@claymore
is it possible to disable temp and fan monitoring for the NV cards?
Topic: Claymore's Dual Ethereum AMD+NVIDIA GPU Miner v15.0 (Windows/Linux) - page 1125. (Read 6590718 times)
September 12, 2016, 04:13:42 AM
September 12, 2016, 01:41:39 AM
Is there a command in Claymores miner that allows you to restart the miner if a card drops below a certain hashrate?
I noticed that I was running a 29 mhs low power mod and after 1 hour GPU 1 dropped to 25. Restarted the miner and both GPUs are hashing at 29 MHs with no issues for 4 hours.
Thanks
I noticed that I was running a 29 mhs low power mod and after 1 hour GPU 1 dropped to 25. Restarted the miner and both GPUs are hashing at 29 MHs with no issues for 4 hours.
Thanks
Had the same problem -- upgraded to Crimson 16.9.1 -- no problems now!
September 12, 2016, 01:39:02 AM
SOMETHING NOTABLE ABOUT NVIDIA CARDS--
Today I tried increasing the intensity for the secondary coin on my EVGA GTX CLASSIFIED 980ti card. I was able to more than double the hashrate for SiaCoin (SC) from 210MH/s to 540MH/s by increasing the "-dcri" intensity value from default to "-dcri 75". The card heated up from 77 deg F to 83 deg F, but the Ethereum (ETH) hashrate remained close to 22MH/s. The temperature of 83 deg F is the default target for this 980ti card.
The EVGA CLASSIFIED 980ti does not hash ETH much faster than a GTX 970, but it has greater compute power. This difference is shown in the large increase in hashrate for the secondary coin (SC) by just increasing the "dcri" intensity value. There is room for experimentation here. --scryptr.
you better sell this hungry monster and buy 1070 or rx480. You'll have 30 eth and 450sc easy with only 150-200 watts (depend on your downvolting and optimization skills)Today I tried increasing the intensity for the secondary coin on my EVGA GTX CLASSIFIED 980ti card. I was able to more than double the hashrate for SiaCoin (SC) from 210MH/s to 540MH/s by increasing the "-dcri" intensity value from default to "-dcri 75". The card heated up from 77 deg F to 83 deg F, but the Ethereum (ETH) hashrate remained close to 22MH/s. The temperature of 83 deg F is the default target for this 980ti card.
The EVGA CLASSIFIED 980ti does not hash ETH much faster than a GTX 970, but it has greater compute power. This difference is shown in the large increase in hashrate for the secondary coin (SC) by just increasing the "dcri" intensity value. There is room for experimentation here. --scryptr.
NOT BRAGGING ABOUT MY CARD--
Thanks for the advice, but I am posting a note regarding the relative hashing rate increase on the secondary coin (in my case, SiaCoin [SC]) on nVidia cards. The compute units, CUDA cores, kick in on the intensity increase (dcri = 75). The secondary coin is "compute intensive", rather than "memory intensive", as is ETH, or XMR for that matter.
This will be true for the new nVidia 10XX series as well. When the rebates kikck in, I might buy the new stuff. Again, the thing here is that a top-notch 980ti can't hash ETH much faster than a 970, but if you apply the computing unts, the secondary coin hash rate really increases. My EVGA GTX 960 SC cards are now hashing at 10.5MH/s ETH / 250MH/s SC, where at the default setting (dcri = 30) they were hashing 10.5MH/s ETH / 120MH/s SC. The intensity setting is now "-dcri 75" on both my Win 7 980ti and my Linux 5X 960 rig. The rig gets 1.25GH/s hash for SC.
If the cards stay stable overnight, I'll experiment some more. And, after I sell my 280X cards and replace them with RX 470 cards, I'll worry about my nVidia equipment. Whenever I increase the intensity on the 280X cards, they overheat. Too much for one weekend. --scryptr
September 12, 2016, 12:33:54 AM
SOMETHING NOTABLE ABOUT NVIDIA CARDS--
Today I tried increasing the intensity for the secondary coin on my EVGA GTX CLASSIFIED 980ti card. I was able to more than double the hashrate for SiaCoin (SC) from 210MH/s to 540MH/s by increasing the "-dcri" intensity value from default to "-dcri 75". The card heated up from 77 deg F to 83 deg F, but the Ethereum (ETH) hashrate remained close to 22MH/s. The temperature of 83 deg F is the default target for this 980ti card.
The EVGA CLASSIFIED 980ti does not hash ETH much faster than a GTX 970, but it has greater compute power. This difference is shown in the large increase in hashrate for the secondary coin (SC) by just increasing the "dcri" intensity value. There is room for experimentation here. --scryptr.
you better sell this hungry monster and buy 1070 or rx480. You'll have 30 eth and 450sc easy with only 150-200 watts (depend on your downvolting and optimization skills) Today I tried increasing the intensity for the secondary coin on my EVGA GTX CLASSIFIED 980ti card. I was able to more than double the hashrate for SiaCoin (SC) from 210MH/s to 540MH/s by increasing the "-dcri" intensity value from default to "-dcri 75". The card heated up from 77 deg F to 83 deg F, but the Ethereum (ETH) hashrate remained close to 22MH/s. The temperature of 83 deg F is the default target for this 980ti card.
The EVGA CLASSIFIED 980ti does not hash ETH much faster than a GTX 970, but it has greater compute power. This difference is shown in the large increase in hashrate for the secondary coin (SC) by just increasing the "dcri" intensity value. There is room for experimentation here. --scryptr.
September 11, 2016, 11:44:41 PM
I just tried ETH+LBRY on Windows 7
372.54 driver for Pascal, GTX 1070
LBC is reporting 27MH/s
Anything I do for dcri has little effect on total mining profits. The rise in LBC hash rate decreases the ETH hash rates at a very well scaled rate.
It is more profitable for me to single mine LBRY
Is this normal?
372.54 driver for Pascal, GTX 1070
LBC is reporting 27MH/s
Anything I do for dcri has little effect on total mining profits. The rise in LBC hash rate decreases the ETH hash rates at a very well scaled rate.
It is more profitable for me to single mine LBRY
Is this normal?
September 11, 2016, 11:19:03 PM
SOMETHING NOTABLE ABOUT NVIDIA CARDS--
Today I tried increasing the intensity for the secondary coin on my EVGA GTX CLASSIFIED 980ti card. I was able to more than double the hashrate for SiaCoin (SC) from 210MH/s to 540MH/s by increasing the "-dcri" intensity value from default to "-dcri 75". The card heated up from 77 deg F to 83 deg F, but the Ethereum (ETH) hashrate remained close to 22MH/s. The temperature of 83 deg F is the default target for this 980ti card.
The EVGA CLASSIFIED 980ti does not hash ETH much faster than a GTX 970, but it has greater compute power. This difference is shown in the large increase in hashrate for the secondary coin (SC) by just increasing the "dcri" intensity value. There is room for experimentation here. --scryptr.
Today I tried increasing the intensity for the secondary coin on my EVGA GTX CLASSIFIED 980ti card. I was able to more than double the hashrate for SiaCoin (SC) from 210MH/s to 540MH/s by increasing the "-dcri" intensity value from default to "-dcri 75". The card heated up from 77 deg F to 83 deg F, but the Ethereum (ETH) hashrate remained close to 22MH/s. The temperature of 83 deg F is the default target for this 980ti card.
The EVGA CLASSIFIED 980ti does not hash ETH much faster than a GTX 970, but it has greater compute power. This difference is shown in the large increase in hashrate for the secondary coin (SC) by just increasing the "dcri" intensity value. There is room for experimentation here. --scryptr.
September 11, 2016, 09:57:44 PM
Is there a command in Claymores miner that allows you to restart the miner if a card drops below a certain hashrate?
I noticed that I was running a 29 mhs low power mod and after 1 hour GPU 1 dropped to 25. Restarted the miner and both GPUs are hashing at 29 MHs with no issues for 4 hours.
Thanks
I noticed that I was running a 29 mhs low power mod and after 1 hour GPU 1 dropped to 25. Restarted the miner and both GPUs are hashing at 29 MHs with no issues for 4 hours.
Thanks
September 11, 2016, 09:46:33 PM
So I got an H81 BTC PRO board, currently 2x 1060s, 480, 2x 470. Can only use 4 GPUs at a time, Win7 sees all, but 1 will be inactive. Does the 6 gpu mod for Win7 work with Polaris cards? I can switch the 1060s with more 470s if I need to.
Also, is undervolting supported on Linux currently? This is a huge feature since I can save ~150w+ for the rig going from 1150mv to ~960-1000mv, and get better boost clocks. Is EthOS easy to use CDM?
Also, is undervolting supported on Linux currently? This is a huge feature since I can save ~150w+ for the rig going from 1150mv to ~960-1000mv, and get better boost clocks. Is EthOS easy to use CDM?
I was not able to run 6 cards ( rx480 ) with windows 7 at all . 4 was running ok , 5 had some stability issue , 6 never worked ... Tried everything
Switched to win 10 , everything working fine . Installed WIN 10 ( disabled all updates , all automatic downloads ) , Intel Inf Utility , and rx 480 driver . With default settings all card hashing with 24.7 MH/s taking 1100 watts at wall ( 120 V ) , Then i "underclocked" my cards ( power limit , gpu speed - voltage , mem voltage ) , power usage at the wall become around 750Watts without pushing the default mem clock higher , with the same 24.7 Mh/s . Saved 350Watts per rig around 700watts ( with my 2 setup in the same frame) Used wattman for underclocking .Built a 12 cards ETH Miner , with 12 RX 480 / 2 asrock h97 anniversary Mobo , left a place in the middle for a additional micro itx board and 4 more cards Still try to find a stable mem overclocking settings .
http://beszeljukmeg.com/ETH/ETH01m.jpg
Still try to find a stable mem overclocking settings .http://beszeljukmeg.com/ETH/ETH01m.jpg
Nice job on this. Where did you get those rails for the structure?
September 11, 2016, 09:14:36 PM
So I got an H81 BTC PRO board, currently 2x 1060s, 480, 2x 470. Can only use 4 GPUs at a time, Win7 sees all, but 1 will be inactive. Does the 6 gpu mod for Win7 work with Polaris cards? I can switch the 1060s with more 470s if I need to.
Also, is undervolting supported on Linux currently? This is a huge feature since I can save ~150w+ for the rig going from 1150mv to ~960-1000mv, and get better boost clocks. Is EthOS easy to use CDM?
Also, is undervolting supported on Linux currently? This is a huge feature since I can save ~150w+ for the rig going from 1150mv to ~960-1000mv, and get better boost clocks. Is EthOS easy to use CDM?
I was not able to run 6 cards ( rx480 ) with windows 7 at all . 4 was running ok , 5 had some stability issue , 6 never worked ... Tried everything
Switched to win 10 , everything working fine . Installed WIN 10 ( disabled all updates , all automatic downloads ) , Intel Inf Utility , and rx 480 driver . With default settings all card hashing with 24.7 MH/s taking 1100 watts at wall ( 120 V ) , Then i "underclocked" my cards ( power limit , gpu speed - voltage , mem voltage ) , power usage at the wall become around 750Watts without pushing the default mem clock higher , with the same 24.7 Mh/s . Saved 350Watts per rig around 700watts ( with my 2 setup in the same frame) Used wattman for underclocking .Built a 12 cards ETH Miner , with 12 RX 480 / 2 asrock h97 anniversary Mobo , left a place in the middle for a additional micro itx board and 4 more cards Still try to find a stable mem overclocking settings .
Still try to find a stable mem overclocking settings .
is not possible to use more than 7 vga on the same system?
using an adapter 1 to 3 ?
NOT ALL PCIE-SPLITTERS ARE THE SAME--
I have a 1-to-4 PCIe splitter that worked with a GTX 550ti and a GT 9800+, with a GTX 750ti on the motherboard. The 3 cards mined scrypt, but were not very fast. The GT9800+ got 14kh/s.
I was never able to get the splitter to work with any cards more modern; the GTX 550ti was compute 2.1 and the GT 9800+ was compute 1.0. An American made splitter will likely cost more than a motherboard. The model I bought cost $75, a new one today runs $45. I tried with AMD cards also. with no success. Asian 1-to-3 and 1-to-4 splitters run $35 to $50. Good luck. --scryptr
September 11, 2016, 04:25:05 PM
So I got an H81 BTC PRO board, currently 2x 1060s, 480, 2x 470. Can only use 4 GPUs at a time, Win7 sees all, but 1 will be inactive. Does the 6 gpu mod for Win7 work with Polaris cards? I can switch the 1060s with more 470s if I need to.
Also, is undervolting supported on Linux currently? This is a huge feature since I can save ~150w+ for the rig going from 1150mv to ~960-1000mv, and get better boost clocks. Is EthOS easy to use CDM?
Also, is undervolting supported on Linux currently? This is a huge feature since I can save ~150w+ for the rig going from 1150mv to ~960-1000mv, and get better boost clocks. Is EthOS easy to use CDM?
I was not able to run 6 cards ( rx480 ) with windows 7 at all . 4 was running ok , 5 had some stability issue , 6 never worked ... Tried everything
Switched to win 10 , everything working fine . Installed WIN 10 ( disabled all updates , all automatic downloads ) , Intel Inf Utility , and rx 480 driver . With default settings all card hashing with 24.7 MH/s taking 1100 watts at wall ( 120 V ) , Then i "underclocked" my cards ( power limit , gpu speed - voltage , mem voltage ) , power usage at the wall become around 750Watts without pushing the default mem clock higher , with the same 24.7 Mh/s . Saved 350Watts per rig around 700watts ( with my 2 setup in the same frame) Used wattman for underclocking .Built a 12 cards ETH Miner , with 12 RX 480 / 2 asrock h97 anniversary Mobo , left a place in the middle for a additional micro itx board and 4 more cards Still try to find a stable mem overclocking settings .
Still try to find a stable mem overclocking settings .is not possible to use more than 7 vga on the same system?
using an adapter 1 to 3 ?
September 11, 2016, 04:19:14 PM
Hello,
Command -r -1 to restart my claymore bat is not working, my bat closed but the mew bat did not open.
How i fix it?
Thanks
Command -r -1 to restart my claymore bat is not working, my bat closed but the mew bat did not open.
How i fix it?
Thanks
September 11, 2016, 03:53:30 PM
Hey Clarmore didn't u see my post?
Regarding the etha 0 switch works on AMD but not on nvidia card..any explanation why? Thanks
Regarding the etha 0 switch works on AMD but not on nvidia card..any explanation why? Thanks
September 11, 2016, 02:59:34 PM
@Claymore
Do you have this one on your roadmap ? Maybe you have an information for target version ?
Thanks
Do you have this one on your roadmap ? Maybe you have an information for target version ?
Thanks
I think I will add it in 1-2 weeks.
September 11, 2016, 02:23:59 PM
Is there any way to make this miner run on Windows 10 32-bit system?
September 11, 2016, 12:34:42 PM
Third try.
@Claymore
Please update us on this one.
@Claymore
Please update us on this one.
I noticed that I miss some messages sometimes, sorry.
I'm not linux dev and reading/parsing some files to get temps looks strange for me and will take some time to implement and test it properly. Therefore I will not do it right now, though I will do it in future anyway because it seems there is no "normal" way now, again thanks to AMD devs, they don't care about such things as backward compatibility. Also, this way allows monitoring only, fans management won't work.
Right now I add new coin and I think it's more interesting feature than getting temps for Linux for 4xx cards.
No Problem, thats why i try it more often
I agree that this seems to be strange for windows users, but in linux this is quite normal.
The /sysfs/ folder mentioned above is a virtual file system from the Kernel (and there drivers) itself.
The above paths mentioned are not files, they are more like APIs where you can control the driver.
For example you can also control the FAN Speed.
Code:
echo 255 > /sys/class/drm/card1/device/hwmon/hwmon0/pwm1
will set the fanspeed of card1 to 100%
This realy is an important feature of your miner, and im pretty sure that bigger mining farms are not running windows
Ok, I will check it soon.
@Claymore
Do you have this one on your roadmap ? Maybe you have an information for target version ?
Thanks
September 11, 2016, 12:31:10 PM
Guys,
I'm facing some stability problems in both my rigs
RIG specs:
Asrock H81 Pro BTC
8 GB RAM + 16 GB Pagefile
6 x Sapphire RX 480 Nitro + OC 4GB
USB Powered Risers
2 x Corsair RM850X
Software:
Windows 10 x64
AMD drivers 16.9.1
Miner Settings:
Claymore v6.4
-dcri 25
-ethi 16
Stock settings:
GPU: 1305 MHz, 1150 mV
Mem: 1750 MHz, 1000 mV
Speed per card: 22.1 MHs ETC, 276 MHs SC
Power per card: ~130W (GPU-Z)
Underclock:
GPU: 1250 MHz, 1025 mV
Mem: 1850 MHz, Auto
Speed per card: 22 MHs ETC, 275 MHs SC
Power per card: ~105W (GPU-Z)
This is a 25W per card saving while keeping the same hashing speed. However, I have problems with the memory settings
Unfortunately, I cannot get the miner to stabilize with a higher memory speed, it crashes after a few minutes if I set it at 1900 MHz
Anyone using the Sapphire RX 480 Nitro 4GB? Are you undervolting? Care to share your settings? Driver version?
Thanks!
I'm facing some stability problems in both my rigs
RIG specs:
Asrock H81 Pro BTC
8 GB RAM + 16 GB Pagefile
6 x Sapphire RX 480 Nitro + OC 4GB
USB Powered Risers
2 x Corsair RM850X
Software:
Windows 10 x64
AMD drivers 16.9.1
Miner Settings:
Claymore v6.4
-dcri 25
-ethi 16
Stock settings:
GPU: 1305 MHz, 1150 mV
Mem: 1750 MHz, 1000 mV
Speed per card: 22.1 MHs ETC, 276 MHs SC
Power per card: ~130W (GPU-Z)
Underclock:
GPU: 1250 MHz, 1025 mV
Mem: 1850 MHz, Auto
Speed per card: 22 MHs ETC, 275 MHs SC
Power per card: ~105W (GPU-Z)
This is a 25W per card saving while keeping the same hashing speed. However, I have problems with the memory settings
Unfortunately, I cannot get the miner to stabilize with a higher memory speed, it crashes after a few minutes if I set it at 1900 MHz
Anyone using the Sapphire RX 480 Nitro 4GB? Are you undervolting? Care to share your settings? Driver version?
Thanks!
September 11, 2016, 12:15:16 PM
So I got an H81 BTC PRO board, currently 2x 1060s, 480, 2x 470. Can only use 4 GPUs at a time, Win7 sees all, but 1 will be inactive. Does the 6 gpu mod for Win7 work with Polaris cards? I can switch the 1060s with more 470s if I need to.
Also, is undervolting supported on Linux currently? This is a huge feature since I can save ~150w+ for the rig going from 1150mv to ~960-1000mv, and get better boost clocks. Is EthOS easy to use CDM?
Also, is undervolting supported on Linux currently? This is a huge feature since I can save ~150w+ for the rig going from 1150mv to ~960-1000mv, and get better boost clocks. Is EthOS easy to use CDM?
I was not able to run 6 cards ( rx480 ) with windows 7 at all . 4 was running ok , 5 had some stability issue , 6 never worked ... Tried everything
Switched to win 10 , everything working fine . Installed WIN 10 ( disabled all updates , all automatic downloads ) , Intel Inf Utility , and rx 480 driver . With default settings all card hashing with 24.7 MH/s taking 1100 watts at wall ( 120 V ) , Then i "underclocked" my cards ( power limit , gpu speed - voltage , mem voltage ) , power usage at the wall become around 750Watts without pushing the default mem clock higher , with the same 24.7 Mh/s . Saved 350Watts per rig around 700watts ( with my 2 setup in the same frame) Used wattman for underclocking .Built a 12 cards ETH Miner , with 12 RX 480 / 2 asrock h97 anniversary Mobo , left a place in the middle for a additional micro itx board and 4 more cards Still try to find a stable mem overclocking settings .
Still try to find a stable mem overclocking settings . September 11, 2016, 11:00:51 AM
rx 480 under linux support is dropped in 6.4 and 7.0 versions.
I'm getting:
Now I'm using 6.3 version.
I'm getting:
Code:
Sep 11 17:42:51 rig3 kernel: [ 6802.533701] amdgpu 0000:05:00.0: GPU fault detected: 147 0x01f80801
Sep 11 17:42:51 rig3 kernel: [ 6802.534129] amdgpu 0000:05:00.0: VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x0810483B
Sep 11 17:42:51 rig3 kernel: [ 6802.534567] amdgpu 0000:05:00.0: VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x04084001
Sep 11 17:42:51 rig3 kernel: [ 6802.534997] VM fault (0x01, vmid 2) at page 135284795, read from 'TC7' (0x54433700) (132)
Sep 11 17:42:51 rig3 kernel: [ 6802.534129] amdgpu 0000:05:00.0: VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x0810483B
Sep 11 17:42:51 rig3 kernel: [ 6802.534567] amdgpu 0000:05:00.0: VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x04084001
Sep 11 17:42:51 rig3 kernel: [ 6802.534997] VM fault (0x01, vmid 2) at page 135284795, read from 'TC7' (0x54433700) (132)
Now I'm using 6.3 version.
Read OP or Readme, "COMMAND LINE OPTIONS" section:
-etha Ethereum algorithm mode. 0 - optimized for fast cards, 1 - optimized for slow cards, 2 - for gpu-pro Linux drivers. -1 - autodetect (default, automatically selects between 0 and 1).
You can also set this option for every card individually, for example "-etha 0,1,0".
PS. In next update miner will set "-etha 2" automatically if Linux+4xx detected.
September 11, 2016, 09:50:21 AM
rx 480 under linux support is dropped in 6.4 and 7.0 versions.
I'm getting:
Now I'm using 6.3 version.
I'm getting:
Code:
Sep 11 17:42:51 rig3 kernel: [ 6802.533701] amdgpu 0000:05:00.0: GPU fault detected: 147 0x01f80801
Sep 11 17:42:51 rig3 kernel: [ 6802.534129] amdgpu 0000:05:00.0: VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x0810483B
Sep 11 17:42:51 rig3 kernel: [ 6802.534567] amdgpu 0000:05:00.0: VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x04084001
Sep 11 17:42:51 rig3 kernel: [ 6802.534997] VM fault (0x01, vmid 2) at page 135284795, read from 'TC7' (0x54433700) (132)
Sep 11 17:42:51 rig3 kernel: [ 6802.534129] amdgpu 0000:05:00.0: VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x0810483B
Sep 11 17:42:51 rig3 kernel: [ 6802.534567] amdgpu 0000:05:00.0: VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x04084001
Sep 11 17:42:51 rig3 kernel: [ 6802.534997] VM fault (0x01, vmid 2) at page 135284795, read from 'TC7' (0x54433700) (132)
Now I'm using 6.3 version.
September 11, 2016, 09:12:46 AM
Claymore, will you be looking at zcash - equihash to see if you can put out a GPU miner?
I second that. I think it is very important for the smaller miners to have access to a good miner in the early stage.
Right now I don't have any plans to create that miner.
Jump to: