excavator by NiceHash - multi-algorithm NVIDIA GPU miner [1.5.6a] - page 63.

NiceHashSupport

hero member

Activity: 588

Merit: 501

Quote from: Amph on March 16, 2017, 08:27:18 AM

Quote from: NiceHashSupport on March 16, 2017, 07:56:11 AM

Quote from: Amph on March 16, 2017, 04:55:41 AM

i'm not making at all the 440 reported here, but only 1650(412 each) with 4 gpu, 60% tdp +100 core +600 ram

Make sure you are using at least 2 threads. If only 1, then your GPU is not 100% loaded. 1 thread solution is good if you plan to do other stuff with your card. Imagine it as being an intensity switch.

We reached 440 with 115W TDP limited card that has OC set to 2120 core and 4700 mem. With next version, it is about 445 with these same settings.

Another example is 126W TDP limited card, OCed to 2110 core and 4400 mem (micron with BIOS patch) reaching 440 sol.

You can view how far is your card OCed when you launch excavator with overclocking set.

yeah i'm running it with -ca -ca , three thread result in a crash

how can you consume only 115 watt with core 2120? it's impossible to set the gpu at 2120 core with 60% tdp, without having a crash

what you have set on the core? the only thing may be the mem, your is at +900 mine only 600

Due to limited TDP, core is actually at around ~1820. Watt measurement is sw - from excavator or nvidia-smi, not HW one, so HW one is a bit higher.

Amph

legendary

Activity: 3248

Merit: 1070

Quote from: NiceHashSupport on March 16, 2017, 07:56:11 AM

Quote from: Amph on March 16, 2017, 04:55:41 AM

i'm not making at all the 440 reported here, but only 1650(412 each) with 4 gpu, 60% tdp +100 core +600 ram

Make sure you are using at least 2 threads. If only 1, then your GPU is not 100% loaded. 1 thread solution is good if you plan to do other stuff with your card. Imagine it as being an intensity switch.

We reached 440 with 115W TDP limited card that has OC set to 2120 core and 4700 mem. With next version, it is about 445 with these same settings.

Another example is 126W TDP limited card, OCed to 2110 core and 4400 mem (micron with BIOS patch) reaching 440 sol.

You can view how far is your card OCed when you launch excavator with overclocking set.

yeah i'm running it with -ca -ca , three thread result in a crash

how can you consume only 115 watt with core 2120? it's impossible to set the gpu at 2120 core with 60% tdp, without having a crash

what you have set on the core? the only thing may be the mem, your is at +900 mine only 600

NiceHashSupport

hero member

Activity: 588

Merit: 501

Quote from: Amph on March 16, 2017, 04:55:41 AM

i'm not making at all the 440 reported here, but only 1650(412 each) with 4 gpu, 60% tdp +100 core +600 ram

Make sure you are using at least 2 threads. If only 1, then your GPU is not 100% loaded. 1 thread solution is good if you plan to do other stuff with your card. Imagine it as being an intensity switch.

We reached 440 with 115W TDP limited card that has OC set to 2120 core and 4700 mem. With next version, it is about 445 with these same settings.

Another example is 126W TDP limited card, OCed to 2110 core and 4400 mem (micron with BIOS patch) reaching 440 sol.

You can view how far is your card OCed when you launch excavator with overclocking set.

sp_

legendary

Activity: 2926

Merit: 1087

Team Black developer

Quote from: Kompik on March 13, 2017, 12:46:09 PM

using twice -ca -ca, settings. Anyone with similar results or is it just me?

Did you try -ca -ca -ca?

Kompik

sr. member

Activity: 463

Merit: 250

Quote from: Amph on March 16, 2017, 04:55:41 AM

i'm not making at all the 440 reported here, but only 1650(412 each) with 4 gpu, 60% tdp +100 core +600 ram

Yeah i also cannot get 440 at 60% TDP. Because 60% TDP statement does not make sense, what does it mean 50W? 100W? 150W? For MSI cards 60% TDP can be 150W, for EVGA (not FTW) cards the 60% TDP is about 110W.

Amph

legendary

Activity: 3248

Merit: 1070

i'm not making at all the 440 reported here, but only 1650(412 each) with 4 gpu, 60% tdp +100 core +600 ram

Slava_K

hero member

Activity: 677

Merit: 500

Quote from: giagge on March 15, 2017, 04:13:28 AM

Last driver for nvidia is good Huh

.

In windows - yes.

giagge

legendary

Activity: 1134

Merit: 1001

Last driver for nvidia is good Huh

.

m1n1ngP4d4w4n

full member

Activity: 224

Merit: 100

CryptoLearner

Quote from: NiceHashSupport on March 14, 2017, 01:12:34 PM

What cards do you have?

EVGA FTW 1070 x60 all running 378.49

Code:

+-----------------------------------------------------------------------------+
| NVIDIA-SMI 378.49 Driver Version: 378.49 |
|-------------------------------+----------------------+----------------------+
| GPU Name TCC/WDDM | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 GeForce GTX 1070 WDDM | 0000:01:00.0 Off | N/A |
| 33% 63C P2 109W / 111W | 712MiB / 8192MiB | 99% Default |
+-------------------------------+----------------------+----------------------+
| 1 GeForce GTX 1070 WDDM | 0000:02:00.0 Off | N/A |
| 29% 61C P2 107W / 111W | 666MiB / 8192MiB | 99% Default |
+-------------------------------+----------------------+----------------------+
| 2 GeForce GTX 1070 WDDM | 0000:03:00.0 Off | N/A |
| 33% 63C P2 109W / 111W | 666MiB / 8192MiB | 99% Default |
+-------------------------------+----------------------+----------------------+
| 3 GeForce GTX 1070 WDDM | 0000:04:00.0 Off | N/A |
| 34% 64C P2 109W / 111W | 666MiB / 8192MiB | 99% Default |
+-------------------------------+----------------------+----------------------+
| 4 GeForce GTX 1070 WDDM | 0000:05:00.0 Off | N/A |
| 31% 63C P2 111W / 111W | 666MiB / 8192MiB | 99% Default |
+-------------------------------+----------------------+----------------------+
| 5 GeForce GTX 1070 WDDM | 0000:06:00.0 Off | N/A |
| 31% 62C P2 113W / 111W | 666MiB / 8192MiB | 99% Default |
+-------------------------------+----------------------+----------------------+

Smi working correctly too.

NiceHashSupport

hero member

Activity: 588

Merit: 501

What cards do you have?

m1n1ngP4d4w4n

full member

Activity: 224

Merit: 100

CryptoLearner

Quote from: NiceHashSupport on March 13, 2017, 05:47:00 PM

Does nvidia-smi works for reading your power usage?

equihash is just public kernel, but we have one small optimisation ready (more speed with same TDP) for next version.

Yup reading work perfectly with nvidia SMI

Amph

legendary

Activity: 3248

Merit: 1070

Quote from: NiceHashSupport on March 14, 2017, 06:02:13 AM

The log lines you shown are all OK, nothing wrong there.

BTW: can nvidia-smi show you power usage in W ?

yeah

NiceHashSupport

hero member

Activity: 588

Merit: 501

The log lines you shown are all OK, nothing wrong there.

BTW: can nvidia-smi show you power usage in W ?

Amph

legendary

Activity: 3248

Merit: 1070

Quote from: NiceHashSupport on March 13, 2017, 05:47:00 PM

Quote from: Amph on March 13, 2017, 12:47:52 PM

apparently excavator is sending duplicate share can you check this?

Algorithm, pool? Run with -f 0 and paste log here.

Quote

[[2017-03-14 07:51:56.112918] [thread=0x00001684] [info]]

   gpuinfo | GPU#5   TEMP: 58 C   USAGE: 100 %   MEMCTRL: 0 %   POWER: -1 W

[[2017-03-14 07:51:56.119927] [thread=0x00001684] [info]]

   =====> Total power: -6 W   Efficiency: -955531840.000000 H/W

[[2017-03-14 07:51:56.278136] [thread=0x00001774] [debug]]

   device#8 | Found nonce: 0258278f diff=31.9389

[[2017-03-14 07:51:56.278136] [thread=0x00001774] [debug]]

   net | Sending share #7398

[[2017-03-14 07:51:56.278136] [thread=0x00001774] [trace]]

   net | Sending: {"id":7398,"method":"mining.submit","params":["60149-97","6983","2320202020202020","58c792fd","0258278f"]}

[[2017-03-14 07:51:56.317181] [thread=0x0000190c] [trace]]

   net | Received: {"id": 7398, "result": true, "error": null}

[[2017-03-14 07:51:56.317181] [thread=0x0000190c] [info]]

   net | Share #7398 accepted

[[2017-03-14 07:51:59.070607] [thread=0x000012ac] [debug]]

   device#3 | Found nonce: 753260e7 diff=40.3342

[[2017-03-14 07:51:59.071608] [thread=0x000012ac] [debug]]

   net | Sending share #7399

[[2017-03-14 07:51:59.071608] [thread=0x000012ac] [trace]]

   net | Sending: {"id":7399,"method":"mining.submit","params":["60149-97","6983","2120202020202020","58c792fd","753260e7"]}

[[2017-03-14 07:51:59.108663] [thread=0x0000190c] [trace]]

   net | Received: {"id": 7399, "result": true, "error": null}

[[2017-03-14 07:51:59.108663] [thread=0x0000190c] [info]]

   net | Share #7399 accepted

[[2017-03-14 07:51:59.595260] [thread=0x000012ac] [debug]]

   device#3 | Fetched new work, job '6983'

[[2017-03-14 07:51:59.613294] [thread=0x0000111c] [debug]]

NiceHashSupport

hero member

Activity: 588

Merit: 501

Quote from: Amph on March 13, 2017, 12:47:52 PM

apparently excavator is sending duplicate share can you check this?

Algorithm, pool? Run with -f 0 and paste log here.

Quote from: reb0rn21 on March 13, 2017, 01:52:14 PM

@NiceHashSupport

In afterburner ctrl+F

move the desired dot to clock you wish, or better just move the core clock slider (which will move the whole graph up by freq) so desired frequency match voltage you wish, select the freq/volt dot and click "L" which lock like say 2000Mhz to 0.95v (I use that setting for LBC)

problem is you must do same for each GFX on the rig as it can be saved for all cards at once

I see now, didn't even realize this feature existed, thanks

But again, I doubt we will be going that deep, perhaps it would be possible to do (since Afterburner is doing it), but it is too much effort for too little result. Most of the things we need are already achieved with standard overclocking and TDP. Whoever needs more, can use Afterburner.

Quote from: m1n1ngP4d4w4n on March 13, 2017, 01:00:30 PM

Thanks for the new version, power reading doesn't seems to work on my nvidia's 1070 (reading @ -1w) i like the idea of the power efficiency tool. i do about 2570 sols/s @60% tdp (about 112w for me) still a bit slower than ewbf (~2700), but you're catching up !

Does nvidia-smi works for reading your power usage?

equihash is just public kernel, but we have one small optimisation ready (more speed with same TDP) for next version.

reb0rn21

legendary

Activity: 1901

Merit: 1024

@NiceHashSupport

In afterburner ctrl+F

move the desired dot to clock you wish, or better just move the core clock slider (which will move the whole graph up by freq) so desired frequency match voltage you wish, select the freq/volt dot and click "L" which lock like say 2000Mhz to 0.95v (I use that setting for LBC)

problem is you must do same for each GFX on the rig as it can be saved for all cards at once

m1n1ngP4d4w4n

full member

Activity: 224

Merit: 100

CryptoLearner

Thanks for the new version, power reading doesn't seems to work on my nvidia's 1070 (reading @ -1w) i like the idea of the power efficiency tool. i do about 2570 sols/s @60% tdp (about 112w for me) still a bit slower than ewbf (~2700), but you're catching up !

Amph

legendary

Activity: 3248

Merit: 1070

apparently excavator is sending duplicate share can you check this?

Kompik

sr. member

Activity: 463

Merit: 250

Thanks for the new version, this is the best news since there is again a new competition between zCash nvidia miners

. Althought i still find it a little bit slower than ewbf on the same settings. 2450 sols vs 2221 sols (5x 1070 + 1x 1060) using twice -ca -ca, settings. Anyone with similar results or is it just me?

NiceHashSupport

hero member

Activity: 588

Merit: 501

Quote from: Amph on March 13, 2017, 06:04:07 AM

ok, i'm experience a strange bug or something, don't know how to call it, when at launch of the mienr, some gpu don't run, it happen randomly, other time it go right and all gpu begin to hash

btw can you add solo mining? only stratum is supported currently

Run with -f 0 and paste log file here.

Solo mining will never be supported.

Topic: excavator by NiceHash - multi-algorithm NVIDIA GPU miner [1.5.6a] - page 63. (Read 179268 times)