Pages:
Author

Topic: [Cast XMR] high speed XMR/CryptoNight miner for RX Vega GPUs (2 KHash/s) - page 5. (Read 206404 times)

newbie
Activity: 29
Merit: 0
HI guys,

1)I have 7 vega 56 rig GPU running the new cast XMR for v8. With the 18.51 driver the miner on miningpoolhub.com statistics stayed at 9150 H/S for two weeks. We have not done the HBCC off optimization yet so I was not worried about the drop in hashrate--we used to do 11-12 KH/S on v7. Just this week the hashrate even went up to 10.5 KH/S. But started this Thursday the hashrate on the pool went down. Now it is at 7.64 KH/S. I am a newbie but is it possible that something to do with miningpoolhub.com would have caused this? The hashrate on miningpoolhub.com is the hashrate that the pool thinks we are mining at. Could it be possible that something on their end is causing this? When I look at the graphs it does not seem we are getting many blocks. Is it that the pool has not been optimized for V8 yet?

2) Can someone post their config for CAST-XMR please for vega 56 (I flashed mine to 64)? Want to see if I did everything ok.

3) What other hash rate optimizations can you recommend for my miner config? I looked through this read. I see some people talk about a PPT optimization. Thanks!

Thanks!
newbie
Activity: 17
Merit: 1
If I'd had a crystal ball I may have bought a Vega or 3 instead of all my 1070s but I reckon 3600 h/s @ 500W is pretty good no?

You can get that speed with 2 vegas @ 400W total power (cpu not hashing)

Not in Cryptonight-heavy.

About 1400 per Vega in that algo.

1070s seem on par with Vega in Cryptonite-Heavy to me.

However 5 RX 570's can easily get 4000-4300 with about 500-550 watt.

My 4x6 Vega 64 rigs in Cryptonight Heavy yields 37,2 kH/s with the power usage of 4440W (7 day average)
This is just under 120W/kH. There is ~10% variability between my single GPU's.
newbie
Activity: 72
Merit: 0
Hello,dev,why my miners use cast mining, there are too many Duplicate share on pool side?Do you have any comments to reduce it?  When use xmr-stak or SRBminer , no duplicare share.

This ofen happen on  rigs of vega56, vega 64 ,RX580,RX570

cast version include 1.30 1.50 all  happen
newbie
Activity: 8
Merit: 0
After changing to CryptoNightV8, the hash rate has decreased by 10%. I used SRBMiner before, but since I crash frequently, I am using Cast XMR.
newbie
Activity: 2
Merit: 0
Hello everybody, I'm new here

I try Cast XMR on Linux too.
But only 6 GPU are detected. Rig is 8 cards Vega 56.

(8 are detected on Windows 10)

How can i do for start my 8 vega on Linux ?? I connect 7 cards and linux don't start with errors

Thanks for your help  Wink
jr. member
Activity: 47
Merit: 1
Try to add this option "-G 0,1,2,3,4,5" in the command line args.

I try Cast XMR  cast_xmr-vega-ubuntu_150.tar.gz on Linux. But only one GPU 0 is detected. Rig is 6 cards RX470
Tryied with Driver 18.10 and 18.30 (was NO driver error like below)
Kernel 4.17.19

Please advise

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
[23:53:46] Initializing GPU, loading kernel ...
Detected OpenCL Platform: OpenCL 2.1 AMD-APP (2580.4)
[23:53:47] Driver version not supported.
Please make sure AMDGPU-Pro Driver 18.30 (or later) is installed.
GPU0: Ellesmere (card0) | 32 Compute Units | Memory (MB): 4090 | Intensity: 8 / 10
[23:53:47] Connecting to xmr-eu1.nanopool.org:14444 ...
newbie
Activity: 6
Merit: 0
I try Cast XMR  cast_xmr-vega-ubuntu_150.tar.gz on Linux. But only one GPU 0 is detected. Rig is 6 cards RX470
Tryied with Driver 18.10 and 18.30 (was NO driver error like below)
Kernel 4.17.19

Please advise

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
[23:53:46] Initializing GPU, loading kernel ...
Detected OpenCL Platform: OpenCL 2.1 AMD-APP (2580.4)
[23:53:47] Driver version not supported.
Please make sure AMDGPU-Pro Driver 18.30 (or later) is installed.
GPU0: Ellesmere (card0) | 32 Compute Units | Memory (MB): 4090 | Intensity: 8 / 10
[23:53:47] Connecting to xmr-eu1.nanopool.org:14444 ...
newbie
Activity: 61
Merit: 0
Would love to be able to mine ARTO cash (RTO) with cast XMR ... Is it planned to include this cn-lite variant ?
newbie
Activity: 14
Merit: 0
Compute Mode Not Enabled

why do I get this?
http://prntscr.com/lczaa9

try this:

Code:
switch-radeon-gpu --compute=on autorestart
sr. member
Activity: 379
Merit: 250
Compute Mode Not Enabled

why do I get this?
http://prntscr.com/lczaa9
newbie
Activity: 164
Merit: 0
member
Activity: 131
Merit: 13
In the fray since 2013.
If I'd had a crystal ball I may have bought a Vega or 3 instead of all my 1070s but I reckon 3600 h/s @ 500W is pretty good no?

You can get that speed with 2 vegas @ 400W total power (cpu not hashing)

Not in Cryptonight-heavy.

About 1400 per Vega in that algo.

1070s seem on par with Vega in Cryptonite-Heavy to me.

However 5 RX 570's can easily get 4000-4300 with about 500-550 watt.
jr. member
Activity: 144
Merit: 2
If I'd had a crystal ball I may have bought a Vega or 3 instead of all my 1070s but I reckon 3600 h/s @ 500W is pretty good no?

You can get that speed with 2 vegas @ 400W total power (cpu not hashing)
sr. member
Activity: 372
Merit: 250
The road of excess leads to the palace of wisdom
Hi there AMD cryptonight miners.

I am getting at least 3.6kh/s in cryptonight-heavy consuming around 500W at the wall, with 4 Nvidia cards (and an i7 CPU) and some cool software called CryptoDredge.

I just wondered how that compares to what is possible with AMD rigs in terms of hash rate vs. power consumption.

I am honestly not trying to start a stupid tribal war here, I'm just curious because electricity is SO ridiculously expensive here in Australia and cryptonight variant mining seems to be the only gig left if you are getting ripped off for energy.

If I'd had a crystal ball I may have bought a Vega or 3 instead of all my 1070s but I reckon 3600 h/s @ 500W is pretty good no?
member
Activity: 131
Merit: 13
In the fray since 2013.

Is there an option to autorestart miner if hashing stops or falls below a specific thresshold?


my cryptonite heavy rigs have this problem where pool forcibly disconnects then they go to "SWITCHING ALGORITHM" and just stop mining until i restart the miner!
newbie
Activity: 16
Merit: 1
is it possible to undervolt the cards in the batch file for this miner? Im using RX580's card, my overdriventool voltage won't stick after a few minutes
jr. member
Activity: 128
Merit: 5
Great work to the Cast XMR dev crew, but we are very disappointed that you do not provide support for Linux, specifically Ubuntu OS. This is the leading operating system used by pro miners and by anyone who wants reliable, stable mining, with proper security, monitoring, and non-pirate software. We would love to include this miner in our included profiles for PiMP OS / miner.farm.

Thanks again in advance for considering making it easy for PiMP OS to support this miner!

melt (getpimp.org)

Hi,

be disappointed no more and check here:

http://www.gandalph3000.com/cast_xmr/linux/linux_version_of_cast-xmr-cryptonight-monero-miner-for-rx-vega-gpus/

Regards,
glph3k


Hooray!!! Thank you, I will look into this for the pimps. Have a great day and thank you for responding, and for all your hard work!
newbie
Activity: 16
Merit: 0
Hi dear dev,
How to add failover servers?
member
Activity: 91
Merit: 29
Great work to the Cast XMR dev crew, but we are very disappointed that you do not provide support for Linux, specifically Ubuntu OS. This is the leading operating system used by pro miners and by anyone who wants reliable, stable mining, with proper security, monitoring, and non-pirate software. We would love to include this miner in our included profiles for PiMP OS / miner.farm.

Thanks again in advance for considering making it easy for PiMP OS to support this miner!

melt (getpimp.org)

Hi,

be disappointed no more and check here:

http://www.gandalph3000.com/cast_xmr/linux/linux_version_of_cast-xmr-cryptonight-monero-miner-for-rx-vega-gpus/

Regards,
glph3k
newbie
Activity: 14
Merit: 0

@ut3wt: make sure crossfire and ULPS are disabled.

Maybe DDU and install 18.7.1 drivers? I see you have 000 and 001 there but no GPU are installed on these reg values. You can try to delete them before doing DDU and see if it's working. I had 20x Vega64 on Blockchain drivers, for Monero V8 I did DDU Blockchan drivers, installed 18.7.1 drivers, applied Soft Power tables, Disabled Crossfire and ULPS and all 20 of them are hashing at ~1900 h/s now on MoneroV8.

@ut3wt

I got the solution. I went through the exact same suckage yesterday till I found the trick.
The driver does not give a single f*ck what's in the registry!
I reinstalled the 18.6.1 driver multiple times and manually set those values in the registry and after reboot I checked in AMD Settings and CF was still ON!
I had to manually turn CF off in AMD Setting and the 2 out of 6 cards that were previously hashing at only ~100h/s got back to normal with ~1850h/s.
Now all 6 cards are hashing fine with 18.6.1.
I had to remove prev driver with amdcleanutility and install 18.6.1 from scratch. I also switched to Compute mode, but that my not matter for Vegas.

FYI: BC drivers are shit at V8 indeed. Only ~700h/s

Hi,
many thanks for your help resolve this bug, after solution from @Zorg33. I observate this bug in AMD setting, after switch on\off HBCC memory  and CF in AMD setings All GPUs hashing very good) everything began to work well, I use 18.7.1 driver.

Quote
Conclusion: Utility Switch Radeon GPU 0.9.5 not correct switch trigger...

log:

Code:
[01:22:43] GPU0 | 47°C | Fan 3529 RPM | 1555.9 H/s
[01:22:43] GPU5 | 43°C | Fan 3665 RPM | 1676.6 H/s
[01:22:45] GPU1 | 52°C | Fan 3486 RPM | 1908.1 H/s
[01:22:45] GPU4 | 43°C | Fan 3402 RPM | 1901.9 H/s
[01:22:45] GPU3 | 44°C | Fan 3489 RPM | 1894.0 H/s
[01:22:45] GPU2 | 45°C | Fan 3608 RPM | 1891.2 H/s

[Pool: 'pool.supportxmr.com:7777' | Connected: 2018-10-27 19:36:37]
  5:46:09 (100%) Online
  0:00:00 (  0%) Offline

[Job: #422 | Difficulty: 321407 | Running: 7.0 sec | Avg Job Time: 49.3 sec]

[Hash Rate Avg: 11315.7 H/s]
  1890.5 H/s GPU0
  1909.4 H/s GPU1
  1885.2 H/s GPU2
  1886.6 H/s GPU3
  1885.9 H/s GPU4
  1858.2 H/s GPU5

[Shares Found: 713 | Avg Search Time: 29.1 sec]
713 (100%) Accepted
0 (  0%) Rejected by pool
0 (  0%) Invalid result computation failed
0 (  0%) Could not be submitted because of network error
0 (  0%) Outdated because of job change

[01:22:47] GPU5 | 43°C | Fan 3654 RPM | 1836.4 H/s
Pages:
Jump to: