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

[twente-mining]

Hey ppl

I just finished testing the new v 0.6 with most of the Nvidia GTX 10xx cards, results below

GTX 1050 4GB Ti    = 180 Sol/s
GTX 1060 3GB       = 325 Sol/s
GTX 1060 6GB       = 345 Sol/s
GTX 1070 8GB       = 460 Sol/s
GTX 1070 11GB Ti  = 480 Sol/s
GTX 1080 11GB Ti  = 650 - 685 Sol/s

for more details check http://1stminingrig.com/new-dstms-zcash-cuda-miner-0-6-tested-reviewed/

averages from my rigs :

GTX 1070 8GB       = 490 ~ 500 Sol/s
GTX 1070 TI 8GB   = 510 ~ 525 Sol/s ( depending on type )
GTX 1080 TI 11GB = 718 ~ 760 Sol/s ( depending on type )

[1713547577]

1713547577

[1713547577]

1713547577

[1713547577]

1713547577

[CarlOrff]

if anybody did make the switch to DSTM from EWBF for mining with 1080 TI's, how did you modify your OC settings, and did you perceive any notable hash rate improvement over EWBF with the latest 0.6 version of DSTM ?

p.s. just read the notes above: for example on EWBF for ZCASH the memory overclock makes no tangible difference whatsoever, in fact some people even put it to negative to save a bit more on watts. Does it behave differently on DSTM ? I would presume for ZCASH only the Core clock matters.

Thanks in advance

I'm running 8 1080ti's I've been shooting for 4.0 Sols/watt.
My Settings are 67% power, temp limit 70, +100 core clock, fan speed 70%. All of the cards stay around 50-55c at these settings.

With EWBF I was averaging 4850 Sols/s.
With DSTM I'm averaging 5150 Sols/s.

With 1080Ti, I can going to 4,53 sol/W.
You have to down the power and you will stay at 640 sol/s, but with fewer watts.
Make you try...
This is my best result, no other GPU make better result.

[shuweee]

quite new to the whole mining gig, started with claymore doing eth only but reading that ZEC is more suitable for Nvidia as the rig i got at the moment is nothing but 1060 6GB, would it be more worthwhile to do ZEC than ETH? i know there are a lot of considerations, no right or wrong answer to it, but want to see more user's opinions and decide.

[wetblanket]

I just announced new mining proxy software, see the BCT post here: https://bitcointalk.org/index.php?topic=3179895

Open source, no devfee, multi coin/algo, just aiming to help miners out. Building it has been a great learning experience in the mining technology ecosystem. I appreciate the feedback - happy mining. Spread the word!

[jpl]

I just announced new mining proxy software, see the BCT post here: https://bitcointalk.org/index.php?topic=3179895

Open source, no devfee, multi coin/algo, just aiming to help miners out. Building it has been a great learning experience in the mining technology ecosystem. I appreciate the feedback - happy mining. Spread the word!

How exactly does a 'proxy' work?

[dstm]

I was curious if anyone is using HIVE OS with DSTM?  Ever since HIVE OS was upgraded to DSTM v0.6 I started getting the following errors every time I start the miner on each rig (like after a HIVE OS upgrade, or applying any OC setting changes):

If zm isn't able to parse the configuration file it reports the line number on which the error occurred. 
So 'zm.conf:8 invalid parameter' means there was an error on line 8. However your configuration file looks fine.

[dstm]

Hi! I've got a problem.
2018-03-10 21:14:34|gpu_id 2 94 1 unknown error
2018-03-10 21:14:34|gpu 2 unresponsive - check overclocking
2018-03-10 21:14:34|cudaMemcpy 1 failed
2018-03-10 21:14:34|nvmlDeviceGetTemperature failed


2018-03-10 10:41:32|gpu_id 2 72 1 unknown error
2018-03-10 10:41:32|gpu 2 unresponsive - check overclocking
2018-03-10 10:41:33|cudaMemcpy 1 failed
2018-03-10 10:41:33|nvmlDeviceGetTemperature failed


2018-03-10 1:31:29|gpu_id 2 78 2 unknown error
2018-03-10 1:31:29|gpu 2 unresponsive - check overclocking
2018-03-10 1:31:29|cudaMemcpy 1 failed
2018-03-10 1:31:29|nvmlDeviceGetTemperature failed


2018-03-09 8:57:02|nvmlDeviceGetTemperature failed
2018-03-09 8:57:03|gpu_id 2 75 2 unknown error
2018-03-09 8:57:03|gpu 2 unresponsive - check overclocking
2018-03-09 8:57:03|cudaMemcpy 1 failed
2018-03-09 8:57:03|nvmlDeviceGetTemperature failed
2018-03-09 8:57:03|nvmlDeviceGetTemperature failed


2018-03-08 16:27:03|gpu_id 2 67 2 unknown error
2018-03-08 16:27:03|gpu 2 unresponsive - check overclocking
2018-03-08 16:27:03|cudaMemcpy 1 failed
2018-03-08 16:27:03|nvmlDeviceGetTemperature failed



2018-03-08 3:02:00|nvmlDeviceGetTemperature failed
2018-03-08 3:02:00|nvmlDeviceGetTemperature failed
2018-03-08 3:02:00|nvmlDeviceGetTemperature failed
2018-03-08 3:02:00|gpu_id 2 52 0 unknown error
2018-03-08 3:02:00|gpu 2 unresponsive - check overclocking
2018-03-08 3:02:00|cudaMemcpy 1 failed
2018-03-08 3:02:00|nvmlDeviceGetTemperature failed
2018-03-08 3:02:00|nvmlDeviceGetTemperature failed
2018-03-08 3:02:00|nvmlDeviceGetTemperature failed
2018-03-08 3:02:00|nvmlDeviceGetTemperature failed


As i understand, that is the same videocard? Can you paste what mean the number of error? I've got 52, 67, 75, 78, 72, 94.

Right, gpu 2 fails to respond to all kind of calls.
The '52, 67, 75, 78, 72, 94' numbers are there to help debugging, there is nothing wrong with them in your case.

[dstm]

What's this? Never seen it before on any miner, ever.

2018-03-12 08:35:06|#  GPU0  rejected share: [21,"Cannot find job"]

4 times in a couple of hours, using 0.6 on a single 1070.

The pool wasn't able to find the job for the submitted solution.
This could be a bug on pool side.
If this happens more often - pls report the pool you're on - I'll take a look at it.

[dstm]

Anyone know how to convert the targets printed by zm to difficulty?

Code:

2018-03-14 02:54:23 PM|#  server set difficulty to: 001dca01dca01dca01dca01d...

Which difficulty is that?

It's the target set by the pool, difficulty is misleading here.
ZM 0.6.1 will print the difficulty in square brackets.

Code:

server set difficulty to: 00080d62b88772a980000000... [0.99350460]

[dstm]

Here's a real newbie question or two. Running SMOS and dstm on zhash.pro with a single GTX1050 for now. It is running, but always reports sols/W as zero. I presume something is goofed up in this line:   miner.exe --server zhash.pro --user wallet.richard1 --pass x --port 3058 (edited out wallet).

Also, temp is always reported as 61C no matter how I set it (lower or higher) on SM so either it isn't being read or it isn't being changed.

Any ideas?

John

Nvidia has disabled reporting of power consumption for the 1050ti in it's drivers. You can try to downgrade your driver.

[dstm]

Given that this miner uses a bit of CPU power...is there an issue with me running an ETN miner from my CPU while running this miner through my GPUs?

There is no general issue to run a CPU miner in parallel, however the solution rate might decrease.

[dstm]

The miner has been working for a week without problems but it takes two days that this happens every 2 or 3 hours, some solution?

2018-03-15 3:27:04|gpu_id 2 76 0 an illegal instruction was encountered
2018-03-15 3:27:04|gpu 2 unresponsive - check overclocking
2018-03-15 3:27:04|cudaMemcpy 1 failed

This is most likely the result of unstable memory due to overclocking.

[dstm]

@dstm,

Any chance to see % of GPU fan usage at telemetry?

Thanks!

This was already requested - it's on my todo list.

[dstm]

Hi DSTM!
I have a question about telemetry. The telemetry message does not display the latest, up-to-date information, but only the average value? It is quite logical to see the actual, the latest information, taken from the log of the miner, because possibly overclocking was increased, or other manipulations were performed and the average readings would not immediately reflect this change. In my opinion, the current hash is more important in monitoring than the average, which is smeared in time.

You can access current performance values via json-rpc - it reports both current and average values.
The idea behind the averages on the webui: you don't look at it all the time - so if you check it let's say once a day you really want to know how your systems performs on average - however this has the drawbacks you've mentioned. I think it makes sense to add current values to the webui.

[wetblanket]

I just announced new mining proxy software, see the BCT post here: https://bitcointalk.org/index.php?topic=3179895

Open source, no devfee, multi coin/algo, just aiming to help miners out. Building it has been a great learning experience in the mining technology ecosystem. I appreciate the feedback - happy mining. Spread the word!

How exactly does a 'proxy' work?

Simply put, it sits in the middle between your miner(s) and your pool(s), in an effort to consolidate multiple miners/rigs through a single connection to the pool. Your miners connect to to the proxy, and the proxy connects to the pool (just once, even for multiple miners), resulting in a single difficulty for all connected miners.

Without a proxy, your miner connects directly to a pool. There's nothing wrong with this, generally speaking - but with miner crashes/restarts due to overclocking or bugs, the miner has to reconnect to the pool. This can mean that your miner is seen by the pool as 'new' after reconnecting, which is generally considered a bad thing (you can lose out on shares/sharerate as you might be treated as a pool hopper). It's important to note that there are multiple factors involved, and over time pools are likely trying to get better at detecting pool hopping more precisely.

With a mining proxy sitting in the middle, the proxy maintains the connection to the pool on miner disconnect, the proxy keeps track of current difficulty and mining job, etc. When miners reconnect, they receive those same values - essentially picking up where they left off. The pool doesn't see any disconnects, and so any impact from pool-hopping isn't attributed to your account.

In the past, others have stated that a 20% improvement in sharerate/shares/reduction in stale/invalid shares have been achieved through the use of a proxy. I'm interested whether this still holds true, assuming that miners and pools have actually improved over the last couple of years, which may not be that true... 🤔

[jpl]

logfile full error message?..  does it hit a max file size?

[jpl]

I just announced new mining proxy software, see the BCT post here: https://bitcointalk.org/index.php?topic=3179895

Open source, no devfee, multi coin/algo, just aiming to help miners out. Building it has been a great learning experience in the mining technology ecosystem. I appreciate the feedback - happy mining. Spread the word!

How exactly does a 'proxy' work?

Simply put, it sits in the middle between your miner(s) and your pool(s), in an effort to consolidate multiple miners/rigs through a single connection to the pool. Your miners connect to to the proxy, and the proxy connects to the pool (just once, even for multiple miners), resulting in a single difficulty for all connected miners.

Without a proxy, your miner connects directly to a pool. There's nothing wrong with this, generally speaking - but with miner crashes/restarts due to overclocking or bugs, the miner has to reconnect to the pool. This can mean that your miner is seen by the pool as 'new' after reconnecting, which is generally considered a bad thing (you can lose out on shares/sharerate as you might be treated as a pool hopper). It's important to note that there are multiple factors involved, and over time pools are likely trying to get better at detecting pool hopping more precisely.

With a mining proxy sitting in the middle, the proxy maintains the connection to the pool on miner disconnect, the proxy keeps track of current difficulty and mining job, etc. When miners reconnect, they receive those same values - essentially picking up where they left off. The pool doesn't see any disconnects, and so any impact from pool-hopping isn't attributed to your account.

In the past, others have stated that a 20% improvement in sharerate/shares/reduction in stale/invalid shares have been achieved through the use of a proxy. I'm interested whether this still holds true, assuming that miners and pools have actually improved over the last couple of years, which may not be that true... 🤔

Thanks... is it secure and not hackable?

[OlimpFil]

You can access current performance values via json-rpc - it reports both current and average values.

Hi DSTM!

Tell me where and what should be entered in the file json-rpc.th to display the current amount of hashed? Here is the current file:

Request:
-->{"id":1, "method":"getstat"}

Response:
<--{
    "id": 1,
    "result": [{
        "gpu_id":            0,         // GPU CUDA id
        "gpu_name":          "string",  // GPU name
        "gpu_pci_bus_id":    0,         // PCI bus id
        "gpu_pci_device_id": 0,         // PCI device id
        "gpu_uuid":          "string"   // GPU UUID
        "temperature":       0,         // current GPU temperature
        "sol_ps":            0.00,      // current Sol/s
        "avg_sol_ps":        0.00,      // average Sol/s
        "sol_pw":            0.00,      // current Sol/s / Watt
        "avg_sol_pw":        0.00,      // average Sol/s / Watt
        "power_usage":       0.00,      // current power usage
        "avg_power_usage":   0.00,      // average power usage
        "accepted_shares":   0,         // total amount of accepted shares
        "rejected_shares":   0,         // total amount of rejected shares
        "latency":           0          // network latency
    }],

    "uptime":   0,                      // uptime in seconds
    "contime":  0,                      // connection time in seconds (gets reset on reconnect)
    "server":   "string",               // server name
    "port":     0000,                   // port
    "user":     "string",               // username
    "version":  "string",                // zm version
    "error":    null
}



What do you need to write and where?

[wetblanket]

I just announced new mining proxy software, see the BCT post here: https://bitcointalk.org/index.php?topic=3179895

Open source, no devfee, multi coin/algo, just aiming to help miners out. Building it has been a great learning experience in the mining technology ecosystem. I appreciate the feedback - happy mining. Spread the word!

How exactly does a 'proxy' work?

Simply put, it sits in the middle between your miner(s) and your pool(s), in an effort to consolidate multiple miners/rigs through a single connection to the pool. Your miners connect to to the proxy, and the proxy connects to the pool (just once, even for multiple miners), resulting in a single difficulty for all connected miners.

Without a proxy, your miner connects directly to a pool. There's nothing wrong with this, generally speaking - but with miner crashes/restarts due to overclocking or bugs, the miner has to reconnect to the pool. This can mean that your miner is seen by the pool as 'new' after reconnecting, which is generally considered a bad thing (you can lose out on shares/sharerate as you might be treated as a pool hopper). It's important to note that there are multiple factors involved, and over time pools are likely trying to get better at detecting pool hopping more precisely.

With a mining proxy sitting in the middle, the proxy maintains the connection to the pool on miner disconnect, the proxy keeps track of current difficulty and mining job, etc. When miners reconnect, they receive those same values - essentially picking up where they left off. The pool doesn't see any disconnects, and so any impact from pool-hopping isn't attributed to your account.

In the past, others have stated that a 20% improvement in sharerate/shares/reduction in stale/invalid shares have been achieved through the use of a proxy. I'm interested whether this still holds true, assuming that miners and pools have actually improved over the last couple of years, which may not be that true... 🤔

Thanks... is it secure and not hackable?

Can you explain your concerns more deeply?

It's secure if your network is secure. The worst a bad actor could do (if your LAN is unsecured or they are already present on your network) is send invalid shares to your proxy. If that's a concern, then adding proxy-local authentication would address that (already on the list if important to others).

The described method of installation always installs the latest canonical version, so if people are following instructions, they will have the official version, not a modified version.

Without knowing your concerns more deeply, it's hard to answer. But maybe it's better to have this discussion via PM or on the announcement thread to avoid polluting dstm's thread here. https://bitcointalk.org/index.php?topic=3179895

[ManeBjorn]

Thanks for the stats. Nice work.

Hey ppl

I just finished testing the new v 0.6 with most of the Nvidia GTX 10xx cards, results below

GTX 1050 4GB Ti    = 180 Sol/s
GTX 1060 3GB       = 325 Sol/s
GTX 1060 6GB       = 345 Sol/s
GTX 1070 8GB       = 460 Sol/s
GTX 1070 11GB Ti  = 480 Sol/s
GTX 1080 11GB Ti  = 650 - 685 Sol/s

for more details check http://1stminingrig.com/new-dstms-zcash-cuda-miner-0-6-tested-reviewed/