hsrminer - Nvidia mining software for various algos by palgin&alexkap

[ferall]

Indeed.

I get 1350 or ccminer anyway, 1450 on this miner. 1070 Ti @ 80

how?

Im +155, +600, 80% and getting 1325kh/s

what other settings have you changed?  Intensity?

Core 200 mem 700 here. 80%

[1714687735]

1714687735

[1714687735]

1714687735

[1714687735]

1714687735

[1714687735]

1714687735

[badfad]

Guys please take into account that different brands of nvidia models have different TDPs and factory OC, so 80% tdp and +100 core will mean diffrent power and clock values depending on brand.
Zotac 1080s AMP! Extreme do close to 1500 kh/s at 75 % TDP and +0 core +625 mem. But the thing is they come with a 270W TDP and a lot of +core overclock.
MSI 1070tis do 1370 at 80% TDP (140W) and 200+ core 600 mem.
MSI 1080tis do 1800+ at 90% TDP(250W) and 100+ core 655 mem, probably can do more.

[giveen1]

Interesting point of thought, using NH miner on their server, it causes my mV to jump up to 1000mV, while hsrminer_neoscrypt sits it at 930mV.

[canhbtce]

 INFO : [12:21:20] : GPU#3 - Zotac GTX 1060 3GB, Temp is 65C, Fans at 80%, Power consumption 90W
 INFO : [12:21:20] : GPU#0 - Zotac GTX 1060 3GB, Temp is 65C, Fans at 80%, Power consumption 85W
 INFO : [12:21:48] : GPU#1 - Zotac GTX 1060 3GB, Temp is 69C, Fans at 80%, Power consumption 73W
 INFO : [12:21:48] : Share Accepted
 INFO : [12:21:57] : Share Accepted
 INFO : [12:22:04] : Share Accepted
 INFO : [12:22:30] : New block #37316 received, new difficulty is 812.118
 INFO : [12:23:30] : Share Accepted, overall speed is 4438.75kH/s, efficiency 99.45%
 INFO : [12:23:34] : Share Accepted
 INFO : [12:23:52] : New block #37317 received, new difficulty is 779.949
 INFO : [12:24:13] : Share Accepted
 INFO : [12:24:52] : Connection interrupted
 INFO : [12:24:58] : MESSAGE FROM SERVER: welcome on ALTMINER.NET - happy mining!
 INFO : [12:24:58] : Share difficulty reset -- 64 (0.00098)
 INFO : [12:25:02] : Share Accepted
 INFO : [12:25:02] : GPU#4 - Zotac GTX 1060 3GB, Temp is 68C, Fans at 80%, Power consumption 78W
 INFO : [12:25:02] : Share Accepted
 INFO : [12:25:03] : Share Accepted
 INFO : [12:25:04] : GPU#5 - Zotac GTX 1060 3GB, Temp is 69C, Fans at 80%, Power consumption 89W
 INFO : [12:25:04] : Share Accepted
 INFO : [12:25:06] : Share Accepted
 INFO : [12:25:06] : Connection interrupted
 INFO : [12:25:15] : Next connection attempt in 10 seconds
 INFO : [12:25:36] : stratum_subscribe timed out
 INFO : [12:25:36] : Next connection attempt in 10 seconds
 INFO : [12:25:50] : MESSAGE FROM SERVER: welcome on ALTMINER.NET - happy mining!
 INFO : [12:25:50] : New block #37318 received, new difficulty is 762.411
Cuda error in func 'neoscrypt_cpu_hash_1060' at line 1420 : an illegal memory access was encountered.
Cuda error in func 'neoscrypt_cpu_hash_1060' at line 1408 : an illegal memory access was encountered.
Cuda error in func 'neoscrypt_cpu_hash_1060' at line 1408 : an illegal memory access was encountered.

i have 6x1060 3g.
i can't fix this error. i got it many times.....

[GrassBite]

Indeed.

I get 1350 or ccminer anyway, 1450 on this miner. 1070 Ti @ 80

how?

Im +155, +600, 80% and getting 1325kh/s

what other settings have you changed?  Intensity?

Core 200 mem 700 here. 80%

How do you cool that?  I run 100, 150, 70% on my 1070tis and the one in the middle still hits 74 degrees

[stevascha]

As with Palgin's other miners, my pool results do not match my local reported speed. Since changing to this miner on my home mining rig the miner reported speed is around 500kh higher than my previous combination of tpruvot/1080ti and klausT/1070ti. However the reported speed at hashrefinery for that rig has now dropped to 300kh lower than before over the 12 hour test period.

Previous combined speed: 6.3kh
HSRMiner reported speed: 6.8kh
Pool average last 24hrs: 5.95kh

Since changing back an hour ago I am starting to see my pool graph go back to the previous 6.3kh. I don't know what you're doing with these miners you release Palgin, but I have yet to find one of your miners where the local reported speed is anywhere close to the actual performance.

Note: Tested Neoscrypt only, 388.something, win7 x64, coin switching pool.

Hope you can get to the bottom of this recurring issue, your miner software is very stable and I like the simplicity of the console output. I also like that I don't have to use a different miner for each card. The fee is reasonable, and will be well earned once the pool workrate matches the locally reported workrate.

I nearly didnt add this last part as I'm less than confident in my observations here, but it may help so here we go anyway. It's hard to tell but the hashrate dips may be related to large difficulty changes when switching between coin chains. I have not tested the miner directly mining a coin. This may give you a direction of investigation, please bear in mind my lack of confidence in this observation though! It would need testing with more granularity than is available on my pool graph.

totally agree, the palgin programs have some output issue to pool
ex in local shows 1400khs and in pool shows 1300khs

[ironandrey]

Powerfull miner Thanks!
My result
Asus Strix 1060 6gb. - 941khash tempr.(50c)
EVGA 1070 - 1335khash tempr.(52c)
EVGA 1080 - 1427khesh tempr.(46c)

[ferall]

Indeed.

I get 1350 or ccminer anyway, 1450 on this miner. 1070 Ti @ 80

how?

Im +155, +600, 80% and getting 1325kh/s

what other settings have you changed?  Intensity?

Core 200 mem 700 here. 80%

How do you cool that?  I run 100, 150, 70% on my 1070tis and the one in the middle still hits 74 degrees

My cards are on two levels and are spaced well apart. They run at 45c. They are also in a cold building which helps.

[fedxiii]

Powerfull miner Thanks!
My result
Asus Strix 1060 6gb. - 941khash tempr.(50c)
EVGA 1070 - 1335khash tempr.(52c)
EVGA 1080 - 1427khesh tempr.(46c)

bat file 1060 6gb and msi config .pls.

[dragonmike]

I was having the same problems with the 1080ti, I was getting 1850kh/s with one card but after some time it would drop down to 1550 and stay there. Sometimes it was a few hours and then drop and sometimes it was longer. It was suggested to me that my video card driver was crashing in the background and although it wasn't crashing the system, the overclock software was not able to reconnect to the card and so the card was running at base stats and not being overclocked even though the software showed it as overclocked.
...

Interesting premise that I am trying out now. I rebooted the machine again, lowered the o/c from +140/+350 to +125/+300 and restarted hsrminer. Reported speed is now around 1280; let's see if it holds.

Welp, that didn't work... Exactly 1 hour after restarting at a lower o/c the hashrate dropped to ~1150 (instead of ~1170).

I've just started poking around in event viewer but haven't found anything yet on driver crashes. Not sure where to look, actually, but I'll figure it out eventually... I'm sure.

EDIT - actually, backing off the o/c may, indeed, have worked. I decided to reboot the machine and try +125/+300 again and it has been running fairly steady at 1220 kH/s for 6 hours now. I never did find the gpu driver crash in the logs, but that isn't saying much.

If that didn't do the trick, you could try the opposite, i.e. reapplying o/c every few minutes.
Using nvidiainspector, I have a little script that runs in the background. It's got a 10min timer loop, and reapplies overclocks everytime it restarts. This way, if one card drops off for whatever reason, this both helps reviving it and makes sure overclocks are always current.

Happy to share it later tonight if you're interested.

[p337miner]

I was having the same problems with the 1080ti, I was getting 1850kh/s with one card but after some time it would drop down to 1550 and stay there. Sometimes it was a few hours and then drop and sometimes it was longer. It was suggested to me that my video card driver was crashing in the background and although it wasn't crashing the system, the overclock software was not able to reconnect to the card and so the card was running at base stats and not being overclocked even though the software showed it as overclocked.

I was running at maximum overclock and although it was stable with CCminer, it was not stable with HSRminer. I've resolved this issue by lowering the overclock just a bit to the point where the drops no longer occurred. I've also vastly increased the virtual memory and now I no longer have this drop in hashrate. I do not mess with -i or -c and so I can't confirm if any of that works, I can only tell you I was seeing the drops daily, and now I do not.

You can check your windows logs to see if the drivers are crashing, I've never checked the logs because once I made the changes it stopped happening. Hope this helps you guys..

May I ask what your cc/cm/tdp is for ccminer? Trying to find some good oc settings for 1080ti

[MagicSmoker]

Welp, that didn't work... Exactly 1 hour after restarting at a lower o/c the hashrate dropped to ~1150 (instead of ~1170).

I've just started poking around in event viewer but haven't found anything yet on driver crashes. Not sure where to look, actually, but I'll figure it out eventually... I'm sure.

EDIT - actually, backing off the o/c may, indeed, have worked. I decided to reboot the machine and try +125/+300 again and it has been running fairly steady at 1220 kH/s for 6 hours now. I never did find the gpu driver crash in the logs, but that isn't saying much.

If that didn't do the trick, you could try the opposite, i.e. reapplying o/c every few minutes.
Using nvidiainspector, I have a little script that runs in the background. It's got a 10min timer loop, and reapplies overclocks everytime it restarts. This way, if one card drops off for whatever reason, this both helps reviving it and makes sure overclocks are always current.

Happy to share it later tonight if you're interested.

I don't want to jinx myself here, but hashrate has been holding steady at 1220 for over 24 hours now, and the pool mostly agrees based on the coins earned vs. average difficulty for the past 24 hours.

So I don't think I'll need your script, but I am interested in seeing it because it sounds really useful and I don't have much experience with powershell scripting (I do have some remnant knowledge of optimizing config.sys and autoexec.bat, however... which probably gives away my age...  )

[Dealazer]

Hello guys,

I'm getting the below error on three different rigs. Any ideas?

Cuda error in func 'neoscrypt_cpu_init' at line 1352 : out of memory.

This miner is using code by ccminer by KlausT.

I believe it can go further than just the mere 700kH/s of 1060.

[cmbconcretefire]

I was having the same problems with the 1080ti, I was getting 1850kh/s with one card but after some time it would drop down to 1550 and stay there. Sometimes it was a few hours and then drop and sometimes it was longer. It was suggested to me that my video card driver was crashing in the background and although it wasn't crashing the system, the overclock software was not able to reconnect to the card and so the card was running at base stats and not being overclocked even though the software showed it as overclocked.
...

Interesting premise that I am trying out now. I rebooted the machine again, lowered the o/c from +140/+350 to +125/+300 and restarted hsrminer. Reported speed is now around 1280; let's see if it holds.

Welp, that didn't work... Exactly 1 hour after restarting at a lower o/c the hashrate dropped to ~1150 (instead of ~1170).

I've just started poking around in event viewer but haven't found anything yet on driver crashes. Not sure where to look, actually, but I'll figure it out eventually... I'm sure.

EDIT - actually, backing off the o/c may, indeed, have worked. I decided to reboot the machine and try +125/+300 again and it has been running fairly steady at 1220 kH/s for 6 hours now. I never did find the gpu driver crash in the logs, but that isn't saying much.

If that didn't do the trick, you could try the opposite, i.e. reapplying o/c every few minutes.
Using nvidiainspector, I have a little script that runs in the background. It's got a 10min timer loop, and reapplies overclocks everytime it restarts. This way, if one card drops off for whatever reason, this both helps reviving it and makes sure overclocks are always current.

Happy to share it later tonight if you're interested.

I am extremely interested in your script. I want it. I want to look at it. And then I want to be able to possibly PM you and ask you more questions about it. I've literally been looking for something like this for WEEKS!! Please share / explain install as soon as possible.

[jugger1028]

I was having the same problems with the 1080ti, I was getting 1850kh/s with one card but after some time it would drop down to 1550 and stay there. Sometimes it was a few hours and then drop and sometimes it was longer. It was suggested to me that my video card driver was crashing in the background and although it wasn't crashing the system, the overclock software was not able to reconnect to the card and so the card was running at base stats and not being overclocked even though the software showed it as overclocked.

I was running at maximum overclock and although it was stable with CCminer, it was not stable with HSRminer. I've resolved this issue by lowering the overclock just a bit to the point where the drops no longer occurred. I've also vastly increased the virtual memory and now I no longer have this drop in hashrate. I do not mess with -i or -c and so I can't confirm if any of that works, I can only tell you I was seeing the drops daily, and now I do not.

You can check your windows logs to see if the drivers are crashing, I've never checked the logs because once I made the changes it stopped happening. Hope this helps you guys..

May I ask what your cc/cm/tdp is for ccminer? Trying to find some good oc settings for 1080ti

I don't use CCMiner, but when I did I was able to do tdp 90% core +150 mem +650 with no problems in stability.. For HSRMiner I have had to lower that to 90 tdp +150core +625mem. Note that my evga cards are set to this, my Asus cards are not as they are not stable with the higher core clock so I'm at +100core +600mem on those. Also note HSRMiner is less stable at higher clocks than CCMiner, HSRMiner crashes more frequently if my OC settings are higher.

Also note I have not confirmed that the hashrates HSRMiner are showing match what the pool says yet.. I've been in the middle of expanding so I couldn't confirm. I will confirm by this weekend, I find it very shady that Palgin drops a miner with a dev fee and disappears with no commentary so I'm very skeptical that the results are real. We've also never heard a single word from the supposed co author, Alex.

[d5coupe]

So has anyone bench-marked the pool results of HSR vs KlausT recently?

My miner-speeds are within 100-150 k/h (HSR being faster) - but is it real, is my biggest question.

I'm currently using nicehash pool since that's where I started & with the new zero-fee transfers to coinbase I am happy with it. -- Does anyone know if there is a more profitable way that nicehash pool? Obviously the BTC exchange rate has a lot to do with it.

Just got my first 12-gpu ring running stable, trying to optimize my operation, thanks for your help!

[NetopyrMan]

Cuda error in func 'neoscrypt_cpu_hash_1060' at line 1408 : an illegal memory access was encountered.

try i- 2 (or 3)
dont OC too much, neoscrypt is wierd

[MagicSmoker]

So has anyone bench-marked the pool results of HSR vs KlausT recently?

My miner-speeds are within 100-150 k/h (HSR being faster) - but is it real, is my biggest question.
...

Before I got my o/c problems sorted out I would say that hsrminer overestimated its hashrate, let's say... but now the pool and hsrminer are in good agreement at ~1220 for a single GTX 1080; that's almost 200 more than I got with ccminer-KlausT.

[Just_a_miner]

6 major cons:
1. No options like -r and -R (similar to ccminer), can't loop for failover, etc
2. No API.
3. After startup I see per-card hashrate... but some time after it not displaying anymore
4. Sometimes it just hangs after swith to dev fee mining.
5. If program terminated, windows appear, with suggestion to End Task... not good for looping.
6. Build f....g linux version!

Please add API support as soon as possible.
Also, it would be great if the API behaves in the same way as the one in CCMiner. That way we can just replace CCMiner with your program.

Guys, check out my fork, API is working now, -r option, and more!

[cl37007]

Definate speed increase over ccminer for my 1080 Ti, not so much for my 1070. Possibly need some tweaking I guess as I have not played with any of the hsrminer options such as -i

It seems to run for longer without crashing the GPU driver than ccminer but on the other hand it crashes the entire system sometimes, something ccminer has never done, obviously I cannot leave it unattended if it is going to do this

Overall seems a little more sensitive to OC settings. Looking forward to future developments, I like the speed increase