Claymore's Dual Ethereum AMD+NVIDIA GPU Miner v15.0 (Windows/Linux)

[comp]

Is there a problem on sia.nanopool.org
becouse my SIA balance is the same for 5-6 hours despite my Miners runnning properly.

Edit: And I examined Top Miners Balances are the same.

ask nanopool

[P00P135]

The DAG epoch issue: it seems AMD fixed it for Vega in drivers and now they are working on Polaris. New drivers will be available soon, may be in two weeks or so.

Thanks for the update.  I wonder how many people are going to be kicking themselves for toss their AMD cards because of DAG issues.

[Naughtid]

The DAG epoch issue: it seems AMD fixed it for Vega in drivers and now they are working on Polaris. New drivers will be available soon, may be in two weeks or so.

Will that be for R9 nano as well. I tried by R9 nano, it is even worse than Rx

[nadrojcote]

The DAG epoch issue: it seems AMD fixed it for Vega in drivers and now they are working on Polaris. New drivers will be available soon, may be in two weeks or so.

Hallelujah! Excited to be a member of the 30 MH's per card club again.

[I_M]

The DAG epoch issue: it seems AMD fixed it for Vega in drivers and now they are working on Polaris. New drivers will be available soon, may be in two weeks or so.

That's awesome news.  Time to scoop up all the GPUs on ebay!

[vodzilla]

The DAG epoch issue: it seems AMD fixed it for Vega in drivers and now they are working on Polaris. New drivers will be available soon, may be in two weeks or so.

finally good news

[Msobh]

The DAG epoch issue: it seems AMD fixed it for Vega in drivers and now they are working on Polaris. New drivers will be available soon, may be in two weeks or so.

Perfecto

[noviceminer]

For who would like to optimize their rig's performance, let me suggest this blog post: jukesays.wordpress.com

Dude you are advertising your blog. This isnt the place.

[laik2]

Hi guys,
i am news on the mining scene, i build my first 8 rig comment please  

Specs build :

BASEBOARD : ASUS PRIME Z270-P
CPU : Celeron G3900
GPU : 8 x ASUS 1070GTX DUAL 8OG-OC
RAM : 4G Kingston Hyper X
PSU : 2 x EVGA SuperNova G2 750W
OS : nvOC18 linux Version on CORSAIR USB

I actually just run on DUAL_ETH_SC with some good results i think i would ask u if it seems ok

GPU #0: GeForce GTX 1070, 8105 MB available, 15 compute units, capability: 6.1     >>>> As u see this first one have less memory is normal ?

GPU #1: GeForce GTX 1070, 8114 MB available, 15 compute units, capability: 6.1

GPU #2: GeForce GTX 1070, 8114 MB available, 15 compute units, capability: 6.1

GPU #3: GeForce GTX 1070, 8114 MB available, 15 compute units, capability: 6.1

GPU #4: GeForce GTX 1070, 8114 MB available, 15 compute units, capability: 6.1

GPU #5: GeForce GTX 1070, 8114 MB available, 15 compute units, capability: 6.1   >>>> Some problem here but no less memory ...

GPU #6: GeForce GTX 1070, 8114 MB available, 15 compute units, capability: 6.1

GPU #7: GeForce GTX 1070, 8114 MB available, 15 compute units, capability: 6.1

ETH - Total Speed: 253.062 Mh/s, Total Shares: 566(53+71+76+68+75+75+87+83), Rejected: 0, Time: 03:01
ETH: GPU0 31.686 Mh/s, GPU1 31.672 Mh/s, GPU2 31.726 Mh/s, GPU3 31.638 Mh/s, GPU4 31.850 Mh/s, GPU5 30.694 Mh/s, GPU6 31.916 Mh/s, GPU7 31.879 Mh/s

SC - Total Speed: 4217.689 Mh/s, Total Shares: 303(32+37+33+45+37+43+42+34), Rejected: 0
SC: GPU0 528.101 Mh/s, GPU1 527.869 Mh/s, GPU2 528.766 Mh/s, GPU3 527.302 Mh/s, GPU4 530.833 Mh/s, GPU5 511.573 Mh/s, GPU6 531.936 Mh/s, GPU7 531.309 Mh/s

Incorrect ETH shares: none
1 minute average ETH total speed: 251.905 Mh/s
Pool switches: ETH - 0, SC - 0
Current ETH share target: 0x00000000dbe6fece (diff: 5000MH), epoch 135(2.05GB)
Current SC share target: 0x0000000007547ff5 (diff: 150GH)

GPU0 t=65C fan=70%, GPU1 t=69C fan=55%, GPU2 t=68C fan=60%, GPU3 t=70C fan=55%, GPU4 t=69C fan=55%, GPU5 t=69C fan=70%, GPU6 t=68C fan=60%, GPU7 t=70C fan=50%


OC settings :

Core clock +100
Mem clock +1600
PL : 150W per GPU
-dcri 50


About my problem now

On GPU0 = can i do something for this ? Is normal ? Is unstable lost ~2-3 mh/s up and down ...

On GPU5 = actually this GPU is the first one always to softcrash ... Actually i have to clock all GPU individually cause GPU5 cant go up then Core clock +100 / Mem clock +1300 at PL = 150W per gpu, cant handle more then Mem clock at +1250 at 140W per gpu ... Is plug on share PCIe connectors with GPU6, no issue on GPU6 ...

Conclusion : without this 2 little GPU fix i think is fine, with this setting the miner can experiment a soft reset after ~4h and can handle for ~9h, no hard reset needed for now.

I grab anytips !



Another little request for u Claymore

Anyway to add some "average ETH totalspeed" with more details on "s" like : 1m / 30m / 1h / 2h / 4h / 8h, or a way to setup this one and increment like "-dcri +/-"

You are seeing 8105  MB of memory on GPU 0 because its connected to monitor - this is completely normal.

[Vann]

The DAG epoch issue: it seems AMD fixed it for Vega in drivers and now they are working on Polaris. New drivers will be available soon, may be in two weeks or so.

Thanks AMD. Smart move! Everyone believing click bait articles overstating the DAG epoch problem, congrats you got played!

[yany]

For who would like to optimize their rig's performance, let me suggest this blog post: jukesays.wordpress.com

Dude you are advertising your blog. This isnt the place.

There are recurring questions here that this post answers relevantly. I don't get any fee for that, so I don't really get your problem. I honestly like to see the increasing efficiency of the mining rigs around the world. If this post can help in it, why not...

[androstan1234]

My 6-gpu rig runs fine on genoil, but in claymore one of the GPUs has 0 hashrate from the beginning, hangs in OpenCLL, claymore restarts, then the system becomes non-responsive.

Since it runs fine on genoil, I'm not sure what the problem could be?

[yany]

My 6-gpu rig runs fine on genoil, but in claymore one of the GPUs has 0 hashrate from the beginning, hangs in OpenCLL, claymore restarts, then the system becomes non-responsive.

Since it runs fine on genoil, I'm not sure what the problem could be?

PSU, the riser and the distribution of the power cables also can cause this. Especially pay attention on the PSU rails distribution among the VGA cards and try to move the power connectors to ensure they connect well. However some details could help: psu, vgas, risers, anything.

[Odetas]

If you put one of those Nvidia Tesla GPUs in the system will the miner recognize it and use it?

[androstan1234]

My 6-gpu rig runs fine on genoil, but in claymore one of the GPUs has 0 hashrate from the beginning, hangs in OpenCLL, claymore restarts, then the system becomes non-responsive.

Since it runs fine on genoil, I'm not sure what the problem could be?

PSU, the riser and the distribution of the power cables also can cause this. Especially pay attention on the PSU rails distribution among the VGA cards and try to move the power connectors to ensure they connect well. However some details could help: psu, vgas, risers, anything.

I assumed the problem was unique to claymore, since genoil is working.  I don't understand why a riser would work on genoil by not claymore.

The cards are:  XFX RX 480 8GB; MSI RX 580 Armor 8GB; MSI Twin Frozr 4 GB; Powercolor RX 580 Red Dragon 4 GB; and ASUS Rog Strix RX 580 8GB.

PSU is EVGA 1kW.

I don't remember where the risers came from.

Running on Windows 10.

[eth_mining]

Help pls !!

When I try to mine in Windows 10 with 1 RX 470 and 2 RX 570, I get 24 MHS per card after bios mod.. HDMI/monitor is connected to mother board hdmi.. But afterburber will not let me make any change.. Everything is disabled.. If i switch my hdmi connection to one of the cards, Afterburner works and I am able to overclock all cards.. But I only get 27-28 MHS on the card that has HDMI connected.. Remaining 2 cards get 4 MHS .. Any suggestions?

I used dvi-d to vga (with 68 ohm resistors) dummy plug on other 2 cards.. But it wouldnt help at all.. Still 4 MHS..

I also tried using Sapphire trixx but it is not applying overclock settings.. And unable to monitor temperature as well.. Very badly need afterburner or some way to overclock and monitor temperature on claymore...

[crypto.bin101]

Help pls !!

When I try to mine in Windows 10 with 1 RX 470 and 2 RX 570, I get 24 MHS per card after bios mod.. HDMI/monitor is connected to mother board hdmi.. But afterburber will not let me make any change.. Everything is disabled.. If i switch my hdmi connection to one of the cards, Afterburner works and I am able to overclock all cards.. But I only get 27-28 MHS on the card that has HDMI connected.. Remaining 2 cards get 4 MHS .. Any suggestions?

I used dvi-d to vga (with 68 ohm resistors) dummy plug on other 2 cards.. But it wouldnt help at all.. Still 4 MHS..

Try to update windows 10 with latest patch.

[Saphire]

OC settings :

Core clock +100
Mem clock +1600
PL : 150W per GPU
-dcri 50

+1600 for Memory?
Can you paste a screen shot please. I afraid over +800 OC.

[yany]

My 6-gpu rig runs fine on genoil, but in claymore one of the GPUs has 0 hashrate from the beginning, hangs in OpenCLL, claymore restarts, then the system becomes non-responsive.

Since it runs fine on genoil, I'm not sure what the problem could be?

PSU, the riser and the distribution of the power cables also can cause this. Especially pay attention on the PSU rails distribution among the VGA cards and try to move the power connectors to ensure they connect well. However some details could help: psu, vgas, risers, anything.

I assumed the problem was unique to claymore, since genoil is working.  I don't understand why a riser would work on genoil by not claymore.

The cards are:  XFX RX 480 8GB; MSI RX 580 Armor 8GB; MSI Twin Frozr 4 GB; Powercolor RX 580 Red Dragon 4 GB; and ASUS Rog Strix RX 580 8GB.

PSU is EVGA 1kW.

I don't remember where the risers came from.

Running on Windows 10.

I understand, however as they are two different codes, they may load the cards in a bit different way. That may be enough to unveil an unstable connection between your components or some poor quality chips on one of the risers. At least I recommend to try out to change a riser on that card and put it on another one. Same stuff with the cables connected (power/data). This can lead you to the weak point easily, if there's any. The fact that genoil runs with no issue does not mean your rig is fully stable. However, I can see you mixed VGA cards from rather different manufacturers. Claymore's code might have some bug handling this case, but I suppose, there's no any vendor-specific code in the source. Maybe (!) it's sensitive to the mixed sizes of VRAMs.

Anyway, please ensure, that you installed all the updates on your OS (including Creative update in the case of Win10) and use the latest official drivers. with 17.6.2 I had no issues, but I tested Radeon only in a single setup.

[vapourminer]

When I try to mine in Windows 10 with 1 RX 470 and 2 RX 570, I get 24 MHS per card after bios mod.. HDMI/monitor is connected to mother board hdmi.. But afterburber will not let me make any change.. Everything is disabled.. If i switch my hdmi connection to one of the cards, Afterburner works and I am able to overclock all cards.. But I only get 27-28 MHS on the card that has HDMI connected.. Remaining 2 cards get 4 MHS .. Any suggestions?

I used dvi-d to vga (with 68 ohm resistors) dummy plug on other 2 cards.. But it wouldnt help at all.. Still 4 MHS..

the problem is dvi-d.. the dvi to vga adapter with resister trick only works with dvi-i.

you need an active dummy plug.