nicehash / excavator Goto Github PK
View Code? Open in Web Editor NEWNiceHash's proprietary low-level CUDA miner
Home Page: https://www.nicehash.com
NiceHash's proprietary low-level CUDA miner
Home Page: https://www.nicehash.com
Without extracting the file and being alerted by macafee and getting fired from my job, is there a readme file inside which explains usage of the command ?
2.1)
Does nihashminer support this also?
2.2)
Which version?
Is this supported by AMD and which algorithm does it do , or which coin?
Thank you
Hello Support,
I am using nvidia gtx 1070 and latest Nicehash version. It works fine for daggerhashimoto/decreed while it switches to equihash the miner gets stuck with cuda error. This is seriously effecting the mining revenues.. had to check the miner manually every half an hour to restart it
Please see error below
Let me know if anyone have any suggestions on sorting this out
Thanks
Best regards
How to enable events remotely?
For example how to create an event "on_quit"? Or make an analog "{"time":15,"loop":20,"commands":[..."?
Running (4) GTX 1070's on Windows 10 using the excavator+web+restart_script.bat to test I'm only getting about 1.65 kH/s. Trying to figure out what I'm missing. I see all 4 cards in the web status window running at about 90% utilization connected to the US equihash server. Thanks in advance for the help.
I'm getting cuda illegal memory access error, stopping my miner while running daggerhashimoto.
Would be great if you could help.
Regards,
Daniel
I am interested in only mining during certain times, so I would love to add the capability to schedule mining times or to be able to stop the miner via a command line command (Which I can put on an external scheduler).
As topic above says. If yes any estimated time frame?
I noticed (by mistake) that if you call worker.reset in all workers and immediately call algorithm.list, the response is incomplete:
{"algorithms":[{"algorithm_id":0,"name":"equihash","connected":true,"got_job":true,"details":{"total_shares":333,"total_accepted":333,"total_rejected":0,"last_efficiency":1.0},"address":"equihash.eu.nicehash.com:3357","login":"BTCADDRESS:x","workers":[{"worker_id":0,"device_id":0,"params":[],"speed":312.5},{"worker_id":1,"device_id":0,"params":[],"speed":312.5},{"worker_id":2,"device_id":1,"params":[],"speed":187.5},{"worker_id":3,"device_id":1,"params":[],"speed":187.5},{"worker_id":4,"device_id":2,"params":[],"speed":187.5},{"worker_id":5,"device_id":2,"params":[],"speed":312.5},{"worker_id":6,"device_id":3,"params":[],"speed":187.5},{"worker_id":7,"device_id":3,"params":[],"speed":312.5},{"worker_id":8,"device_id":4,"params":[],"speed":62.5},{"worker_id":9,"device_id":4,"params":[],"speed":375.0},{"worker_id":10,"device_id":5,"params":[],"speed":0.0},{"worker_id":11,"device_id":5,"params":[],"speed":250.0},{"worker_id":12,"device_id":6,"params":[],"speed":
The way the response is truncated suggests that there was a problem when retrieving a worker's speed.
Regards.
Hi,
I have a MSI RX 480 Gaming X 8G on stock bios. Since nicehash miner version 2.0.0.9 excavator opens and remains opened indiferent that i use it or not. (cannot be closed, starts itself automatically). Problem still occurs with version 2.0.0.12,
If i start to mine my system gets instant black screen of death or in about 2-3mins. While it works i am getting awesome performance from the video card but it wont last much. Artifacts appear (image uploaded below) and the full system freezes, have to reset from button. Tested with 17.7.1,.17.7,2, and blockchain beta driver. Its the same with every driver.
I guess it has to do something with the hardware acceleration or excavator does some overclocking wich is to much for my card.
It doesnt matter that i mine with excavator or with for example claymore.
Temporary solution is that i start with -na parameter, but then Nicehash miner v2 needs at least 5 minutes to start, stucked on "starting excavator server" message. With -na paraeter my system is rock stable even on 24/7.
In the future could you implement an adjustable overclock for AMD cards? (since it shuts down radeon setting and afterbuner) or start excavator like in miner 2.0.0.8 where it shut itself down after initialization or idk.
Thank you for reading this message.
Hi..
My rig will trigger BSOD if i run this version. Below is my detail:
OS: Windows 7 SP1 x64
GPU: NVIDIA GTX1060
Our framework starts excavator and adds/removes workers based on the most profitable algorithm.
We have noticed that after excavator has been running for some time (1-2 hours), if we stop all workers and add new ones for Decred algorithm (default arguments), the GPUs are stuck at less than 5% load.
Interestingly, the speed reported by the workers seems buggy too, it being 100 times more than GPUs@100% load.
When this happens, if excavator is stopped and started again, the problem goes away.
Hi (cTnko from forums) here, i am working on an GUI for the excavator and i ran into an odd problem.
For some reason the address sent through JSON gets received as lowercase (which obviously isn't valid).
You can use this example (from the API) to test it out, in chrome.
http://127.0.0.1:38080/api?command={"id":1,"method":"algorithm.add","params":["equihash","equihash.eu.nicehash.com:3357","34HKWdzLxWBduUfJE9JxaFhoXnfC6gmePG.test2:x"]}
This is on the latest version 1.2.8 alpha.
Can you confirm?
How to use the algorithm parameters?
Where should I enter them?
For example, I want to use https://github.com/nicehash/excavator/tree/master/amd#equihash , But the documentation does not say where to insert the parameters
1 0 or 1 Disable/enable binary kernels 2 0 or 1 Disable/enable the Global Data Share (requires admin privileges)
It is logical that we need a method algorithm.add https://github.com/nicehash/excavator/tree/master/api#algorithm-add
Example usage: {"id":1,"method":"algorithm.add","params": ["equihash","equihash.eu.nicehash.com:3357","34HKWdzLxWBduUfJE9JxaFhoXnfC6gmePG.test2:x"]}
????????????????????????????
Even use the default json file and bat,but still failed like:
Build number: 412
Provided startup commandline:
excavator.exe -c default_command_file.json -d 2 -f 6 -p 0 -wp 38080 -wi
127.0.0.1 -wa
[23:59:06][0x001ebe18][info] Log started
Excavator has crashed... restarting in 10 seconds...
The driver is 378.78.any help~ thanks
Hi, i was wondering if anyone reported the -or parameter not resetting the overclock after miner closes?
I closed the console via the X button, i also tried CTRL+C, it doesn't seem to work either way.
I would really like for the miner to reset the overclock + TDP/power limit when it closes.
Drivers: 382.53
OS: Win 10 x64
I am using single MSI 1080 X+
Hi,
Is it possible to add a new method to the web api to print the speed of a specific algorithm?
Something like:
algorithm.print.speed
Prints speed of algorithm to console output.
Example usage:
{"id":1,"method":"algorithm.print.speed","params":["0"]}
Command parameter # | Type | Description
1 | string | Algorithm ID.
Example response:
{
"id":1,
"error":null
}
It would be really useful to me.
Thank you in advance.
Keep the great work!
Please release precompiled binary for Linux. You know there is a huge amount of users who use Linux, especially on big mining farms. A binary which works in Ubuntu will be best for most users I guess.
I added some functionality to your index.html - now workers can be added/removed.
Here.
version excavator_v1.3.0 change API? Where can I see the description?
Excavator no longer works on windows 10.
Tested on Creatue update
Previous versions (1.1.4.a) are working fine
P.S. Checked on the other computers - half the computers with Windows 7 it also does not work.
Apparently the problem is not in the version of Windows, but in what that exotic distribution package?
Hi,
I am changing my framework (that uses excavator to mine), to adjust memory speeds for each algorithm in order to maximise my GTX 1080 Tis.
I've noticed that decred, lbry, pascal, sia and blake2s doesn't require fast memory. The lower the memory, the faster excavator mines on this algorithms, because there is more power available to the core, hence higher core clocks with the same TDP.
I have P2 disabled to avoid any problems with going to P0 with +500 memory offset (to mine daggerhashimoto/equihash). Reliability is important with 23 rigs, but this renders me unable to go below 5000Mhz (-500 offset) on the memory using excavator.
I did some investigations in order to find out how faster the mentioned algorithms get with lower memory speeds. NvidiaInspector allowed me to set -1000 offset. This, on top of P2 allowed me to go as low as 4000MHz.
The results relative to P0 are:
decred +0 = 29.20
decred -500 = 29.74 (+1.8%)
decred -1000 = 29.85 (+2.2%)
decred -1500 = 30.34 (+3.9%)
lbry +0 = 3
lbry -500 = 3 (+0%)
lbry -1000 = 3.0397 (+1.3%)
lbry -1500 = 3.0657 (+2.2%)
pascal +0 = 11.0225
pascal -500 = 11.117 (+0.8%)
pascal -1000 = 11.198 (+1.6%)
pascal -1500 = 11.297 (+2.5%)
sia +0 = 17.743
sia -500 = 17.920 (+1%)
sia -1000 = 18.067 (+1.8%)
sia -1500 = 18.263 (+2.9%)
blake2s +0 = 42.84
blake2s -500 = 43.25 (+1%)
blake2s -1000 = 43.64 (+1.8%)
blake2s -1500 = 44.01 (+2.7%)
I know I can achieve -1000 (relative to P0) with excavator setting -500 (maximum) offset in P2, but I don't want to enable P2 due to reliability.
I was wondering if it is possible to allow excavator to apply bigger offsets to memory, in order to go to 4000MHz or even lower (I would like to find out when those algorithms start getting worse due too slow memory speed, but I have the feeling that they could improve even further).
Thank you in advance and keep the awesome work!
This mining dont see my VGA R9 280, why?
i know you can use ExcavatorExtraStartupParameters to set different options. but none of the intensity options seem to work. Even if they did work i think the minimum intensity is around 90% gpu usage.
How can i limit the gpu usage to like 50%, or 20%? if i could do that then i could leave it running while i'm browsing the web and whatever without noticing lag
Is there a release of the miner GUI available that integrates excavator 1.2.9?
I'm not having much luck in trying to replace the version in the bin and update the run scripts.
My rig: 2x480, 2x1070. Mine only 1070s on Excavator.
If do NOT disable AMD support(-na), amd fans reset to 0% and dont changes to other value.
P.S. And phymem.sys, phymem_vista_win7_win8.sys crashing system on Win-7x64, so this issue dont fixed.
Add UPTIME to the excavator API.
This is a indicator that allows you to determine how long the miner works without crashing or rebooting.
It is good to have a command SUMMARY - Output of all information in one JSON
for example:
{ "devices": [{ "device_id": 0, "name": "GeForce GTX 1080 Ti", "gpgpu_type": 1, "details": { "cuda_id": 0, "sm_major": 6, "sm_minor": 1, "bus_id": 7 }, "more_detail": { "id": 1, "error": null, "device_id": 0, "name": "GeForce GTX 1080 Ti", "gpgpu_type": 1, "details": { "cuda_id": 0, "sm_major": 6, "sm_minor": 1, "bus_id": 7 }, "uuid": "GPU-238aa18d-b50f-05b9-3c56-d8c9f1e0d8bd", "gpu_temp": 82, "gpu_load": 100, "gpu_load_memctrl": 77, "gpu_power_usage": 292.1730041503906, "gpu_power_limit_current": 300.0, "gpu_power_limit_min": 125.0, "gpu_power_limit_max": 300.0, "gpu_tdp_current": 120.0, "gpu_clock_core_max": 2075, "gpu_clock_memory": 5705, "gpu_fan_speed": 75, "gpu_fan_speed_rpm": 2869 } }], "algorithms": [{ "algorithm_id": 0, "name": "equihash", "connected": true, "got_job": true, "address": "zen.suprnova.cc:3618", "login": "Kayfolom.node10:x", "workers": [{ "worker_id": 0, "device_id": 0, "params": [], "speed": 356.15 }, { "worker_id": 1, "device_id": 0, "params": [], "speed": 354.01 }] }] }
`Build time: 2017-06-21 20:31:09
Build number: 412
Provided startup commandline:
..\excavator -c ..\default_command_file.json
[14:29:39][0x000008d4][info] Log started
[14:29:42][0x000008d4][info] opencl | Loaded phymem for direct PCIe access.
Press any key to continue . . .`
Then program exits.
Running "excavator+web+restart_script.bat" in admin mode is just an endless loop of excavator has crashed
Hello,
First of, good work on this new miner. Keep up the progress.
I'm having an issue with my 1080TI rigg.
On default it uses only 80% of the power limit set.
When i try to change this trough the config it won't do anything with that setting.
i'll include my config, if i missed something please let me know.
If this is indeed a bug, is there an ETA on a fix?
{"id":1,"method":"device.set.power_limit","params":["0","275"]}, {"id":1,"method":"device.set.power_limit","params":["1","275"]}, {"id":1,"method":"device.set.power_limit","params":["2","275"]}, {"id":1,"method":"device.set.power_limit","params":["3","275"]}
it also gives me an error from time to time
Net error: JSON member "result" not expected type: Array
After this error all cards seem to start using only 1 thread.
Regards,
Twente-Farm
Hello again,
i'm testing the new version. And i'm being met with an error:
'an illegal memory access was encountered' in func 'cuda_eq_
This also crashes the NVidia driver.
running in admin mode gives
CUDA error 'unspecified launch failure' in func 'cuda_eq_run' line 2380
running in admin does not crash driver.
log_2017-06-29_16-31-10.1913.zip
thank you in advance
From time to time, it gives out an unrelated connection status, returns false, although in the miner's logs everything is fine, the work is on, the share are sent
How come when nicehash is installed I get a trojan virus and can barely search the internet?
Add a shortcut to exit the application (Q for example), the combination Ctrl+C is not very convenient (Entails a request to interrupt the bat file).
Thank you in advance!
So I was using a Java Program to start Excavator and it seemed to work fine until I didn't get any shares. It seems that it finds nonces, but just doesn't send it to the server.
However when I was starting excavator using a batch file, everything worked as usual.
What could be reasons for excavator not sending shares?
Log File is attached.
Thanks for the help.
for version 1.2.10a running Windows 7 Sp1 x64 with GTX card
Hi, this is an odd bug i found recently by accident.
What will happen now, all the shares will be rejected (reasons: above target or job not found).
Am i using the API wrong, or is this a bug? (doing the same thing with equihash works just fine)
Can you guys confirm this?
run no win 10, gtx 1060 3gb
[09:09:04][0x000018ac][fatal] wrkr0-0 | CUDA error 'out of memory' in func 'cuda_daggerhashimoto::run' line 1292
Hi,
I am making my own script to use Excavator through its web API.
I am looking for a way to stop a worker, but I want to let it finish and submit the current share.
The closest API method is "worker.free" but it doesn't seem to finish and submit current job.
Thank you.
Cards are not overclocked.
C:\Users\Wojtek\Desktop\Mine\excavator>excavator -a equihash -s eu1-zcash.flypool.org:3333 -u .... -p 0 -d 2 -ca
==================== www.nicehash.com ====================
Excavator v1.1.0a GPU Miner for NiceHash.
Copyright (C) 2017 NiceHash. All rights reserved.
Developed by djeZo with help and contributors of code:
pallas, Vorksholk, bitbandi, ocminer
==================== www.nicehash.com ====================
[13:14:35][0x00001cf0][info] Log started
[13:14:35][0x00001cf0][info] Miner starting up, press [CTRL + C] to quit
[13:14:35][0x00001cf0][info] net | Connecting to eu1-zcash.flypool.org:3333
[13:14:35][0x00002e84][info] net | Connected!
[13:14:35][0x00001f2c][info] device#0 | Initializing thread #0 GPU#0 GeForce GTX 750 Ti
[13:14:35][0x00001554][info] device#2 | Initializing thread #2 GPU#2 GeForce GTX 750 Ti
[13:14:35][0x00000ed4][info] device#4 | Initializing thread #4 GPU#0 GeForce GTX 750 Ti
[13:14:35][0x00002720][info] device#5 | Initializing thread #5 GPU#1 GeForce GTX 750 Ti
[13:14:35][0x00000664][info] device#6 | Initializing thread #6 GPU#2 GeForce GTX 750 Ti
[13:14:35][0x000020f0][info] device#7 | Initializing thread #7 GPU#3 GeForce GTX 750 Ti
[13:14:35][0x000016bc][info] device#1 | Initializing thread #1 GPU#1 GeForce GTX 750 Ti
[13:14:35][0x0000251c][info] device#3 | Initializing thread #3 GPU#3 GeForce GTX 750 Ti
[13:14:35][0x00002e84][info] net | Authorized as ....
[13:14:35][0x00001554][info] device#2 | Algorithm: CUDA-equihash parameters: M=2
[13:14:35][0x00000ed4][info] device#4 | Algorithm: CUDA-equihash parameters: M=2
[13:14:35][0x00001f2c][info] device#0 | Algorithm: CUDA-equihash parameters: M=2
[13:14:35][0x00002720][info] device#5 | Algorithm: CUDA-equihash parameters: M=2
[13:14:36][0x00000664][info] device#6 | Algorithm: CUDA-equihash parameters: M=2
[13:14:36][0x000016bc][info] device#1 | Algorithm: CUDA-equihash parameters: M=2
[13:14:36][0x000020f0][info] device#7 | Algorithm: CUDA-equihash parameters: M=2
[13:14:36][0x0000251c][info] device#3 | Algorithm: CUDA-equihash parameters: M=2
[13:14:36][0x00002e84][info] net | New job '42ff826a869294842a97', diff=470.588
[13:14:36][0x0000251c][error] device#3 | CUDA error 'an illegal memory access was encountered' in func 'cuda_eq_run' line 2019
[13:14:36][0x00000664][error] device#6 | CUDA error 'an illegal memory access was encountered' in func 'cuda_eq_run' line 2043
[13:14:36][0x000016bc][error] device#1 | CUDA error 'an illegal memory access was encountered' in func 'cuda_eq_run' line 2043
[13:14:36][0x0000251c][error] device#3 | CUDA error 'an illegal memory access was encountered' in func 'cuda_equihash::deinit' line 2127
[13:14:36][0x00000664][error] device#6 | CUDA error 'an illegal memory access was encountered' in func 'cuda_equihash::deinit' line 2127
[13:14:36][0x00000ed4][error] device#4 | CUDA error 'an illegal memory access was encountered' in func 'cuda_eq_run' line 2043
[13:14:36][0x000020f0][error] device#7 | CUDA error 'an illegal memory access was encountered' in func 'cuda_eq_run' line 2043
[13:14:36][0x000016bc][error] device#1 | CUDA error 'an illegal memory access was encountered' in func 'cuda_equihash::deinit' line 2127
[13:14:36][0x0000251c][warning] device#3 | Thread detected runtime error, quitting miner...
I noticed that sometimes nist5 algorithm is more profitable on my GTX 1080 Tis (65% power, P2 state disabled) than any other algorithm on NiceHash using ccminer on Windows (I keep track of nist5, neoscrypt, lyra2rev2, blake256r8, daggerhashimoto, decred, cryptonight, lbry, equihash, pascal, x11gost, sia and blake2s profitability). Most of the time lbry/equihash/blake2s are the winners, but sometimes nist5 is more profitable.
Any plans to add this algo in the near future?
Thank you very much in advance.
Keep the awesome work!
Nowhere is a list of supported algorithms:
https://github.com/nicehash/excavator/tree/master/api#info - Nothing
https://github.com/nicehash/excavator - Nothing
https://github.com/nicehash/excavator/blob/master/excavator-man-API.txt - Nothing
https://bitcointalk.org/index.php?topic=1777827.0 - Nothing
README.md - Nothing
I was trying to remember if the Excavator supports the Ethereum, but was surprised to find out that the information was nowhere to get...
Add optional API password (Security token).
I think that the Excavator has a great future, but we need to take care of safety now. Otherwise, it is very easy for an attacker to scan the ports, find the port of the excavator and reconfigure the mining for himself.
example:
excavator.exe -wi 0.0.0.0 -wp 38080 -apipassword c4rc34rcq34
excavator.exe -i 0.0.0.0 -p 38080 -apipassword c4rc34rcq34
for request
{"id":1, "password": "c4rc34rcq34", "method":"device.list","params":[]}
Around 9% of my shares are rejected with the error "Share above target."
Rejected shares appear mostly after "Generating DAG cache..." appears in the output log.
How to stop work via API?
Can add the stop command to the API? Which will return the miner to idle with zero acceleration.
I started excavator using debug level 5 (excavator.exe -na -d 5 -p 5000 -wp 6000 -wi 127.0.0.1) and I was unable to add a worker. I was getting the following error when trying to add a worker for equihash:
[17:31:41][0x00001ff8][fatal] wrkr0-1 | CUDA error 'invalid argument' in func 'cuda_eq_run' line 2356
[17:31:41][0x00000d00][fatal] wrkr0-0 | CUDA error 'invalid argument' in func 'cuda_eq_run' line 2356
[17:31:41][0x00001238][fatal] wrkr1-2 | CUDA error 'invalid argument' in func 'cuda_eq_run' line 2356
and this is the error when trying to add a worker for blake2s:
[17:33:35][0x00002300][fatal] wrkr0-0 | CUDA error 'invalid argument' in func 'cuda_blake2s::run' line 546
[17:33:35][0x000018d8][fatal] wrkr1-1 | CUDA error 'invalid device pointer' in func 'cuda_blake2s::deinit' line 458
[17:33:35][0x00001134][fatal] wrkr7-7 | CUDA error 'invalid device pointer' in func 'cuda_blake2s::deinit' line 458
I didn't try other algorithms.
Happens the same with debug levels 4 and 3. Also, debug level 6 seems to be the same as 0.
If you need anything else from me please let me know.
Unable to use hardware acceleration. Any idea what/how to fix this?
GPU: Sapphire rx480 nitro+ 8gb samsung, Stock Rom
W10 1703, Driver 16.11.5, tried latest relive driver as well, doesn't fix the problem.
[00:24:09][0x00000838][info] core | Found OpenCL device: AMD Radeon RX 480
[00:24:09][0x00000838][info] opencl | PCI Vendor/Device ID: 0x1002, 0x67df
[00:24:09][0x00000838][info] opencl | PCI Subsystem Vendor/Device ID: 0x174b, 0xe347
[00:24:09][0x00000838][info] opencl | Memory Type: GDDR5
[00:24:09][0x00000838][info] opencl | Memory Size: 8192MB
[00:24:09][0x00000838][info] opencl | Video BIOS: 57KB
[00:24:09][0x00000838][error] opencl | Failed to gather information for hardware acceleration.
[00:24:09][0x00000838][error] opencl | Hardware acceleration will not be available.
What criteria needed in order to get it working?
some of my rigs are facing unable to detect video bios issue.
Thanks.
rx480 bios uploaded.
bios.zip
When utilizing excavator it starts up and runs for a few minutes then hangs and eventually it continues and then it hangs again
Only when 4+ Nvidia GPUs. (CPU / Memory are all medium or low usage)
To rule out HW: I switched over to ClaymoreDualMiner and its handling all 4 GPUs fine.
I have multi rigs like lot of people and I would love to view all of them in a single place.
Will it be possible to have one "master" web server in order to display informations of all rigs ?
excavator will connect to an existing management web server on the network like a simple client
If it possible it will be AMAZING π₯
Hi, Excavator version, 1.2.11.a crashes on startup while trying to generate the DAG.
Not to sure how easy a fix will be or not. Would be great to have support for VEGA though.
Thank you very much.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
π Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. πππ
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google β€οΈ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.