EWBF's CUDA Zcash miner - page 100. | Bitcointalksearch.org

ShowMeCoins

member

Activity: 129

Merit: 11

Quote from: sublimus on July 11, 2017, 12:12:36 PM

Quote from: hallway_monitor on July 11, 2017, 11:07:18 AM

Anyone here with knowledge of CPU's and PCIe channel utilization? I ask because @Blackhawk67 over at the z.cash forum is claiming that he is getting an extra 70-100 sol/s per gpu out of his 1080ti's when mining with both cpu and gpu. Based on his post it looks like he is getting about 830 sol/s per gpu, but when just mining with gpu's he's back down to normal range. I suspect it has something to do with cpu and pcie channel utilization, but my knowledge only goes so far..

EWBF uses only GPUs to mine.
From the screen shots I see he is using nheqminer to mine on the computer's CPU. IMO.
Is it worth it to use the CPU to mine?
Maybe if you have a spare one handy for replacement to restart your machine ASAP so
you dont lose too much hashing time.
Having a machine down for 3 to five 5 days waiting for a replacement CPU can be costly.
Direct link to external forum's are prohibited but an excerpt of @Blackhawk67's comment would have
been useful to form an enlightened opinion.

4x 1080 ti (GigaByte Aorus) + 1 1060 6G (Gigabyte)
Don't know how , but I've seen it "jump" even higher.
I'll try firing up a CPU miner and see what it does.

**edit**
Nehh, not the result I was hoping for. Could it be that my spikes are vardiff related?
Still a 200 Sol overall increase after letting "zcashd" do some mining.
**/edit**

Code:

INFO 17:38:13: GPU0 Accepted share 41ms [A:29, R:0]
Temp: GPU0 65C GPU1 55C GPU2 59C GPU3 60C GPU4 71C
GPU0: 948 Sol/s GPU1: 1000 Sol/s GPU2: 902 Sol/s GPU3: 1028 Sol/s GPU4: 361 Sol/s
Total speed: 4239 Sol/s
INFO 17:38:18: GPU2 Accepted share 41ms [A:22, R:0]
INFO 17:38:22: GPU0 Accepted share 30ms [A:30, R:0]
INFO 17:38:24: GPU1 Accepted share 32ms [A:26, R:0]
INFO 17:38:25: GPU3 Accepted share 19ms [A:29, R:0]
INFO 17:38:25: GPU0 Accepted share 20ms [A:31, R:0]
INFO 17:38:30: GPU3 Accepted share 24ms [A:30, R:0]
INFO 17:38:33: GPU1 Accepted share 78ms [A:27, R:0]
INFO 17:38:35: GPU0 Accepted share 50ms [A:32, R:0]
INFO 17:38:36: GPU0 Accepted share 23ms [A:33, R:0]
INFO 17:38:37: GPU3 Accepted share 20ms [A:31, R:0]
INFO 17:38:46: GPU4 Accepted share 36ms [A:14, R:0]
Temp: GPU0 65C GPU1 55C GPU2 59C GPU3 60C GPU4 71C
GPU0: 949 Sol/s GPU1: 941 Sol/s GPU2: 934 Sol/s GPU3: 914 Sol/s GPU4: 359 Sol/s
Total speed: 4097 Sol/s
INFO 17:38:47: GPU0 Accepted share 25ms [A:34, R:0]
INFO 17:38:47: GPU2 Accepted share 26ms [A:23, R:0]
INFO 17:38:50: GPU3 Accepted share 36ms [A:32, R:0]
INFO 17:38:53: GPU4 Accepted share 31ms [A:15, R:0]
INFO 17:38:54: GPU0 Accepted share 31ms [A:35, R:0]
INFO 17:38:56: GPU1 Accepted share 31ms [A:28, R:0]
INFO 17:38:57: GPU3 Accepted share 32ms [A:33, R:0]
INFO 17:39:00: GPU0 Accepted share 32ms [A:36, R:0]
INFO 17:39:00: GPU0 Accepted share 24ms [A:37, R:0]
INFO 17:39:02: GPU1 Accepted share 28ms [A:29, R:0]
Temp: GPU0 65C GPU1 56C GPU2 59C GPU3 60C GPU4 71C
GPU0: 964 Sol/s GPU1: 962 Sol/s GPU2: 922 Sol/s GPU3: 926 Sol/s GPU4: 346 Sol/s
Total speed: 4120 Sol/s
INFO 17:39:23: GPU3 DevFee Accepted share
INFO 17:39:25: GPU1 Accepted share 26ms [A:30, R:0]
INFO 17:39:25: GPU2 Accepted share 20ms [A:24, R:0]
INFO 17:39:26: GPU0 Accepted share 21ms [A:38, R:0]

hallway_monitor

newbie

Activity: 25

Merit: 0

Quote from: scryptr on July 11, 2017, 01:06:21 PM

Quote from: hallway_monitor on July 11, 2017, 11:07:18 AM

Anyone here with knowledge of CPU's and PCIe channel utilization? I ask because @Blackhawk67 over at the z.cash forum is claiming that he is getting an extra 70-100 sol/s per gpu out of his 1080ti's when mining with both cpu and gpu. Based on his post it looks like he is getting about 830 sol/s per gpu, but when just mining with gpu's he's back down to normal range. I suspect it has something to do with cpu and pcie channel utilization, but my knowledge only goes so far..

NHEQMINER WILL MINE ON CPU AND GPUS SIMULTANEOUSLY--

If you have a strong CPU (i7, Vishera, Ryzen), it will kick in extra hash. Be sure to monitor CPU temperature. This is not a good idea for a laptop, but a desktop CPU with a good quality cooler should not be a problem.

A laptop may burn if mining on both CPU and GPU. Or, on GPU alone, actually. --scryptr

Well I definitely know the CPU can give you extra hash (small though). However, @blackhawk67 is claiming an addition 50-70 sol/s on each gpu WHEN the cpu was mining alongside, so I was wondering if there was a way leverage those findings for us..

bf-jack

newbie

Activity: 7

Merit: 0

scryptr

legendary

Activity: 1797

Merit: 1028

Quote from: hallway_monitor on July 11, 2017, 11:07:18 AM

Anyone here with knowledge of CPU's and PCIe channel utilization? I ask because @Blackhawk67 over at the z.cash forum is claiming that he is getting an extra 70-100 sol/s per gpu out of his 1080ti's when mining with both cpu and gpu. Based on his post it looks like he is getting about 830 sol/s per gpu, but when just mining with gpu's he's back down to normal range. I suspect it has something to do with cpu and pcie channel utilization, but my knowledge only goes so far..

NHEQMINER WILL MINE ON CPU AND GPUS SIMULTANEOUSLY--

If you have a strong CPU (i7, Vishera, Ryzen), it will kick in extra hash. Be sure to monitor CPU temperature. This is not a good idea for a laptop, but a desktop CPU with a good quality cooler should not be a problem.

A laptop may burn if mining on both CPU and GPU. Or, on GPU alone, actually. --scryptr

sublimus

full member

Activity: 240

Merit: 100

Quote from: hallway_monitor on July 11, 2017, 11:07:18 AM

Anyone here with knowledge of CPU's and PCIe channel utilization? I ask because @Blackhawk67 over at the z.cash forum is claiming that he is getting an extra 70-100 sol/s per gpu out of his 1080ti's when mining with both cpu and gpu. Based on his post it looks like he is getting about 830 sol/s per gpu, but when just mining with gpu's he's back down to normal range. I suspect it has something to do with cpu and pcie channel utilization, but my knowledge only goes so far..

EWBF uses only GPUs to mine.
From the screen shots I see he is using nheqminer to mine on the computer's CPU. IMO.
Is it worth it to use the CPU to mine?
Maybe if you have a spare one handy for replacement to restart your machine ASAP so
you dont lose too much hashing time.
Having a machine down for 3 to five 5 days waiting for a replacement CPU can be costly.
Direct link to external forum's are prohibited but an excerpt of @Blackhawk67's comment would have
been useful to form an enlightened opinion.

hallway_monitor

newbie

Activity: 25

Merit: 0

Anyone here with knowledge of CPU's and PCIe channel utilization? I ask because @Blackhawk67 over at the z.cash forum is claiming that he is getting an extra 70-100 sol/s per gpu out of his 1080ti's when mining with both cpu and gpu. Based on his post it looks like he is getting about 830 sol/s per gpu, but when just mining with gpu's he's back down to normal range. I suspect it has something to do with cpu and pcie channel utilization, but my knowledge only goes so far..

https://i.imgur.com/D3jtIyV.png

https://forum.z.cash/t/nvidia-gtx-1080-ti-gpus/16765/98

sublimus

full member

Activity: 240

Merit: 100

Quote from: usefree on July 11, 2017, 05:52:10 AM

Quote from: QuintLeo on July 09, 2017, 06:31:08 PM

Quote from: usefree on July 09, 2017, 10:48:02 AM

is there some kind of software for updating GPU firmware? Like win Ati Flash and polaris bios editor for amd?

NVidia does some sort of encryption or some such to protect their BIOS from being updated - only option is that SOMETIMES you can flash a BIOS from a "higher level" card of the same manufacturer.

Thanks. I found some usefull notes here
https://www.reddit.com/r/nvidia/comments/63gbit/you_can_now_flash_asus_strix_bios_to_nvidia_gtx/
and here
www.overclock.net/t/1627212/how-to-flash-a-different-bios-on-your-1080-ti

thanks for the info.

usefree

sr. member

Activity: 390

Merit: 251

Mamihlapinatapai

Quote from: QuintLeo on July 09, 2017, 06:31:08 PM

Quote from: usefree on July 09, 2017, 10:48:02 AM

is there some kind of software for updating GPU firmware? Like win Ati Flash and polaris bios editor for amd?

NVidia does some sort of encryption or some such to protect their BIOS from being updated - only option is that SOMETIMES you can flash a BIOS from a "higher level" card of the same manufacturer.

Thanks. I found some usefull notes here
https://www.reddit.com/r/nvidia/comments/63gbit/you_can_now_flash_asus_strix_bios_to_nvidia_gtx/
and here
www.overclock.net/t/1627212/how-to-flash-a-different-bios-on-your-1080-ti

ALTasas

newbie

Activity: 3

Merit: 0

Hello,
I had same issue with miner trying to restart GPU's. Did't find any solution so I wrote ir by myself. It simple .NET console application which starts one or few instances of miner and if miner speed drops below limit just restarts that instance.

https://drive.google.com/drive/folders/0B9xxqRvQzCY_aDJNV0hTVFpLZXM

it's very new, and not tested very well yet.

readme.txt:

Watchdog can run one or more instances of EWBF's CUDA miner and restart it if speed drops below limit. I had issue with this miner when after error it keeps tying to restart GPU's with no luck so I write this simple app.

parameters:

--arguments path to text file which contains string whith arguments should be passed to miner. Dont use --config argument, all arguments should be passed in one line. Don't use --api argument it will start api anyway on 127.0.0.1 because it read miner speed from api. File shold look like this:
--server hushpool.cloud --user t1VhD1ENAiwjEt4Nw24N1xyDSmKfpP3LpS9.RIG002 --pass x --port 5555 --log 1 logfile hush.log --templimit 85 --cuda_devices 0 1 2 3 4 5

--miners how many instances of miner should run. Default and I think optimal is 2

--port api port of first miner instance, all other instances opens port+instance_number

--min_speed speed limit, if speed is equal or below min_speed miner will be restarted. Default is 1

--start_interval you can set intervals between starting instances in seconds. Default is 15

--check_interval you can set interval to check miners speed in seconds. Default is 60

--miner_file path to miner. Default "miner.exe"

If my app helped you, just run it without any parameters for minute or two. Thanks.

NvidiaMining

newbie

Activity: 26

Merit: 0

Hey guys,

i'm very new to mining and have the following issue occuring. Any input is highly appreciated!

One of my two rigs consists of 5 MSI QuickSilver 1070s and a 1000w PSU (Corsair HX1000i) using 0.3.4b. Afterburner settings are:
Power Limit: 55%
Core Clock: +150
Mem Clock: +500

1 of the 1070s is running at 450 Sol/s at 120w and is tweakable, is quickly adjusts to Afterburner adjustments but is usually above 55% power limit, more like 60. This cards is the coldest one at 59 to 60 degrees. All other 4 cards do not really care about their power limit at 55% but stay below 50%, averaging 45%. They vary in temperatures from 64 to 69 degrees, pull 100w and run 400 Sol/s. The core clock on those 4 cards remain at stock levels, too.

I'd like to push the other 4 cards to 450 as well and could live with 125w on all cards, yet my Afterburner Settings do not seem to have any effect. The Afterburner box "apply to similar GPUs" is checked and all GPUs do have the same settings, even if i uncheck the box and view them manually. The entire rig pulls 580w from the wall.

Is my PSU on the light side? It does not seem stressed at all, half of the time its fan does not even spin. Should i reinstall Afterburner?

Thank you all for your help, if you need more info please let me know.
Happy hashing!

http://up.picr.de/29745984pv.jpg

TheHypnoToad

newbie

Activity: 31

Merit: 0

Quote from: moeinbiz on July 07, 2017, 06:32:46 PM

I personally use 127.0.0.1:42000 to monitor a rig on localhost and if you add firewall exception in windows plus port forwarding in router and a DDNS service, this could be accessible from everywhere, I also use an uptime monitor service to ping this address for each rig and send me notification for any down time.
Hope that help you.

Hi this is a great way of restarting if the miner does crash.

How did you add a firewall exception in Windows?

Thanks

Heguli97

full member

Activity: 223

Merit: 101

Quote from: Brucelats on July 10, 2017, 11:32:24 AM

So, may I ask which Zcash miner is best? I want to try my 17x Gtx 1070 on Zcash a bit instead ethereum. For ETH I use Claymore/Genoils, what should I use for Zcash? I can't use Claymore Zcash as it's for AMD.

What do you guys suggest?

This EWBF's miner is the best

Dr_Victor

sr. member

Activity: 954

Merit: 250

Quote from: Brucelats on July 10, 2017, 11:32:24 AM

So, may I ask which Zcash miner is best? I want to try my 17x Gtx 1070 on Zcash a bit instead ethereum. For ETH I use Claymore/Genoils, what should I use for Zcash? I can't use Claymore Zcash as it's for AMD.

What do you guys suggest?

Use EWBF's CUDA Zcash miner (NVIDIA)

Brucelats

sr. member

Activity: 326

Merit: 250

So, may I ask which Zcash miner is best? I want to try my 17x Gtx 1070 on Zcash a bit instead ethereum. For ETH I use Claymore/Genoils, what should I use for Zcash? I can't use Claymore Zcash as it's for AMD.

What do you guys suggest?

sublimus

full member

Activity: 240

Merit: 100

Quote from: Muspelheim on July 09, 2017, 11:24:50 PM

hello bros,

i ve been posted before about another problem. today i woke up and i saw erros.

i checked log file and i saw them too much.

ERROR: Looks like GPU3 are stopped. Restart attempt.
ERROR: Looks like GPU3 are stuck he not respond.
ERROR: Looks like GPU1 are stopped. Restart attempt.
ERROR: Looks like GPU1 are stuck he not respond.
ERROR: Looks like GPU3 are stopped. Restart attempt.
ERROR: Looks like GPU3 are stuck he not respond.
ERROR: Looks like GPU0 are stopped. Restart attempt.

my hardware list ;

* 4x asus gtx 1070(stock, no oc)
* 2x 750w power supply
* 8gb ram
* i3 processor
* z270-p mo.bo.
* 4x v006 risers

bios settings ;

4g decoding -> enabled
changed speeds auto -> gen1

bios updated to the lastest version
windows updated to the lastest version

how can i solve this problem permanently?

thank you for now.

My guess is either it is the driver or
one of the card is causing a glitch.

1. I'd try to reseat the cards. or
2. Run the cards one at a time to identify the culprit. or
3. Uninstall/Reinstall the driver with no cards then reinstall the cards one at a time.

Muspelheim

newbie

Activity: 8

Merit: 0

hello bros,

i ve been posted before about another problem. today i woke up and i saw erros.

i checked log file and i saw them too much.

ERROR: Looks like GPU3 are stopped. Restart attempt.
ERROR: Looks like GPU3 are stuck he not respond.
ERROR: Looks like GPU1 are stopped. Restart attempt.
ERROR: Looks like GPU1 are stuck he not respond.
ERROR: Looks like GPU3 are stopped. Restart attempt.
ERROR: Looks like GPU3 are stuck he not respond.
ERROR: Looks like GPU0 are stopped. Restart attempt.

my hardware list ;

* 4x asus gtx 1070(stock, no oc)
* 2x 750w power supply
* 8gb ram
* i3 processor
* z270-p mo.bo.
* 4x v006 risers

bios settings ;

4g decoding -> enabled
changed speeds auto -> gen1

bios updated to the lastest version
windows updated to the lastest version

how can i solve this problem permanently?

thank you for now.

QuintLeo

legendary

Activity: 1498

Merit: 1030

Quote from: usefree on July 09, 2017, 10:48:02 AM

is there some kind of software for updating GPU firmware? Like win Ati Flash and polaris bios editor for amd?

NVidia does some sort of encryption or some such to protect their BIOS from being updated - only option is that SOMETIMES you can flash a BIOS from a "higher level" card of the same manufacturer.

jimmykl

full member

Activity: 349

Merit: 102

Quote from: Chipychop on July 09, 2017, 11:07:28 AM

Quote from: Mikanoshi on July 09, 2017, 06:16:38 AM

Quote from: Chipychop on July 09, 2017, 01:36:41 AM

which core voltage do you use?

Whichever gives 80% PL, it adjusts voltage automatically. No more than 1.0v usually.

perfectly, I have better hashrate now

one doubt more: power supply required for 6 gtx 1070 with oc? 1000w gold plus enough?

regards

Probably ok if you keep the PL low. I'd go with 1200W if you can. Always pays to have some headroom and more likely to operate in the efficiency sweet spot.

Chipychop

member

Activity: 91

Merit: 10

Quote from: Mikanoshi on July 09, 2017, 06:16:38 AM

Quote from: Chipychop on July 09, 2017, 01:36:41 AM

which core voltage do you use?

Whichever gives 80% PL, it adjusts voltage automatically. No more than 1.0v usually.

perfectly, I have better hashrate now

one doubt more: power supply required for 6 gtx 1070 with oc? 1000w gold plus enough?

regards

usefree

sr. member

Activity: 390

Merit: 251

Mamihlapinatapai

is there some kind of software for updating GPU firmware? Like win Ati Flash and polaris bios editor for amd?

Topic: EWBF's CUDA Zcash miner - page 100. (Read 2164327 times)