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

[papampi]

Does tt, ttli, ttdcr work on Nvidia cards
I can't get them to work
These are in my config
-tt 65,65,65,65,65,65
-ttli 70,70,70,70,70,70
-ttdcr 68,68,68,68,68,68

Code:

ETH - Total Speed: 177.979 Mh/s, Total Shares: 24(5+1+4+3+5+6), Rejected: 0, Time: 00:09 
ETH: GPU1 29.719 Mh/s, GPU2 29.606 Mh/s, GPU3 29.721 Mh/s, GPU4 29.627 Mh/s, GPU5 29.646 Mh/s, GPU6 29.661 Mh/s 
  LBC - Total Speed: 170.560 Mh/s, Total Shares: 42(11+4+8+5+8+6), Rejected: 0 
  LBC: GPU1 28.480 Mh/s, GPU2 28.372 Mh/s, GPU3 28.482 Mh/s, GPU4 28.392 Mh/s, GPU5 28.410 Mh/s, GPU6 28.424 Mh/s 
Incorrect ETH shares: none 
 1 minute average ETH total speed: 177.310 Mh/s 
Pool switches: ETH - 0, LBC - 0 
Current ETH share target: 0x0000000112e0be82 (diff: 4000MH), epoch 136(2.06GB) Current LBC share target: 0x00000001213e7591 (diff: 3801MH) 
DCR coefs (-ttdcr): GPU1 100%, GPU2 100%, GPU3 100%, GPU4 100%, GPU5 100%, GPU6 100% 
GPU1 t=77C fan=84%, GPU2 t=73C fan=74%, GPU3 t=73C fan=76%, GPU4 t=71C fan=72%, GPU5 t=69C fan=69%, GPU6 t=69C fan=69% 

[P00P135]

Does tt, ttli, ttdcr work on Nvidia cards
I can't get them to work
These are in my config
-tt 65,65,65,65,65,65
-ttli 70,70,70,70,70,70
-ttdcr 68,68,68,68,68,68

Code:

ETH - Total Speed: 177.979 Mh/s, Total Shares: 24(5+1+4+3+5+6), Rejected: 0, Time: 00:09 
ETH: GPU1 29.719 Mh/s, GPU2 29.606 Mh/s, GPU3 29.721 Mh/s, GPU4 29.627 Mh/s, GPU5 29.646 Mh/s, GPU6 29.661 Mh/s 
  LBC - Total Speed: 170.560 Mh/s, Total Shares: 42(11+4+8+5+8+6), Rejected: 0 
  LBC: GPU1 28.480 Mh/s, GPU2 28.372 Mh/s, GPU3 28.482 Mh/s, GPU4 28.392 Mh/s, GPU5 28.410 Mh/s, GPU6 28.424 Mh/s 
Incorrect ETH shares: none 
 1 minute average ETH total speed: 177.310 Mh/s 
Pool switches: ETH - 0, LBC - 0 
Current ETH share target: 0x0000000112e0be82 (diff: 4000MH), epoch 136(2.06GB) Current LBC share target: 0x00000001213e7591 (diff: 3801MH) 
DCR coefs (-ttdcr): GPU1 100%, GPU2 100%, GPU3 100%, GPU4 100%, GPU5 100%, GPU6 100% 
GPU1 t=77C fan=84%, GPU2 t=73C fan=74%, GPU3 t=73C fan=76%, GPU4 t=71C fan=72%, GPU5 t=69C fan=69%, GPU6 t=69C fan=69% 

read the readme

[papampi]

Does tt, ttli, ttdcr work on Nvidia cards
I can't get them to work
These are in my config
-tt 65,65,65,65,65,65
-ttli 70,70,70,70,70,70
-ttdcr 68,68,68,68,68,68

Code:

ETH - Total Speed: 177.979 Mh/s, Total Shares: 24(5+1+4+3+5+6), Rejected: 0, Time: 00:09 
ETH: GPU1 29.719 Mh/s, GPU2 29.606 Mh/s, GPU3 29.721 Mh/s, GPU4 29.627 Mh/s, GPU5 29.646 Mh/s, GPU6 29.661 Mh/s 
  LBC - Total Speed: 170.560 Mh/s, Total Shares: 42(11+4+8+5+8+6), Rejected: 0 
  LBC: GPU1 28.480 Mh/s, GPU2 28.372 Mh/s, GPU3 28.482 Mh/s, GPU4 28.392 Mh/s, GPU5 28.410 Mh/s, GPU6 28.424 Mh/s 
Incorrect ETH shares: none 
 1 minute average ETH total speed: 177.310 Mh/s 
Pool switches: ETH - 0, LBC - 0 
Current ETH share target: 0x0000000112e0be82 (diff: 4000MH), epoch 136(2.06GB) Current LBC share target: 0x00000001213e7591 (diff: 3801MH) 
DCR coefs (-ttdcr): GPU1 100%, GPU2 100%, GPU3 100%, GPU4 100%, GPU5 100%, GPU6 100% 
GPU1 t=77C fan=84%, GPU2 t=73C fan=74%, GPU3 t=73C fan=76%, GPU4 t=71C fan=72%, GPU5 t=69C fan=69%, GPU6 t=69C fan=69% 

read the readme

lol ...
Read it 1000 times, it said fan control doesn't work, but nothing about intensity not working too

[Korsakopf]

As I said before, AMD is going to fix the DAG issue for both Vega and Polaris, and new drivers will be available soon.

Hi, Claymore!
Please, don't think that i'm trying to say you are lieer. But all this more looks like a rumor. No any other signs of AMD activity, only your posts. Can you, please, tell us on what is your confidence based? Maybe you can show some screenshots of your communications with AMD and where we could see words of AMD workers telling that they are going to fix this issue?
Thanks!

I tell you what I know from AMD engineers, I'm not going to prove anything, it's up to you believe me or not.

Thanks for your comments on this subject. I'm looking forward to the updates and until then we will just wait patiently. I'm thankful for the work you put into this!

[fr4nkthetank]

@claymore that is quite impressive indeed.

[P00P135]

Does tt, ttli, ttdcr work on Nvidia cards
I can't get them to work
These are in my config
-tt 65,65,65,65,65,65
-ttli 70,70,70,70,70,70
-ttdcr 68,68,68,68,68,68

Code:

ETH - Total Speed: 177.979 Mh/s, Total Shares: 24(5+1+4+3+5+6), Rejected: 0, Time: 00:09 
ETH: GPU1 29.719 Mh/s, GPU2 29.606 Mh/s, GPU3 29.721 Mh/s, GPU4 29.627 Mh/s, GPU5 29.646 Mh/s, GPU6 29.661 Mh/s 
  LBC - Total Speed: 170.560 Mh/s, Total Shares: 42(11+4+8+5+8+6), Rejected: 0 
  LBC: GPU1 28.480 Mh/s, GPU2 28.372 Mh/s, GPU3 28.482 Mh/s, GPU4 28.392 Mh/s, GPU5 28.410 Mh/s, GPU6 28.424 Mh/s 
Incorrect ETH shares: none 
 1 minute average ETH total speed: 177.310 Mh/s 
Pool switches: ETH - 0, LBC - 0 
Current ETH share target: 0x0000000112e0be82 (diff: 4000MH), epoch 136(2.06GB) Current LBC share target: 0x00000001213e7591 (diff: 3801MH) 
DCR coefs (-ttdcr): GPU1 100%, GPU2 100%, GPU3 100%, GPU4 100%, GPU5 100%, GPU6 100% 
GPU1 t=77C fan=84%, GPU2 t=73C fan=74%, GPU3 t=73C fan=76%, GPU4 t=71C fan=72%, GPU5 t=69C fan=69%, GPU6 t=69C fan=69% 

read the readme

lol ...
Read it 1000 times, it said fan control doesn't work, but nothing about intensity not working too

   Note: for NVIDIA cards only temperature monitoring is supported, temperature management is not supported.


You dont really need to use ttli or ttdcr if you set your GPU target temp to 80-90 in afterburner as it will throttle for you.

[Claymore]

Does tt, ttli, ttdcr work on Nvidia cards
I can't get them to work
These are in my config
-tt 65,65,65,65,65,65
-ttli 70,70,70,70,70,70
-ttdcr 68,68,68,68,68,68

-tt: from readme: "Note: for NVIDIA cards only temperature monitoring is supported, temperature management is not supported."
-ttli, -ttdcr are ignored for nvidia (bug), it will be fixed in next update: https://bitcointalk.org/index.php?topic=1433925.msg20046492#msg20046492

[papampi]

-ttli, -ttdcr are ignored for nvidia (bug), it will be fixed in next update: https://bitcointalk.org/index.php?topic=1433925.msg20046492#msg20046492

Thanks a lot for clarification.

[SwaY]

Hi

I'm new here and want to thank you for your great miner at first.

But unfortunately a have a Problem with the monitor tool.

I have two nvidia rigs running - one with Win10Pro and one with nvOC Ubuntu based OS, both using Claymore v9.7 and 384.xx drivers
The Linux one runs fine and monitoring as well - no problems here.

The miner on the windows rig runs very good as well, but here the monitoring dosen't work more than 1.5 hours.
When I restart the system everything works as it should. Most of the time some fan speeds are shown as 0% despite them running, but thats no big deal to me.
But after about 1.5 hours (sometimes 1, sometimes 2) the monitor loses connection to this rig. The rig still is connected and mining, can be pinged...all good.
I run the monitor on the rig itself - but even if a use a different PC in the same network it makes no difference.

Both rigs have a fixed IP by the router, I tried with windows firewall completely turned off but no change.

Before I was using v9.5 and only had that windows rig and monitoring worked without problems for days. The problems came with the update to v9.7.
I did not use v9.6. And using the moitor from v9.5 shows the same behavior.

So to me it looks like something happens that claymore stops reporting. But the only error I found in the log is the fanspeed error 999.


And a question about the use of the monitor tool.
If I use it on rig1 and monitor rig1 and rig2. If rig2 fails somehow the restart.bat I set for that case is executed on rig1 which runs the monitor.
So even if rig1 runs like it should it is restarted because of rig2 failure.
Maybe I got something wrong...but maybe room for improvement.


By the way - anyone tested the new 393.39 drivers already? Worth to install it?

[awright69]

The miner on the windows rig runs very good as well, but here the monitoring dosen't work more than 1.5 hours.
When I restart the system everything works as it should. Most of the time some fan speeds are shown as 0% despite them running, but thats no big deal to me.
But after about 1.5 hours (sometimes 1, sometimes 2) the monitor loses connection to this rig. The rig still is connected and mining, can be pinged...all good.
I run the monitor on the rig itself - but even if a use a different PC in the same network it makes no difference.

Both rigs have a fixed IP by the router, I tried with windows firewall completely turned off but no change.

Before I was using v9.5 and only had that windows rig and monitoring worked without problems for days. The problems came with the update to v9.7.
I did not use v9.6. And using the moitor from v9.5 shows the same behavior.

Same exact problem here. Running ASUS Strix gaming MOBO with 6 ASUS Strix 1060s. Monitor stops / displays erroneous data (e.g., showing fans at 0 speed or temperature at 0C, but fans still working and temperature still in good range on the actual cards), then hours to a day or 2 later, rig actually stops hashing.

[Marvell1]

New version with Vega support is close, currently I see 35-36 MH/s ETH at stock clocks and huge speed for second coins. But it's too hot even at 1.0v for core and 0.9v for memory.

Hey Claymore, have you lowered core clock massively ?  Should be as or more efficient than an 580, it is possible.

Not yet, also I did not change anything in bios, but I'm sure it's possible to make it much more efficient with bios/voltage/clocks changes.

BTW, at "-dcri 500" I get 4600MH/s DCR (at stock clocks) after some Vega-related improvements in assembler code (i.e. these improvements are not available for Polaris cards), now I'm working on SIA and PASC...
At "-dcri 100" I see 35MH/s ETH and 3500MH/s DCR.
If I set 1050MHz for memory (default is 945MHz), I get 40MH/s ETH.

yeah the Vega cards look like they will be great in the winter here where cooling is not an issue , I might build a rack or so of them for dual mining but 300 watts per card is nuts , theres no PSU outthere that can power six of those

[P00P135]

New version with Vega support is close, currently I see 35-36 MH/s ETH at stock clocks and huge speed for second coins. But it's too hot even at 1.0v for core and 0.9v for memory.

Hey Claymore, have you lowered core clock massively ?  Should be as or more efficient than an 580, it is possible.

Not yet, also I did not change anything in bios, but I'm sure it's possible to make it much more efficient with bios/voltage/clocks changes.

BTW, at "-dcri 500" I get 4600MH/s DCR (at stock clocks) after some Vega-related improvements in assembler code (i.e. these improvements are not available for Polaris cards), now I'm working on SIA and PASC...
At "-dcri 100" I see 35MH/s ETH and 3500MH/s DCR.
If I set 1050MHz for memory (default is 945MHz), I get 40MH/s ETH.

yeah the Vega cards look like they will be great in the winter here where cooling is not an issue , I might build a rack or so of them for dual mining but 300 watts per card is nuts , theres no PSU outthere that can power six of those

This or Server power supplies will work just fine. 
http://www.super-flower.com.tw/products_detail.php?class=2&sn=16&ID=119&lang=

[Marvell1]

New version with Vega support is close, currently I see 35-36 MH/s ETH at stock clocks and huge speed for second coins. But it's too hot even at 1.0v for core and 0.9v for memory.

Hey Claymore, have you lowered core clock massively ?  Should be as or more efficient than an 580, it is possible.

Not yet, also I did not change anything in bios, but I'm sure it's possible to make it much more efficient with bios/voltage/clocks changes.

BTW, at "-dcri 500" I get 4600MH/s DCR (at stock clocks) after some Vega-related improvements in assembler code (i.e. these improvements are not available for Polaris cards), now I'm working on SIA and PASC...
At "-dcri 100" I see 35MH/s ETH and 3500MH/s DCR.
If I set 1050MHz for memory (default is 945MHz), I get 40MH/s ETH.

yeah the Vega cards look like they will be great in the winter here where cooling is not an issue , I might build a rack or so of them for dual mining but 300 watts per card is nuts , theres no PSU outthere that can power six of those

This or Server power supplies will work just fine. 
http://www.super-flower.com.tw/products_detail.php?class=2&sn=16&ID=119&lang=

those psu are godly expensive, server style could work but the issue woth those is excessive noise add on to the fact that the pic-ie and pico sellers with resasonble packages have dissapeared (sidehack and finsky) , the vega cards look like too much of a hassle unless you have a perfect climetcontrolled area and deap pockets to buy 2000 watt modular psus.  Dual mining is great and all that but wattage use needs to be around 250 max

[tytanick]

Hi, Claymore, do you know by any chance why nvdia p106-100 are mining with 2MH/s speed on ethereum ?
Is it something i am missing here ?

[Claymore]

Hi, Claymore, do you know by any chance why nvdia p106-100 are mining with 2MH/s speed on ethereum ?
Is it something i am missing here ?

The only similar issue I saw was in Windows 10 before Anniversary update, Nvidia had ridiculous hashrate for 10xx cards. After Anniversary update they fixed it.

[Marvell1]

New version with Vega support is close, currently I see 35-36 MH/s ETH at stock clocks and huge speed for second coins. But it's too hot even at 1.0v for core and 0.9v for memory.

Hey Claymore, have you lowered core clock massively ?  Should be as or more efficient than an 580, it is possible.

Not yet, also I did not change anything in bios, but I'm sure it's possible to make it much more efficient with bios/voltage/clocks changes.

BTW, at "-dcri 500" I get 4600MH/s DCR (at stock clocks) after some Vega-related improvements in assembler code (i.e. these improvements are not available for Polaris cards), now I'm working on SIA and PASC...
At "-dcri 100" I see 35MH/s ETH and 3500MH/s DCR.
If I set 1050MHz for memory (default is 945MHz), I get 40MH/s ETH.

yeah the Vega cards look like they will be great in the winter here where cooling is not an issue , I might build a rack or so of them for dual mining but 300 watts per card is nuts , theres no PSU outthere that can power six of those

This or Server power supplies will work just fine. 
http://www.super-flower.com.tw/products_detail.php?class=2&sn=16&ID=119&lang=

those psu are godly expensive, server style could work but the issue woth those is excessive noise add on to the fact that the pic-ie and pico sellers with resasonble packages have dissapeared (sidehack and finsky) , the vega cards look like too much of a hassle unless you have a perfect climetcontrolled area and deap pockets to buy 2000 watt modular psus.  Dual mining is great and all that but wattage use needs to be around 250 max

So? You said no PSU out there that can power six... now eat the words.

lol BOXED , I what I meant to say was no reasonably priced PSU, however I had no idea there were 2000 watt Modular ATX psu's out of there

[P00P135]

You can still use two 1000w PSU's to run them so you are still kind of wrong 

[Marvell1]

You can still use two 1000w PSU's to run them so you are still kind of wrong 

dual
psus is dangerous had them catch fire before , crossing rails by accident I NEVER use dual psus anymore

[limen_cha]

I'm a Claymore ETH GPU miner. I want to develop a software that has a management window based on the Claymore's kernel. What should I do? How can I get the kernel code for Claymore's? I would even like to pay for it. Can you help me? Thank you very much.

[cashen]

I'm a Claymore ETH GPU miner. I want to develop a software that has a management window based on the Claymore's kernel. What should I do? How can I get the kernel code for Claymore's? I would even like to pay for it. Can you help me? Thank you very much.

LOL - Claymore giving you his CL and ASM sources? Shaq has a better chance of joining the KKK.

I was thinking more like snoop dogg stop smoking weed.