= 499.2
% increase 1.63


Letting the rig run for now to see how stable it is.
I will then see if the same % increase is seen on the pool side of things.
member
Activity: 336
Merit: 15
0.5.6 works fine with GTX 1070 (regular and Ti) - can't see much of a speed increase but also the power consumption is exactly same as with the old version. Gonna try it out on a GTX 1060 later.
full member
Activity: 434
Merit: 107
love your work dstm.
Thank you very much!


So when you say that it loads the gpus slightly more is it still different than how ewbf miner loads its gpus.
I wonder if this will make equihash more profitable to mine that ethash on the 1060s.

full member
Activity: 210
Merit: 100
Thanks dstm..
I'm a happy miner  Cheesy
newbie
Activity: 176
Merit: 0
I don't see boost for 1080ti...may be around 0.2% as Dev said.
full member
Activity: 158
Merit: 100
mine safe o/
New Version 0.5.6
reduce cpu load
minor performance improvements


This release contains many internal optimizations. CPU load is lower by about 15%-20% now. The solver contains some minor performance improvements - it's about 2%-2.5% on an 1060 and about 0.2% on faster GPUs. People on overclocked systems may have to readjust their setting since this version 'pushes' the GPUs slightly more.

That is awesome!
Will check 1070 and 1080ti work with this update asap

UPD:
ID   DEVICE NAME   °C   ∅ Sol/s   ∅ Sol/W   ∅ Watt   Shares   Lat
0   GeForce GTX 1070   64   501.41   2.81   178.59   119 / 0   109
1   GeForce GTX 1070   59   495.71   2.94   168.61   130 / 1   90
2   GeForce GTX 1070   56   496.00   2.92   169.70   154 / 3   104
3   GeForce GTX 1070   61   501.20   2.88   173.77   167 / 0   99
4   GeForce GTX 1070   56   495.25   2.93   169.12   149 / 0   94
5   GeForce GTX 1070   58   496.52   2.98   166.38   89 / 0   94
Total   -                           2986.08   2.91   1026.16   808 / 4   98

ID   DEVICE NAME   °C   ∅ Sol/s   ∅ Sol/W   ∅ Watt   Shares   Lat
0   GeForce GTX 1080 Ti   65   725.23   2.92   248.50   121 / 1   93
1   GeForce GTX 1080 Ti   63   750.32   3.19   235.49   132 / 0   86
2   GeForce GTX 1080 Ti   61   743.02   3.14   236.47   101 / 0   93
3   GeForce GTX 1080 Ti   61   758.17   3.06   248.00   129 / 0   94
4   GeForce GTX 1080 Ti   59   758.35   3.05   248.53   94 / 0   93
5   GeForce GTX 1080 Ti   62   768.11   2.94   260.85   131 / 0   94
6   GeForce GTX 1080 Ti   62   775.15   2.97   260.73   115 / 0   83
7   GeForce GTX 1080 Ti   60   753.62   3.20   235.45   96 / 0   93
Total   -                             6031.95   3.06   1974.01   919 / 1   91

I see no change for 1070 GPUs and 0.4% boost for 1080ti.
sr. member
Activity: 399
Merit: 254
New Version 0.5.6
reduce cpu load
minor performance improvements


This release contains many internal optimizations. CPU load is lower by about 15%-20% now. The solver contains some minor performance improvements - it's about 2%-2.5% on an 1060 and about 0.2% on faster GPUs. People on overclocked systems may have to readjust their setting since this version 'pushes' the GPUs slightly more.

Nice
Thank you
newbie
Activity: 176
Merit: 0
New Version 0.5.6
reduce cpu load
minor performance improvements


This release contains many internal optimizations. CPU load is lower by about 15%-20% now. The solver contains some minor performance improvements - it's about 2%-2.5% on an 1060 and about 0.2% on faster GPUs. People on overclocked systems may have to readjust their setting since this version 'pushes' the GPUs slightly more.
Thank you very much.
full member
Activity: 350
Merit: 126
Hi, what settings for failover servers? Thanks.

ZM doesn't support failover servers currently, I'll add it in one of the future releases.
newbie
Activity: 43
Merit: 0
Hi, what settings for failover servers? Thanks.
member
Activity: 120
Merit: 10
New Version 0.5.6
reduce cpu load
minor performance improvements


This release contains many internal optimizations. CPU load is lower by about 15%-20% now. The solver contains some minor performance improvements - it's about 2%-2.5% on an 1060 and about 0.2% on faster GPUs. People on overclocked systems may have to readjust their setting since this version 'pushes' the GPUs slightly more.

My asus rog stix 1080ti when run DSMT is about 700sol/s core +200 mem+150 TDP 80%.
When you can get my card is performance

thanks you
full member
Activity: 350
Merit: 126
New Version 0.5.6
reduce cpu load
minor performance improvements


This release contains many internal optimizations. CPU load is lower by about 15%-20% now. The solver contains some minor performance improvements - it's about 2%-2.5% on an 1060 and about 0.2% on faster GPUs. People on overclocked systems may have to readjust their setting since this version 'pushes' the GPUs slightly more.
member
Activity: 102
Merit: 11
I give my 1080ti's a little bit more power.  All 4 are EVGA SC Black

85% Power Limit
+125 Core
+145 Memory

For me...... Your latency numbers are Amazing !!!! ........... (4ms) Seriously........ You will be qualified for the hold attack if you had a big farm of miners....
Are you mining on your pool servers......?
member
Activity: 126
Merit: 11
I give my 1080ti's a little bit more power.  All 4 are EVGA SC Black



85% Power Limit
+125 Core
+145 Memory
full member
Activity: 558
Merit: 194
Could someone please write clock settings for 1080ti for getting over 750sols? I have asus strix gaming 1080ti.

750 Sols on air is pushing it I think.  I primarily run EVGA SC's and SC2's and they do around 710 Sol/s at modest OC settings (190W PL, +120 Core and +600 Memory).  Here's one of them:



If you're water cooled, you can push things further.  Here's a rig with 4x MSI Sea Hawks EK's and 1x Zotac ArcticStorm.  The Sea Hawks are running 190W PL, +155 Core and +700 memory.  The Zotac is running 230W PL, +230 Core and +700 memory.

member
Activity: 83
Merit: 10
@DSTM'S  could you change color when displaying the miners web page? maybe put a dark gray, or another color, because all that white shoots Undecided Undecided
jr. member
Activity: 189
Merit: 8
anyone have stable overclocking for Palit 1080 SuperJets?
newbie
Activity: 4
Merit: 0
any ideas how to fix "cudaMemcpy 1 failed"  Huh Will it help to replace windows 10 with windows 7?
reduce oc.
jr. member
Activity: 189
Merit: 8
any ideas how to fix "cudaMemcpy 1 failed"  Huh Will it help to replace windows 10 with windows 7?
newbie
Activity: 2
Merit: 0
I have been running this miner in ethos for 24 hours.  It reports more sols than ewbf, but it seems to drop from the zen suprnova pool regularly.  I can see miners drop and reappear just staying logged in and watching.  Anyone else see this issue?  I plan to leave this running for more days, but curious if others are seeing this or if this may be normal.  ewbf didn't have this same behaviour on suprnova.
I got problem with ZEN same as you.

Dev said what it's normal, said server sent packet which information to close connection, but could ZM ignore this packet? This packet got all card in the same time?  (8 my cards reconnects)


It seems like there was a misunderstanding. Periodical disconnects which happen often aren't 'normal'. The server or something in between e.g. your router closes the connection. Does this also happen on other servers for you?

It doesn't happen with ewbf.  I run 5 miners so its easy to see when one drops.  I don't know if the view in suprnova is misleading or not.  I will know for sure after a week when i can compare numbers.
Jump to: