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

[$@X]

Also, some output recommendations:

-Latency in ms

-Accepted shares/Rejected shares (and possibly %)

[dstm]

So it looks like this is not working for ZenCash? I've got better performance than EWBF, but it seems the pool is not accepting the shares. Any insight?

Zen should work, it's tested.

If you get rejected shares or some pool isn't working properly pls report the pool-url and the way you started zm such that I'm able to test it.

[dr.]

dstm, Sorry for being persistent, but is there any way to use your miner with a proxy? EWBF, ccminer and, basically, any other miner I've used worked flawlessly with my HTTP proxy server.

Don't hesitate to ask

Could you please describe in more detail how you setup your proxy - I'll try to reproduce why it isn't working.

Sure. I'm using latest 3proxy, my current config looks like this

Has all internet access pass trough the proxy? i.e. is everything else blocked?

Nope, there are some basic service ports opened (i.e. 23, 80, 443 etc. + a couple of special ports, like 20017, which I'm using in a config I've posted earlier), but everything else is closed.
So if there's a need to connect to some exotic pool ports like 3333 etc., I have to individually re-route specific miners through the proxy.

[$@X]

zm.exe --server us.miningspeed.com --user zneRT7DDbLKq16dZwPdPYo5NkfEuFCBLt8n --pass x --port 3062

Hashing was fine, pool just wasn't accepting shares.

After I reset and changed pool to eu.miningspeed.com, I then got about 90% of shares rejected also. Seemed like 100% accepted on us.miningspeed.com, but they wouldn't show up on the pool website. The website is saying 0 shares.

Thanks!

[dstm]

Also, some output recommendations:

-Latency in ms

-Accepted shares/Rejected shares (and possibly %)

What latency exactly?

Accepted/Rejected shares are accesibale accessible via the webui or json-rpc.

[gurupool]

The first block of our pool (https://zpool.guru) is found with DSTM 

I just have a test with you pool : with a 200 sol/s miner, no share, O, null.
I stop the test after few minutes.
In the same time, few minutes, with SuprNova, I get a lot od shares.
Why I get 0 share with your pool, smallest than suprnova ?

I update you on the official post : https://bitcointalk.org/index.php?topic=2262047.msg23207497.0

[$@X]

The latency between my machine and the pool when submitting shares.

I guess I don't use either of those, I just look at the command prompt for reject to accept ratio usually.

Also, just tried on zenmine.pro... It is showing last accepted share and some hash, but the graphs are not working. Seems like they are not communicating with the miner.

This is with Windows miner of course.

Thanks!

[ARTRN]

zm.exe --server us.miningspeed.com --user zneRT7DDbLKq16dZwPdPYo5NkfEuFCBLt8n --pass x --port 3062

Hashing was fine, pool just wasn't accepting shares.

After I reset and changed pool to eu.miningspeed.com, I then got about 90% of shares rejected also. Seemed like 100% accepted on us.miningspeed.com, but they wouldn't show up on the pool website. The website is saying 0 shares.

Thanks!

I'd say this was what happened to me also on miningspeed pool's btcz coin.
edited:typo

[dstm]

zm.exe --server us.miningspeed.com --user zneRT7DDbLKq16dZwPdPYo5NkfEuFCBLt8n --pass x --port 3062

Hashing was fine, pool just wasn't accepting shares.

After I reset and changed pool to eu.miningspeed.com, I then got about 90% of shares rejected also. Seemed like 100% accepted on us.miningspeed.com, but they wouldn't show up on the pool website. The website is saying 0 shares.

Thanks!

Can't reproduce currently - works fine for me - I'll do more tests later.

[gurupool]

The first block of our pool (https://zpool.guru) is found with DSTM 

I just have a test with you pool : with a 200 sol/s miner, no share, O, null.
I stop the test after few minutes.
In the same time, few minutes, with SuprNova, I get a lot od shares.
Why I get 0 share with your pool, smallest than suprnova ?

I'm sorry you had 0 share. We currently have more than 100 workers and good results. Did you use port 3333 or SSL on 4333?

[dstm]

dstm, Sorry for being persistent, but is there any way to use your miner with a proxy? EWBF, ccminer and, basically, any other miner I've used worked flawlessly with my HTTP proxy server.

Don't hesitate to ask

Could you please describe in more detail how you setup your proxy - I'll try to reproduce why it isn't working.

Sure. I'm using latest 3proxy, my current config looks like this

Has all internet access pass trough the proxy? i.e. is everything else blocked?

Nope, there are some basic service ports opened (i.e. 23, 80, 443 etc. + a couple of special ports, like 20017, which I'm using in a config I've posted earlier), but everything else is closed.
So if there's a need to connect to some exotic pool ports like 3333 etc., I have to individually re-route specific miners through the proxy.

This is why it's not working, you have to open 3443 for zm. Zm won't work if it can't submit dev fee shares.

[BadAstroZA]

More testing of Windows version mining ZenCash:

Total   -   3208.00   3.55   904.34W   768 / 0


So that's the 7x Asus DUAL OC  GTX 1070s running the latest version of zm. CPU usage is around 20-30%.

Power limit 90% core voltage +33%,, core clock +107, memory clock +520.
One major difference in this version over what I tested yesterday is that the power consumption graph in Afterburner is much much more stable. Hopefully the rig won't crash now.


Ewbf on same overclock settings:

3143Sol/s at 945 W, so 3.33Sol/W


Now going to run zm with same settings as above but core clock pushed to +120. All other parameters identical. Early results:
Total   -   3272.33   3.46   945.80W



Telemetry screen requests:
1. Add uptime totals
2. Add connected server and port (I sometimes have 3 rigs running, and not all on mining the same thing)
2. Add W after the wattage figure in the 'Total' row (minor request, helps for copying and pasting results to forums etc)

[BadAstroZA]

Here's an interesting one...

What I find curious is the difference in difficulty. Is this something set by the server or set by the miner?


The first shows zm running for maybe 25 minutes... this is 7x GTX 1070s and difficulty of 345.6
The latter is ewbf which has been running for 4 hours.. this is 6x GTX 1070s and a difficulty of 2409.


UPDATE: zm miner after about 30 minutes
3263.15Sol/s    3.45Sol/W    945.90W    812 / 8 Shares


This seems consistently about 4% improvement over ewbf 0.3.4b.

[dstm]

Here's an interesting one...

What I find curious is the difference in difficulty. Is this something set by the server or set by the miner?


The first shows zm running for maybe 25 minutes... this is 7x GTX 1070s and difficulty of 345.6
The latter is ewbf which has been running for 4 hours.. this is 6x GTX 1070s and a difficulty of 2409.


UPDATE: zm miner after about 30 minutes
3263.15Sol/s    3.45Sol/W    945.90W    812 / 8 Shares


This seems consistently about 4% improvement over ewbf 0.3.4b.

Difficulty is set by the server.

In zm each GPU works separately and independent. Each GPU receives from the server it's own difficulty based on it's performance (if the server is smart enough). This has some advantages especially if the performance of your GPUs differs. Multiply 345 by 7(7 equal GPUs) and you'll get 2415.

[BadAstroZA]

OK, interesting. Suprnova displays the total of all cards together in the second rig (ewbf using 6 GPUs) as one figure and displays individual card difficulty for zm (using 7 GPUs).

Why do you do it this way?

[Benefit14snake]

Consider making a version without Dev fee? Happy to pay you once if you make a version that has no reoccurring fee.

[dstm]

OK, interesting. Suprnova displays the total of all cards together in the second rig (ewbf using 6 GPUs) as one figure and displays individual card difficulty for zm (using 7 GPUs).

Why do you do it this way?

Beside some technical advantages, each GPU receives it's own difficulty and an own job from the server this way, which is a more optimal way to distribute work across the solution space.

[dstm]

Consider making a version without Dev fee? Happy to pay you once if you make a version that has no reoccurring fee.

Sry won't do for for obvious reasons.
Just lookup this forum and you'll see what happens if a developer does this.

[Benefit14snake]

Ok thanks anyways.

[BadAstroZA]

Beside some technical advantages, each GPU receives it's own difficulty and an own job from the server this way, which is a more optimal way to distribute work across the solution space.

Cool, thanks for explaining.

Well, so far so good. Will keep running on the 7x GTX 1070s for now and monitor over the day. May switch over the 6x GTX 1070 rig as well depending on stability over the next few hours.

Also going to be doing intermittent mining with my GTX 1070 on my desktop PC (normal workstation running Windows 7).