[ANN] dstm's ZCash / Equihash Nvidia Miner v0.6.2 (Linux / Windows)

[Atronoss]

So:
yes, you can turn the devfee off.
yes, if you do, you will mine slower.  reason - IAW the devfee string log, it will mine in the dev's pool.  It just will not deposit the coin.

at the end, we didn't make the program.  Give credit where due.  if you don't want to pay the dev fee, then try to learn to write your own.  

I have no issues paying a fee.  I think 1% is more reasonable, but im not a coder and can't complete with them.

Hey, this "discussion" has started from GDDR5X card performance on DSTM vs. EWBF, and one of the users stated (and was instantly insulted by a dumbass in this thread) that in current DSTM performance state EWBF outperforms it on GDDR5X cards + you can turn off fee or set to 1% if you see it is more reasonable.

From my side this was not question about fee itself but performance, and why someone must pay fee on less performing miner...

Currently i am running 6x1080 rig on flypool on EWBF for 24 hrs and will switch to DSTM to check real situation in GDDR5X EWBF vs. DSTM

Fee or not to fee is not the question, though 1% to me also would be prefferable, if you make 100K or 50K per day i dont see big difference, i would live with 10K per day....

If dev see that this is his lifetime opportunity to build a "fortune" that is his choice. We as users still have choice of different miners/algos/pools etc.

Hey dumbass (yes, you) 

this discussion started many posts before, by fecker and one other guy, that was complaining about fees as well, and developer clearly stated out, that he will not reduce fee (and many of us support that). Anyway fecker still whining and complaining, and ok that’s maybe fine, but it is not fine, to clearly express under this thread his support to guy, that makes software to disable the developer fee in DSTM. Such a support doesn't belong to any thread and especially to this thread. If you disable devfee by a third party software, you are not stealing from just developer, but in some way from all of us who pays him (all of us who understands some basic principles of business and also moral). It had nothing to do with GDDR5X. So please, first read, or just STFU, before you call someone dumbass...

...comprende stupido? Thanks! 

[NoOneLt]

Charging 2% here from all profits is just ridiculous , this guy is making millions monthly yet he hides behind anonymity and brilliant "customer service" (yes, you are paying him).

This pretty much.

Here its a working program for fee remove
https://bitcointalk.org/index.php?topic=2800586.0

I DO hope you can get it working under Windows 10. Cheers 

Any 1080 (Non-Ti) users here??? How about your hashrate when using DSTM miner??? 

Yes, standard 1080 FE, running 0.9V(85% TDP), can get ez 2000MHz, but doing 1974MHz, watercooled, keeping it at about 37C, XOC bios(may affect sols in a negative way), +550 on mems.
Doing 570Sols with EKWB. DSTM is not good for GDDR5X cards(1080, 1080Ti).
If you can do 570 with EKWB with 0 fee, why would u do 570 x 0.98 (2% fee)= 558.6 with DSTM? Doesnt make sense.

Fecker, don't you wanna get the "feck" out of here, please? And oh BTW, it is not EKWB, it is EWBF miner you are using, so use it and stop bitching at DSTM thread about fees again and again... EKWB is company that makes watercooling solution.

Quote specially for you mind disabled fella @Atronoss - dude stated some facts and calculations about his rig and observation on GDDR5X and here you jump in telling him what to do.

[Atronoss]

Charging 2% here from all profits is just ridiculous , this guy is making millions monthly yet he hides behind anonymity and brilliant "customer service" (yes, you are paying him).

This pretty much.

Here its a working program for fee remove
https://bitcointalk.org/index.php?topic=2800586.0

I DO hope you can get it working under Windows 10. Cheers 

Any 1080 (Non-Ti) users here??? How about your hashrate when using DSTM miner??? 

Yes, standard 1080 FE, running 0.9V(85% TDP), can get ez 2000MHz, but doing 1974MHz, watercooled, keeping it at about 37C, XOC bios(may affect sols in a negative way), +550 on mems.
Doing 570Sols with EKWB. DSTM is not good for GDDR5X cards(1080, 1080Ti).
If you can do 570 with EKWB with 0 fee, why would u do 570 x 0.98 (2% fee)= 558.6 with DSTM? Doesnt make sense.

Fecker, don't you wanna get the "feck" out of here, please? And oh BTW, it is not EKWB, it is EWBF miner you are using, so use it and stop bitching at DSTM thread about fees again and again... EKWB is company that makes watercooling solution.

Quote specially for you mind disabled fella @Atronoss - dude stated some facts and calculations about his rig and observation on GDDR5X and here you jump in telling him what to do.

So specially for you take a look at this:

"I DO hope you can get it working under Windows 10. Cheers"

as a reply for:

"Here its a working program for fee remove"

[NoOneLt]

So specially for you take a look at this:

"I DO hope you can get it working under Windows 10. Cheers"

as a reply for:

"Here its a working program for fee remove"

I dont care about fee removal, all my statements related to GDDR5X performance on DSTM vs. EWFB, if you do not even have rigs with this cards just pass my posts.

And as i said iam currently doing some tests to confirm or deny feckers words. And after ~30 hrs i will be able to provide results here.

[Atronoss]

So specially for you take a look at this:

"I DO hope you can get it working under Windows 10. Cheers"

as a reply for:

"Here its a working program for fee remove"

I dont care about fee removal, all my statements related to GDDR5X performance on DSTM vs. EWFB, if you do not even have rigs with this cards just pass my posts.

I never said you have something against fee, but you just jumped into long going discussion and called me stupido in first post, and later even dumbass, without trying to understand what was this all about.

And I never said anything about GDDR5X performance, as I said, this dev fee discussion was long going and what I said to fecker was only because of our previous discussion and because of his support for third party software to remove fee.

And BTW yes, I also have one 1080 Ti in my rig.

Nevermind, I guess, we finally cleared the situation and do not need to insult each other anymore.

[NoOneLt]

Can anyone share a real testing results? What miner is better for ZEC on 1070?
For me all of them look the same: 460-480 sols after fine CPU tuning.

Sols "drawed" by miner can be fales, you'd need test actual results on some pool that provides enough infomation about your hashrate, i am trying some testing on flypool. Though i am not sure it gives correct numbers, but at least difference should be seein if there are any difference.

[Atronoss]

Also I'm registering some problems with DSTM's ZM miner 0.5.8 while reconnecting to the pool (it happened just twice for 14 days - nothing terrible):

Computer si WINDOWS 10 ENTERPRISE LTSB 2016 64-BIT, MSI GTX 1080 Ti, 100 Mbps wired network, uptime around 4 days.

On other computers I'm running 0.5.7 and it never happened yet.

http://www.atronoss.com/files/storage/dstm/2018-01-25_081035.png

[dstm]

DSTM I've got another question regarding the API of your miner.
The fields user and version are included in the response, but they are empty.
"user":     "username",           // username
"version":  "0.5.4",              // zm version

Are you aware of that or is this just an error in my case?

Thx for reporting, will check.

Can't reproduce this, working fine for me on both OS.

[dstm]

May i know the formula of 2% dev-fee.

Is sol/s *0.98 or etc.

Someone can tell me. Thanks!

Yes, that's right.

[dstm]

Hi, DSTM.
Recv 10054 is a network, I think I fix it.
Now 10053 visits me (but its very rare). Is this a network problem too?
Send timeout is network too? (I changed WiFi adapter from noname to HighPower Tplink and dont see send timeout for a couple of days)
2018-01-22 19:11:54|#  recv failed: 10053
2018-01-22 19:11:54|#  reconnecting
2018-01-22 19:12:07|#  connected to: eu1-zcash.flypool.org:3333
2018-01-22 19:12:10|#  server set difficulty to: 0004189374bc6a7ef9db22d0...
2018-01-22 19:12:34|>  GPU0  62C  Sol/s: 332.8  Sol/W: 2.95  Avg: 332.8  I/s: 178.5  Sh: 0.00   . .   
2018-01-22 19:12:35|>  GPU2  61C  Sol/s: 301.2  Sol/W: 3.60  Avg: 301.2  I/s: 163.4  Sh: 0.00   . .   
2018-01-22 19:12:36|>  GPU1  58C  Sol/s: 293.9  Sol/W: 3.43  Avg: 293.9  I/s: 159.3  Sh: 0.00   . .   
2018-01-22 19:12:36|>  GPU3  57C  Sol/s: 316.3  Sol/W: 3.51  Avg: 316.3  I/s: 167.9  Sh: 0.00   . .   
2018-01-22 19:12:38|>  GPU4  60C  Sol/s: 335.4  Sol/W: 3.63  Avg: 335.4  I/s: 180.0  Sh: 0.00   . .   
2018-01-22 19:12:40|>  GPU5  62C  Sol/s: 296.2  Sol/W: 3.27  Avg: 296.2  I/s: 162.8  Sh: 2.99   . .   +
2018-01-22 19:12:40|   ========== Sol/s: 1875.8 Sol/W: 3.40  Avg: 1875.8 I/s: 1011.8 Sh: 2.99   . . 
2018-01-22 19:12:45|#  send timeout
2018-01-22 19:12:54|   GPU0  63C  Sol/s: 332.2  Sol/W: 2.94  Avg: 332.5  I/s: 176.8  Sh: 0.00   . .   
2018-01-22 19:12:55|   GPU2  61C  Sol/s: 302.7  Sol/W: 3.49  Avg: 302.0  I/s: 160.9  Sh: 1.49   . .   +
2018-01-22 19:12:55|#  send timeout
2018-01-22 19:12:56|#  send timeout
2018-01-22 19:12:56|   GPU1  61C  Sol/s: 291.8  Sol/W: 3.42  Avg: 292.9  I/s: 156.7  Sh: 0.00   . .   
2018-01-22 19:12:56|   GPU3  58C  Sol/s: 309.3  Sol/W: 3.49  Avg: 312.8  I/s: 165.8  Sh: 0.00   . .   
2018-01-22 19:12:58|   GPU4  60C  Sol/s: 339.4  Sol/W: 3.53  Avg: 337.4  I/s: 177.5  Sh: 1.50   . .   +
2018-01-22 19:13:00|   GPU5  63C  Sol/s: 289.6  Sol/W: 3.19  Avg: 292.9  I/s: 161.0  Sh: 5.98   . .   +++
2018-01-22 19:13:00|   ========== Sol/s: 1865.0 Sol/W: 3.34  Avg: 1870.4 I/s: 998.5  Sh: 8.97   . . 
2018-01-22 19:13:08|#  send timeout
2018-01-22 19:13:14|   GPU0  63C  Sol/s: 327.1  Sol/W: 2.92  Avg: 330.7  I/s: 176.5  Sh: 0.00   . .   
2018-01-22 19:13:15|   GPU2  61C  Sol/s: 298.5  Sol/W: 3.44  Avg: 300.8  I/s: 159.9  Sh: 1.99   . .   +
2018-01-22 19:13:16|#  send timeout
2018-01-22 19:13:16|   GPU1  61C  Sol/s: 293.8  Sol/W: 3.41  Avg: 293.2  I/s: 156.1  Sh: 0.00   . .   
2018-01-22 19:13:16|   GPU3  59C  Sol/s: 319.4  Sol/W: 3.51  Avg: 315.0  I/s: 165.4  Sh: 0.00   . .   
----------------------------------------------------------------------------------------------------------------------------------
========== Sol/s: 1882.1 Sol/W: 3.31  Avg: 1864.7 I/s: 1005.4 Sh: 6.80   0.99 100
2018-01-23 19:52:16|   GPU2  61C  Sol/s: 301.5  Sol/W: 3.33  Avg: 300.9  I/s: 163.1  Sh: 1.07   0.99 94 
2018-01-23 19:52:16|   GPU3  59C  Sol/s: 321.4  Sol/W: 3.51  Avg: 309.2  I/s: 168.6  Sh: 1.16   0.99 78 
2018-01-23 19:52:16|   GPU0  63C  Sol/s: 327.6  Sol/W: 3.04  Avg: 328.6  I/s: 176.5  Sh: 1.17   0.99 183
2018-01-23 19:52:28|cudaMemcpy 2 failed
This is an overclocking problem, I think? But what GPU? The numeration is similar to 0123?
----------------------------------------------------------------------------------------------------------------------------------
And thew last question. My First rig running on a 0.5.7 and has a brilliant lost shares (0.6%)
uptime:     1 days  14:14:29
contime:     1 days  14:14:26
version:   zm 0.5.7
server:   eu1-zcash.flypool.org:3333
ID   DEVICE NAME   °C   ∅ SOL/S   ∅ SOL/W   ∅ WATT   SHARES   LAT
0   GeForce GTX 1070   58   478.96   3.37   142.23   3962 / 26   78
1   GeForce GTX 1070   58   470.71   3.31   142.18   3917 / 26   78
2   GeForce GTX 1070   58   467.81   3.30   141.56   3911 / 30   78
3   GeForce GTX 1070   59   460.86   3.24   142.21   3843 / 30   78
4   GeForce GTX 1070   60   471.97   3.69   127.96   3979 / 22   78
Total   -   2350.32   3.38   696.13   19612 / 134   78
The second rig running on a 0.5.8 and has more than a 1 percent of lost shares, I change on 0.5.7 and will see (for now less than 1 percent).
Thanks.

Yes, there is still something wrong with your network.

Concerning rejected shares: it depends on multiple things, however it depends also on the quality of your network connection. 0.5.8 has no changes affecting this - don't forget you have to run it for some time to get proper statistics about the rejected share ratio.

[dstm]

I tried switching from ewbf to dstm for the 2nd time yesterday. Initially it started out 5-10% faster than ewbf which is awesome, however after about 10-15 minutes it slowed more & more to the point where it was as much as 40% slower than ewbf, at which point I switched back.
I've talked to others that do not have the same problem, yet they were running less GPUs than I am. I have multiple rigs with 13x1070s & 1070tis.

I have the same problem with DSTM 0.5.6, 0.5.7 and 0.5.8. Stable and fast as hell, but after one hour it starts to make less and less sol/s.. in three hours I'm getting half rate (I start with 680-700 sol/s and after 4 hours I end with 230 sol/s)... so my only solution is restart the miner. All the time no more than 60ºC, fan speed at 60%. I have two rigs, same specs and the same problem: Asrock H110 PRO + 13 Nvidia 1080Ti (up five different brands, mixed). Every rig have 4 PSUs: 3 x 1000W + 1 x 1600W and just one cable to power every riser (4 pin molex version) so no problems about power or temperature. I use simplemining without overclock, just power at 200W, no extra mhz in gpu core or memory. Is it a configuration problem?

Yep same motherboard, never had this issue with any other miner but I'd love to get it working & get the extra sols. Anyone else have any ideas?

Are you all using SMOS?
Some SMOS user reported something similar recently.
https://bitcointalk.org/index.php?topic=2021765.msg27398311#msg27398311

If you take a look at the provided data: affected systems seems to have a lot 'nvidia-settings' processes running, this might be the reason for the performance drop after some time. So pls check if this is also the case on your system - It looks like SMOS is starting nvidia-setting (most likely for monitoring) without properly closing it.

[dstm]

Also I'm registering some problems with DSTM's ZM miner 0.5.8 while reconnecting to the pool (it happened just twice for 14 days - nothing terrible):

Computer si WINDOWS 10 ENTERPRISE LTSB 2016 64-BIT, MSI GTX 1080 Ti, 100 Mbps wired network, uptime around 4 days.

On other computers I'm running 0.5.7 and it never happened yet.

Thx for reporting.
I've a test case for this, something like reconnecting every 10sec/randomly for a couple hours - it works on my systems fine.
Be aware that zm stops all GPUs during connection loss to save energy - if your system is unstable/overclocked it's likely to crash during stop/start cycles.

I'll let the test case run again to make sure there are no issues.

[MrTDHP]

Have you figured out the weird case where hashrate of a card dropping to about half of it after hours of completely stable running?
The rig has to be restarted in order to restore the appropriate hashrate.

[NoOneLt]

So here is the first part of report, i have run EWBF with --fee 0 for ~28 hrs. on 6x1080 rig. Little exploanation:

On the right there is flypool stats with highlightet 24 hrs window start, below is miner screen at earlyest time i was able to scroll to (2018-01-28 ~11:00)
On the left same flypool but end of test highlighted + screen from miner at test end.

Dont look at miner heading it is coming from bat file naming istance, so i just forgot to change to flypool.

So even with --fee 0 real hashrate is ~2% less then reported.

Now DSTM will run for good amount of time and i check it somwhere on saturday.

Screen:



It is big screen actually, so i leave full link https://imageking.eu/images/ewbfcompar.png

[Caze]

Sorry if this has been asked before but with a quick search I couldn't find it mentioned anywhere. Is the sol/s displayed the effective hashrate (meaning minus the devfee) or with the devfee (meaning our effective hashrate is the displayed sol/s*0.98)?

[Atronoss]

Sorry if this has been asked before but with a quick search I couldn't find it mentioned anywhere. Is the sol/s displayed the effective hashrate (meaning minus the devfee) or with the devfee (meaning our effective hashrate is the displayed sol/s*0.98)?

It is total hashrate, so you have to multiply it by 0,98.

[SpecterHome]

========== Sol/s: 1882.1 Sol/W: 3.31  Avg: 1864.7 I/s: 1005.4 Sh: 6.80   0.99 100
2018-01-23 19:52:16|   GPU2  61C  Sol/s: 301.5  Sol/W: 3.33  Avg: 300.9  I/s: 163.1  Sh: 1.07   0.99 94 
2018-01-23 19:52:16|   GPU3  59C  Sol/s: 321.4  Sol/W: 3.51  Avg: 309.2  I/s: 168.6  Sh: 1.16   0.99 78 
2018-01-23 19:52:16|   GPU0  63C  Sol/s: 327.6  Sol/W: 3.04  Avg: 328.6  I/s: 176.5  Sh: 1.17   0.99 183
2018-01-23 19:52:28|cudaMemcpy 2 failed
This is an overclocking problem, I think? But what GPU? The numeration is similar to 0123?

and what about cudaMemcpy 2 failed.
And the numder, is this a 3rd videocard?

[Cryptoguy2017]

Just move to 5.7 its the stablest one. 5.8 was only made for adding anti dev fee remove functions....

Just my 2 cents....

[kaleb]

So for sure liking this miner over EWBF im seeing about a 30sol/s jump in miner and a 50sol/s jump in pool and my projected earnings are ACTUALLY what they should be after dev fee and not 10% less like EWBF... Now my GPU is liquid cooled and mines at 40*c so my results arent exactly normal....but with a GTX 1080 (non ti) Power+12% core +125   Mem +500 for 2100Mhz on core and 6000Mhz on memory Im pushing AVG:643 Sol/s in miner....thats not even the best part... on pool after 24hrs my 24hr average hash rate is 697sol/s

[siforek]

I tried switching from ewbf to dstm for the 2nd time yesterday. Initially it started out 5-10% faster than ewbf which is awesome, however after about 10-15 minutes it slowed more & more to the point where it was as much as 40% slower than ewbf, at which point I switched back.
I've talked to others that do not have the same problem, yet they were running less GPUs than I am. I have multiple rigs with 13x1070s & 1070tis.

I have the same problem with DSTM 0.5.6, 0.5.7 and 0.5.8. Stable and fast as hell, but after one hour it starts to make less and less sol/s.. in three hours I'm getting half rate (I start with 680-700 sol/s and after 4 hours I end with 230 sol/s)... so my only solution is restart the miner. All the time no more than 60ºC, fan speed at 60%. I have two rigs, same specs and the same problem: Asrock H110 PRO + 13 Nvidia 1080Ti (up five different brands, mixed). Every rig have 4 PSUs: 3 x 1000W + 1 x 1600W and just one cable to power every riser (4 pin molex version) so no problems about power or temperature. I use simplemining without overclock, just power at 200W, no extra mhz in gpu core or memory. Is it a configuration problem?

Yep same motherboard, never had this issue with any other miner but I'd love to get it working & get the extra sols. Anyone else have any ideas?

Are you all using SMOS?
Some SMOS user reported something similar recently.
https://bitcointalk.org/index.php?topic=2021765.msg27398311#msg27398311

If you take a look at the provided data: affected systems seems to have a lot 'nvidia-settings' processes running, this might be the reason for the performance drop after some time. So pls check if this is also the case on your system - It looks like SMOS is starting nvidia-setting (most likely for monitoring) without properly closing it.

Having this issue & yes, on SMOS. Would just replace SMOS with Ubuntu if I could but only have remote access to rigs. Might need to mod SMOS source to fix :/