Morph42
Jr. Member
Offline
Activity: 60
Merit: 2
|
|
April 21, 2020, 09:54:50 PM |
|
New to mining, but have my rig setup and it's currently mining away. 12 x MSI/AMD Radeon RX 570 (8GB) I'm reading through a number of posts and I'm seeing that my hashrate is really low per card compared to what others are mentioning. ETH - Total Speed: 122.407 Mh/s ETH: GPU0 10.192 Mh/s, GPU1 10.197 Mh/s, GPU2 10.227 Mh/s, GPU3 10.205 Mh/s, GPU4 10.205 Mh/s, GPU5 10.201 Mh/s, GPU6 10.195 Mh/s, GPU7 10.191 Mh/s, GPU8 10.219 Mh/s, GPU9 10.202 Mh/s, GPU10 10.183 Mh/s, GPU11 10.191 Mh/s
Here's claymore's report on my system GPU0 t=59C fan=91% P=99W, GPU1 t=56C fan=88% P=98W, GPU2 t=56C fan=91% P=97W, GPU3 t=51C fan=90% P=96W, GPU4 t=56C fan=90% P=98W, GPU5 t=57C fan=91% P=99W, GPU6 t=52C fan=92% P=101W, GPU7 t=54C fan=89% P=96W, GPU8 t=55C fan=90% P=98W, GPU9 t=58C fan=91% P=87W, GPU10 t=54C fan=91% P=90W, GPU11 t=62C fan=92% P=102W Total GPUs power consumption: 1161 Watts
I've downloaded the most recent version of the AMD drivers. When I initially ran claymore I did have to run the command "-drivers uninstall" as it was telling me a problem with that. Did that and that error has gone away during startup of mining. I'm thinking the next step is to overclock? But even with my gaming machine I'm touchy about that due to the damage it could inflict. Is there an app that will test where I can safely set it up to? Have you switched to Compute mode instead of Graphics mode on the amd drivers? If you haven't do it for each gpu. Don't worry about overclocking. Just don't put the memory over 2250. so set powerlimit to -10, cclock 1150-1200 mclock 2100-2250. If you go to 2250 and it starts getting invalid shares, back the memory down to 2225, 2200, 2175 ect, until it is stable. Also at one of the -straps to ur command line to apply good bios timings to maximize cards potential. I hadn't actually switched to compute mode. So I'll add [-y 1] but will do that for all cards automatically or do I need to list 0 to b? I'll add the other recommendations -powlim -10 -cclock 1200 -mclock 2250 and I'll try [-strap 1] MOdified mclock to 2200 after a couple lockups. Slightly better hashrate, but still not what I was expecting. ETH - Total Speed: 142.059 Mh/s ETH: GPU0 11.837 Mh/s, GPU1 11.837 Mh/s, GPU2 11.826 Mh/s, GPU3 11.844 Mh/s, GPU4 11.845 Mh/s, GPU5 11.829 Mh/s, GPU6 11.830 Mh/s, GPU7 11.844 Mh/s, GPU8 11.854 Mh/s, GPU9 11.826 Mh/s, GPU10 11.849 Mh/s, GPU11 11.838 Mh/s
GPU0 t=52C fan=93% P=70W, GPU1 t=50C fan=92% P=69W, GPU2 t=50C fan=94% P=68W, GPU3 t=46C fan=93% P=69W, GPU4 t=50C fan=93% P=70W, GPU5 t=50C fan=94% P=70W, GPU6 t=47C fan=94% P=71W, GPU7 t=48C fan=92% P=68W, GPU8 t=51C fan=94% P=70W, GPU9 t=51C fan=94% P=67W, GPU10 t=49C fan=94% P=67W, GPU11 t=54C fan=95% P=71W Total GPUs power consumption: 830 Watts
To help further with your recommendations I figure I should add my BAT file. setx GPU_FORCE_64BIT_PTR 0 setx GPU_MAX_HEAP_SIZE 100 setx GPU_USE_SYNC_OBJECTS 1 setx GPU_MAX_ALLOC_PERCENT 100
EthDcrMiner64.exe -epool ssl://us2.ethermine.org:5555 -ewal [ewallet#] -eworker bigdog1 -dbg -1 -mport 3333 -mpsw [secret#] -etha 0 -tt 80 -tstop 90 -eres 0 -rxboost 1 -y 1 -powlim -10 -cclock 1200 -mclock 2200 -strap 1
pause
|
|
|
|
br2459
|
|
April 21, 2020, 11:01:09 PM |
|
New to mining, but have my rig setup and it's currently mining away. 12 x MSI/AMD Radeon RX 570 (8GB) I'm reading through a number of posts and I'm seeing that my hashrate is really low per card compared to what others are mentioning. ETH - Total Speed: 122.407 Mh/s ETH: GPU0 10.192 Mh/s, GPU1 10.197 Mh/s, GPU2 10.227 Mh/s, GPU3 10.205 Mh/s, GPU4 10.205 Mh/s, GPU5 10.201 Mh/s, GPU6 10.195 Mh/s, GPU7 10.191 Mh/s, GPU8 10.219 Mh/s, GPU9 10.202 Mh/s, GPU10 10.183 Mh/s, GPU11 10.191 Mh/s
Here's claymore's report on my system GPU0 t=59C fan=91% P=99W, GPU1 t=56C fan=88% P=98W, GPU2 t=56C fan=91% P=97W, GPU3 t=51C fan=90% P=96W, GPU4 t=56C fan=90% P=98W, GPU5 t=57C fan=91% P=99W, GPU6 t=52C fan=92% P=101W, GPU7 t=54C fan=89% P=96W, GPU8 t=55C fan=90% P=98W, GPU9 t=58C fan=91% P=87W, GPU10 t=54C fan=91% P=90W, GPU11 t=62C fan=92% P=102W Total GPUs power consumption: 1161 Watts
I've downloaded the most recent version of the AMD drivers. When I initially ran claymore I did have to run the command "-drivers uninstall" as it was telling me a problem with that. Did that and that error has gone away during startup of mining. I'm thinking the next step is to overclock? But even with my gaming machine I'm touchy about that due to the damage it could inflict. Is there an app that will test where I can safely set it up to? Have you switched to Compute mode instead of Graphics mode on the amd drivers? If you haven't do it for each gpu. Don't worry about overclocking. Just don't put the memory over 2250. so set powerlimit to -10, cclock 1150-1200 mclock 2100-2250. If you go to 2250 and it starts getting invalid shares, back the memory down to 2225, 2200, 2175 ect, until it is stable. Also at one of the -straps to ur command line to apply good bios timings to maximize cards potential. I hadn't actually switched to compute mode. So I'll add [-y 1] but will do that for all cards automatically or do I need to list 0 to b? I'll add the other recommendations -powlim -10 -cclock 1200 -mclock 2250 and I'll try [-strap 1] MOdified mclock to 2200 after a couple lockups. Slightly better hashrate, but still not what I was expecting. ETH - Total Speed: 142.059 Mh/s ETH: GPU0 11.837 Mh/s, GPU1 11.837 Mh/s, GPU2 11.826 Mh/s, GPU3 11.844 Mh/s, GPU4 11.845 Mh/s, GPU5 11.829 Mh/s, GPU6 11.830 Mh/s, GPU7 11.844 Mh/s, GPU8 11.854 Mh/s, GPU9 11.826 Mh/s, GPU10 11.849 Mh/s, GPU11 11.838 Mh/s
GPU0 t=52C fan=93% P=70W, GPU1 t=50C fan=92% P=69W, GPU2 t=50C fan=94% P=68W, GPU3 t=46C fan=93% P=69W, GPU4 t=50C fan=93% P=70W, GPU5 t=50C fan=94% P=70W, GPU6 t=47C fan=94% P=71W, GPU7 t=48C fan=92% P=68W, GPU8 t=51C fan=94% P=70W, GPU9 t=51C fan=94% P=67W, GPU10 t=49C fan=94% P=67W, GPU11 t=54C fan=95% P=71W Total GPUs power consumption: 830 Watts
To help further with your recommendations I figure I should add my BAT file. setx GPU_FORCE_64BIT_PTR 0 setx GPU_MAX_HEAP_SIZE 100 setx GPU_USE_SYNC_OBJECTS 1 setx GPU_MAX_ALLOC_PERCENT 100
EthDcrMiner64.exe -epool ssl://us2.ethermine.org:5555 -ewal [ewallet#] -eworker bigdog1 -dbg -1 -mport 3333 -mpsw [secret#] -etha 0 -tt 80 -tstop 90 -eres 0 -rxboost 1 -y 1 -powlim -10 -cclock 1200 -mclock 2200 -strap 1
pause
It doesn't appear that compute mode was enabled successfully. In order to enable compute mode you need to have ran the miner/bat file as admin. It's not just right click and run as administrator. You have to open the command line as administrator, path to the folder (Cd \folder\miner location\) run the .bat file from the command windows. After it runs for about a min, reboot your pc. Then you won't need to run it as administrator the next time. You should see hashrates in the high 28MH/s when done correctly.
|
|
|
|
RAKOT
Jr. Member
Offline
Activity: 152
Merit: 8
|
|
April 22, 2020, 05:41:26 AM |
|
I am not sure if Claymore reads this, but are you working on enabling somehow offloading the Dag after it reaches 4GB? Maybe using a rig with some 8GB cards and some 4GB cards and averaging the Dag between the cards. Would make your miner even more popular.
Would you sit down for hours on programming and bugfixing if you'r $ multimillioner ? So forget about Claymore . He drinks tekila on the beach.
|
|
|
|
Morph42
Jr. Member
Offline
Activity: 60
Merit: 2
|
|
April 22, 2020, 04:29:17 PM |
|
New to mining, but have my rig setup and it's currently mining away. 12 x MSI/AMD Radeon RX 570 (8GB) I'm reading through a number of posts and I'm seeing that my hashrate is really low per card compared to what others are mentioning. ETH - Total Speed: 122.407 Mh/s ETH: GPU0 10.192 Mh/s, GPU1 10.197 Mh/s, GPU2 10.227 Mh/s, GPU3 10.205 Mh/s, GPU4 10.205 Mh/s, GPU5 10.201 Mh/s, GPU6 10.195 Mh/s, GPU7 10.191 Mh/s, GPU8 10.219 Mh/s, GPU9 10.202 Mh/s, GPU10 10.183 Mh/s, GPU11 10.191 Mh/s
Here's claymore's report on my system GPU0 t=59C fan=91% P=99W, GPU1 t=56C fan=88% P=98W, GPU2 t=56C fan=91% P=97W, GPU3 t=51C fan=90% P=96W, GPU4 t=56C fan=90% P=98W, GPU5 t=57C fan=91% P=99W, GPU6 t=52C fan=92% P=101W, GPU7 t=54C fan=89% P=96W, GPU8 t=55C fan=90% P=98W, GPU9 t=58C fan=91% P=87W, GPU10 t=54C fan=91% P=90W, GPU11 t=62C fan=92% P=102W Total GPUs power consumption: 1161 Watts
I've downloaded the most recent version of the AMD drivers. When I initially ran claymore I did have to run the command "-drivers uninstall" as it was telling me a problem with that. Did that and that error has gone away during startup of mining. I'm thinking the next step is to overclock? But even with my gaming machine I'm touchy about that due to the damage it could inflict. Is there an app that will test where I can safely set it up to? Have you switched to Compute mode instead of Graphics mode on the amd drivers? If you haven't do it for each gpu. Don't worry about overclocking. Just don't put the memory over 2250. so set powerlimit to -10, cclock 1150-1200 mclock 2100-2250. If you go to 2250 and it starts getting invalid shares, back the memory down to 2225, 2200, 2175 ect, until it is stable. Also at one of the -straps to ur command line to apply good bios timings to maximize cards potential. I hadn't actually switched to compute mode. So I'll add [-y 1] but will do that for all cards automatically or do I need to list 0 to b? I'll add the other recommendations -powlim -10 -cclock 1200 -mclock 2250 and I'll try [-strap 1] MOdified mclock to 2200 after a couple lockups. Slightly better hashrate, but still not what I was expecting. ETH - Total Speed: 142.059 Mh/s ETH: GPU0 11.837 Mh/s, GPU1 11.837 Mh/s, GPU2 11.826 Mh/s, GPU3 11.844 Mh/s, GPU4 11.845 Mh/s, GPU5 11.829 Mh/s, GPU6 11.830 Mh/s, GPU7 11.844 Mh/s, GPU8 11.854 Mh/s, GPU9 11.826 Mh/s, GPU10 11.849 Mh/s, GPU11 11.838 Mh/s
GPU0 t=52C fan=93% P=70W, GPU1 t=50C fan=92% P=69W, GPU2 t=50C fan=94% P=68W, GPU3 t=46C fan=93% P=69W, GPU4 t=50C fan=93% P=70W, GPU5 t=50C fan=94% P=70W, GPU6 t=47C fan=94% P=71W, GPU7 t=48C fan=92% P=68W, GPU8 t=51C fan=94% P=70W, GPU9 t=51C fan=94% P=67W, GPU10 t=49C fan=94% P=67W, GPU11 t=54C fan=95% P=71W Total GPUs power consumption: 830 Watts
To help further with your recommendations I figure I should add my BAT file. setx GPU_FORCE_64BIT_PTR 0 setx GPU_MAX_HEAP_SIZE 100 setx GPU_USE_SYNC_OBJECTS 1 setx GPU_MAX_ALLOC_PERCENT 100
EthDcrMiner64.exe -epool ssl://us2.ethermine.org:5555 -ewal [ewallet#] -eworker bigdog1 -dbg -1 -mport 3333 -mpsw [secret#] -etha 0 -tt 80 -tstop 90 -eres 0 -rxboost 1 -y 1 -powlim -10 -cclock 1200 -mclock 2200 -strap 1
pause
It doesn't appear that compute mode was enabled successfully. In order to enable compute mode you need to have ran the miner/bat file as admin. It's not just right click and run as administrator. You have to open the command line as administrator, path to the folder (Cd \folder\miner location\) run the .bat file from the command windows. After it runs for about a min, reboot your pc. Then you won't need to run it as administrator the next time. You should see hashrates in the high 28MH/s when done correctly. Thanks for that. That was the apparent issue. After you mentioned it I did some digging and learned I needed to open the AMD software and the setting was still set to GRAPHICS and not COMPUTE. So I fixed that. Then I went back and re-ran my bat file as a proper admin and it's now hashing better. ETH - Total Speed: 279.807 Mh/s, Total Shares: 175, Rejected: 0, Time: 00:44 ETH: GPU0 23.322 Mh/s, GPU1 23.318 Mh/s, GPU2 23.321 Mh/s, GPU3 23.310 Mh/s, GPU4 23.318 Mh/s, GPU5 23.308 Mh/s, GPU6 23.323 Mh/s, GPU7 23.314 Mh/s, GPU8 23.320 Mh/s, GPU9 23.311 Mh/s, GPU10 23.323 Mh/s, GPU11 23.319 Mh/s
I had to remove the powlim, cclock, mclock and strap settings as they kept crashing Windows. I guess the question I have at this point is, is this the highest hashrate I should see out of my cards? Should I be expecting more? After removing the mclock and cclock settings I have played around with the MSI Afterburner app settings. I don't know if they affect Claymore the same or not. I seem to get a lot less errors out of it at least. Currently the afterburner settings are; power Limit +30 temp limit 90 (does this automatically with the increase in power) core clock 1360 mem clock 1904 fan speed 100
|
|
|
|
Morph42
Jr. Member
Offline
Activity: 60
Merit: 2
|
|
April 22, 2020, 04:47:02 PM |
|
Now that I'm getting a better/faster hashrate I'm watching the increase of what my rig is doing. But what I don't understand is the difference in what I see regarding my shares on my screen vs. the Ethermine dashboard.
Currently my screen shows 465 shares On the Ethermine Dashboard it shows 243
I do understand that the dev will mine every so often. But 50% of that doesn't seem right, does it? No. So trying to understand why the discrepancy. The Ethermind dashboard has indicated that it checked in within the past 10min so I would think it should show proper numbers.
|
|
|
|
br2459
|
|
April 22, 2020, 09:13:33 PM Last edit: April 22, 2020, 09:28:19 PM by br2459 |
|
Now that I'm getting a better/faster hashrate I'm watching the increase of what my rig is doing. But what I don't understand is the difference in what I see regarding my shares on my screen vs. the Ethermine dashboard.
Currently my screen shows 465 shares On the Ethermine Dashboard it shows 243
I do understand that the dev will mine every so often. But 50% of that doesn't seem right, does it? No. So trying to understand why the discrepancy. The Ethermind dashboard has indicated that it checked in within the past 10min so I would think it should show proper numbers.
From my understanding the Ethermine dashboard is only for shares is only 1 hour. I don't know if that is the top of every hour or how it calculates that. It also depends on what shares are "Stale" when they reach the pool. That will take away from your total shares ratio. My dashboard has only 4 stale shares in the last hour and it is less than 0% of total shares. I have too many clients to try and figure out what CM is reporting to compare to what the pool is stating. Edit: I just realized why you are seeing the discrepancy. It boils down to the share stats per hour on Ethermine. CM is going to show you the shares based on how long it has been running for with out crashing/restarting. Eithermine is just the last hour. My status changed a bit ago and I only have 2 stale shares in the last hour now. So no need to worry on the shares.
|
|
|
|
hustleman
Jr. Member
Offline
Activity: 155
Merit: 6
|
|
April 22, 2020, 09:30:42 PM |
|
New to mining, but have my rig setup and it's currently mining away. 12 x MSI/AMD Radeon RX 570 (8GB) I'm reading through a number of posts and I'm seeing that my hashrate is really low per card compared to what others are mentioning. ETH - Total Speed: 122.407 Mh/s ETH: GPU0 10.192 Mh/s, GPU1 10.197 Mh/s, GPU2 10.227 Mh/s, GPU3 10.205 Mh/s, GPU4 10.205 Mh/s, GPU5 10.201 Mh/s, GPU6 10.195 Mh/s, GPU7 10.191 Mh/s, GPU8 10.219 Mh/s, GPU9 10.202 Mh/s, GPU10 10.183 Mh/s, GPU11 10.191 Mh/s
Here's claymore's report on my system GPU0 t=59C fan=91% P=99W, GPU1 t=56C fan=88% P=98W, GPU2 t=56C fan=91% P=97W, GPU3 t=51C fan=90% P=96W, GPU4 t=56C fan=90% P=98W, GPU5 t=57C fan=91% P=99W, GPU6 t=52C fan=92% P=101W, GPU7 t=54C fan=89% P=96W, GPU8 t=55C fan=90% P=98W, GPU9 t=58C fan=91% P=87W, GPU10 t=54C fan=91% P=90W, GPU11 t=62C fan=92% P=102W Total GPUs power consumption: 1161 Watts
I've downloaded the most recent version of the AMD drivers. When I initially ran claymore I did have to run the command "-drivers uninstall" as it was telling me a problem with that. Did that and that error has gone away during startup of mining. I'm thinking the next step is to overclock? But even with my gaming machine I'm touchy about that due to the damage it could inflict. Is there an app that will test where I can safely set it up to? Have you switched to Compute mode instead of Graphics mode on the amd drivers? If you haven't do it for each gpu. Don't worry about overclocking. Just don't put the memory over 2250. so set powerlimit to -10, cclock 1150-1200 mclock 2100-2250. If you go to 2250 and it starts getting invalid shares, back the memory down to 2225, 2200, 2175 ect, until it is stable. Also at one of the -straps to ur command line to apply good bios timings to maximize cards potential. I hadn't actually switched to compute mode. So I'll add [-y 1] but will do that for all cards automatically or do I need to list 0 to b? I'll add the other recommendations -powlim -10 -cclock 1200 -mclock 2250 and I'll try [-strap 1] MOdified mclock to 2200 after a couple lockups. Slightly better hashrate, but still not what I was expecting. ETH - Total Speed: 142.059 Mh/s ETH: GPU0 11.837 Mh/s, GPU1 11.837 Mh/s, GPU2 11.826 Mh/s, GPU3 11.844 Mh/s, GPU4 11.845 Mh/s, GPU5 11.829 Mh/s, GPU6 11.830 Mh/s, GPU7 11.844 Mh/s, GPU8 11.854 Mh/s, GPU9 11.826 Mh/s, GPU10 11.849 Mh/s, GPU11 11.838 Mh/s
GPU0 t=52C fan=93% P=70W, GPU1 t=50C fan=92% P=69W, GPU2 t=50C fan=94% P=68W, GPU3 t=46C fan=93% P=69W, GPU4 t=50C fan=93% P=70W, GPU5 t=50C fan=94% P=70W, GPU6 t=47C fan=94% P=71W, GPU7 t=48C fan=92% P=68W, GPU8 t=51C fan=94% P=70W, GPU9 t=51C fan=94% P=67W, GPU10 t=49C fan=94% P=67W, GPU11 t=54C fan=95% P=71W Total GPUs power consumption: 830 Watts
To help further with your recommendations I figure I should add my BAT file. setx GPU_FORCE_64BIT_PTR 0 setx GPU_MAX_HEAP_SIZE 100 setx GPU_USE_SYNC_OBJECTS 1 setx GPU_MAX_ALLOC_PERCENT 100
EthDcrMiner64.exe -epool ssl://us2.ethermine.org:5555 -ewal [ewallet#] -eworker bigdog1 -dbg -1 -mport 3333 -mpsw [secret#] -etha 0 -tt 80 -tstop 90 -eres 0 -rxboost 1 -y 1 -powlim -10 -cclock 1200 -mclock 2200 -strap 1
pause
It doesn't appear that compute mode was enabled successfully. In order to enable compute mode you need to have ran the miner/bat file as admin. It's not just right click and run as administrator. You have to open the command line as administrator, path to the folder (Cd \folder\miner location\) run the .bat file from the command windows. After it runs for about a min, reboot your pc. Then you won't need to run it as administrator the next time. You should see hashrates in the high 28MH/s when done correctly. Thanks for that. That was the apparent issue. After you mentioned it I did some digging and learned I needed to open the AMD software and the setting was still set to GRAPHICS and not COMPUTE. So I fixed that. Then I went back and re-ran my bat file as a proper admin and it's now hashing better. ETH - Total Speed: 279.807 Mh/s, Total Shares: 175, Rejected: 0, Time: 00:44 ETH: GPU0 23.322 Mh/s, GPU1 23.318 Mh/s, GPU2 23.321 Mh/s, GPU3 23.310 Mh/s, GPU4 23.318 Mh/s, GPU5 23.308 Mh/s, GPU6 23.323 Mh/s, GPU7 23.314 Mh/s, GPU8 23.320 Mh/s, GPU9 23.311 Mh/s, GPU10 23.323 Mh/s, GPU11 23.319 Mh/s
I had to remove the powlim, cclock, mclock and strap settings as they kept crashing Windows. I guess the question I have at this point is, is this the highest hashrate I should see out of my cards? Should I be expecting more? After removing the mclock and cclock settings I have played around with the MSI Afterburner app settings. I don't know if they affect Claymore the same or not. I seem to get a lot less errors out of it at least. Currently the afterburner settings are; power Limit +30 temp limit 90 (does this automatically with the increase in power) core clock 1360 mem clock 1904 fan speed 100 Now that you have it in compute mode that will help in your efficiency. no just do the following settings in bat.file and don't even use MSI afterburner. -10 powlim cclock 1200 cvddc 820 mclock 2150 mvddc 900. Give rxboost 1 and -strap 1 another try. If that crashes the system try just -strap 1 without rxboost to see if it is stable. All these cards should be in the 29-31mhs range.
|
|
|
|
wishxy
|
|
April 23, 2020, 03:17:05 AM |
|
Actual ETH hashrate felt down 10-15% in the last two days. Is this a common cause?
|
|
|
|
br2459
|
|
April 23, 2020, 04:10:20 AM |
|
Actual ETH hashrate felt down 10-15% in the last two days. Is this a common cause?
Are you talking about network hashrate or your hashrate on your cards over the last few days?
|
|
|
|
adaseb
Legendary
Offline
Activity: 3878
Merit: 1733
|
|
April 23, 2020, 06:02:27 AM |
|
Well according to, https://etherscan.io/chart/hashrateThe hashrate is completely constant so most likely he has some 4GB GPUs and he is getting slower speeds with them due to the DAG limit being hit on the 4GB cards. Like it was discussed in this thread many times, nothing to do but downgrade to Windows 7 or Switch to Linux or sell the 4GB and buy a 8GB instead. However most likely due to the low profitability nobody will want to go thru the hassle of doing that.
|
|
|
|
RAKOT
Jr. Member
Offline
Activity: 152
Merit: 8
|
|
April 23, 2020, 06:58:25 AM Last edit: April 23, 2020, 09:10:02 AM by RAKOT |
|
Well according to, https://etherscan.io/chart/hashrateThe hashrate is completely constant so most likely he has some 4GB GPUs and he is getting slower speeds with them due to the DAG limit being hit on the 4GB cards. Like it was discussed in this thread many times, nothing to do but downgrade to Windows 7 or Switch to Linux or sell the 4GB and buy a 8GB instead. However most likely due to the low profitability nobody will want to go thru the hassle of doing that. Have a couple of dozen 4Gb's , hashing at the same speed as before. So its about something else. And agree , selling 4Gbs for buying 8Gbs is nothing but pain and suffer ( done it few times not long ago ). We will see tons of 4Gbs very soon on second hand market. BR.
|
|
|
|
br2459
|
|
April 23, 2020, 02:24:15 PM |
|
Well according to, https://etherscan.io/chart/hashrateThe hashrate is completely constant so most likely he has some 4GB GPUs and he is getting slower speeds with them due to the DAG limit being hit on the 4GB cards. Like it was discussed in this thread many times, nothing to do but downgrade to Windows 7 or Switch to Linux or sell the 4GB and buy a 8GB instead. However most likely due to the low profitability nobody will want to go thru the hassle of doing that. Have a couple of dozen 4Gb's , hashing at the same speed as before. So its about something else. And agree , selling 4Gbs for buying 8Gbs is nothing but pain and suffer ( done it few times not long ago ). We will see tons of 4Gbs very soon on second hand market. BR. I am happy I made the decision to only buy 8GB cards when I started. It has paid off wisely. I had a few friends that just bought what they could get and then had to sell when the DAG started to swell.
|
|
|
|
Morph42
Jr. Member
Offline
Activity: 60
Merit: 2
|
|
April 23, 2020, 04:03:49 PM |
|
New to mining, but have my rig setup and it's currently mining away. 12 x MSI/AMD Radeon RX 570 (8GB) I'm reading through a number of posts and I'm seeing that my hashrate is really low per card compared to what others are mentioning. ETH - Total Speed: 122.407 Mh/s ETH: GPU0 10.192 Mh/s, GPU1 10.197 Mh/s, GPU2 10.227 Mh/s, GPU3 10.205 Mh/s, GPU4 10.205 Mh/s, GPU5 10.201 Mh/s, GPU6 10.195 Mh/s, GPU7 10.191 Mh/s, GPU8 10.219 Mh/s, GPU9 10.202 Mh/s, GPU10 10.183 Mh/s, GPU11 10.191 Mh/s
Here's claymore's report on my system GPU0 t=59C fan=91% P=99W, GPU1 t=56C fan=88% P=98W, GPU2 t=56C fan=91% P=97W, GPU3 t=51C fan=90% P=96W, GPU4 t=56C fan=90% P=98W, GPU5 t=57C fan=91% P=99W, GPU6 t=52C fan=92% P=101W, GPU7 t=54C fan=89% P=96W, GPU8 t=55C fan=90% P=98W, GPU9 t=58C fan=91% P=87W, GPU10 t=54C fan=91% P=90W, GPU11 t=62C fan=92% P=102W Total GPUs power consumption: 1161 Watts
I've downloaded the most recent version of the AMD drivers. When I initially ran claymore I did have to run the command "-drivers uninstall" as it was telling me a problem with that. Did that and that error has gone away during startup of mining. I'm thinking the next step is to overclock? But even with my gaming machine I'm touchy about that due to the damage it could inflict. Is there an app that will test where I can safely set it up to? Have you switched to Compute mode instead of Graphics mode on the amd drivers? If you haven't do it for each gpu. Don't worry about overclocking. Just don't put the memory over 2250. so set powerlimit to -10, cclock 1150-1200 mclock 2100-2250. If you go to 2250 and it starts getting invalid shares, back the memory down to 2225, 2200, 2175 ect, until it is stable. Also at one of the -straps to ur command line to apply good bios timings to maximize cards potential. I hadn't actually switched to compute mode. So I'll add [-y 1] but will do that for all cards automatically or do I need to list 0 to b? I'll add the other recommendations -powlim -10 -cclock 1200 -mclock 2250 and I'll try [-strap 1] MOdified mclock to 2200 after a couple lockups. Slightly better hashrate, but still not what I was expecting. ETH - Total Speed: 142.059 Mh/s ETH: GPU0 11.837 Mh/s, GPU1 11.837 Mh/s, GPU2 11.826 Mh/s, GPU3 11.844 Mh/s, GPU4 11.845 Mh/s, GPU5 11.829 Mh/s, GPU6 11.830 Mh/s, GPU7 11.844 Mh/s, GPU8 11.854 Mh/s, GPU9 11.826 Mh/s, GPU10 11.849 Mh/s, GPU11 11.838 Mh/s
GPU0 t=52C fan=93% P=70W, GPU1 t=50C fan=92% P=69W, GPU2 t=50C fan=94% P=68W, GPU3 t=46C fan=93% P=69W, GPU4 t=50C fan=93% P=70W, GPU5 t=50C fan=94% P=70W, GPU6 t=47C fan=94% P=71W, GPU7 t=48C fan=92% P=68W, GPU8 t=51C fan=94% P=70W, GPU9 t=51C fan=94% P=67W, GPU10 t=49C fan=94% P=67W, GPU11 t=54C fan=95% P=71W Total GPUs power consumption: 830 Watts
To help further with your recommendations I figure I should add my BAT file. setx GPU_FORCE_64BIT_PTR 0 setx GPU_MAX_HEAP_SIZE 100 setx GPU_USE_SYNC_OBJECTS 1 setx GPU_MAX_ALLOC_PERCENT 100
EthDcrMiner64.exe -epool ssl://us2.ethermine.org:5555 -ewal [ewallet#] -eworker bigdog1 -dbg -1 -mport 3333 -mpsw [secret#] -etha 0 -tt 80 -tstop 90 -eres 0 -rxboost 1 -y 1 -powlim -10 -cclock 1200 -mclock 2200 -strap 1
pause
It doesn't appear that compute mode was enabled successfully. In order to enable compute mode you need to have ran the miner/bat file as admin. It's not just right click and run as administrator. You have to open the command line as administrator, path to the folder (Cd \folder\miner location\) run the .bat file from the command windows. After it runs for about a min, reboot your pc. Then you won't need to run it as administrator the next time. You should see hashrates in the high 28MH/s when done correctly. Thanks for that. That was the apparent issue. After you mentioned it I did some digging and learned I needed to open the AMD software and the setting was still set to GRAPHICS and not COMPUTE. So I fixed that. Then I went back and re-ran my bat file as a proper admin and it's now hashing better. ETH - Total Speed: 279.807 Mh/s, Total Shares: 175, Rejected: 0, Time: 00:44 ETH: GPU0 23.322 Mh/s, GPU1 23.318 Mh/s, GPU2 23.321 Mh/s, GPU3 23.310 Mh/s, GPU4 23.318 Mh/s, GPU5 23.308 Mh/s, GPU6 23.323 Mh/s, GPU7 23.314 Mh/s, GPU8 23.320 Mh/s, GPU9 23.311 Mh/s, GPU10 23.323 Mh/s, GPU11 23.319 Mh/s
I had to remove the powlim, cclock, mclock and strap settings as they kept crashing Windows. I guess the question I have at this point is, is this the highest hashrate I should see out of my cards? Should I be expecting more? After removing the mclock and cclock settings I have played around with the MSI Afterburner app settings. I don't know if they affect Claymore the same or not. I seem to get a lot less errors out of it at least. Currently the afterburner settings are; power Limit +30 temp limit 90 (does this automatically with the increase in power) core clock 1360 mem clock 1904 fan speed 100 Now that you have it in compute mode that will help in your efficiency. no just do the following settings in bat.file and don't even use MSI afterburner. -10 powlim cclock 1200 cvddc 820 mclock 2150 mvddc 900. Give rxboost 1 and -strap 1 another try. If that crashes the system try just -strap 1 without rxboost to see if it is stable. All these cards should be in the 29-31mhs range. Thanks, that just made a huge difference. I'm now at 32.2mhs Though I'm just looking back and I made a typo in my bat file and wrote [-powlim +18] instead of [-powlim -10] why is it that it's typically a negative number? I've seen this in a couple places and just wondering
|
|
|
|
hustleman
Jr. Member
Offline
Activity: 155
Merit: 6
|
|
April 25, 2020, 04:36:48 PM |
|
The negative powlim is to reduce the amount of electricity the card uses to mine. Reducing the powlim will save you on ur electric bill and will also reduce the amount of heat (temp) of the card. Its all about finding the right efficiency with these GPUs.
|
|
|
|
m4cook
Newbie
Offline
Activity: 1
Merit: 0
|
|
April 26, 2020, 05:15:11 PM |
|
hash decreased from 32 Mh/s to 22 Mh/s. I have two cards : Nvidia GTX 1070it. How to solve this problem ?
|
|
|
|
Digital_Seytan
Jr. Member
Offline
Activity: 222
Merit: 2
digiseytan@walletofsatoshi.com
|
|
April 26, 2020, 06:53:39 PM |
|
|
DonateSATS:Digiseytan@WALLETOFSATOSHi.COM SHOPFREE: https://satsback.com/register/1QEJyGPlg4LN5kwx ETC+Zil Pool:https://k1pool.com/invite/895eb07555
|
|
|
Binary100100
Member
Offline
Activity: 180
Merit: 10
|
|
April 26, 2020, 09:18:25 PM |
|
Maybe. You've been asking for at least two months now and he hasn't answered you. What updates are you hoping for? There are other people here that can help you with basic questions if you need something. Is it something that only the dev can fix for you? If it's just because you haven't had an update then perhaps stop using the program. That way his dev fee gets cut and he'll be forced to develop again. Meanwhile, he's still getting hundreds if not thousands of dollars a day from it.
|
|
|
|
iDisturbia
Jr. Member
Offline
Activity: 84
Merit: 6
|
|
April 26, 2020, 09:43:38 PM |
|
If it ain’t broken ...!!
|
|
|
|
Binary100100
Member
Offline
Activity: 180
Merit: 10
|
If it ain’t broken ...!! Exactly. In my opinion it's because previous updates yielded better results so people like him are hoping for a huge improvement each time and can never get enough. Yet they don't understand that there will always be limitations which are mostly due to the drivers or the equipment itself. Not everything relies on the developer. If it was up to them they would make my GTX 1070's hash at 100MH/s with only 50w from the wall. But it's simply not possible. Yet some people think that with each update it might bring it ever so closer to being a possibility. It's very likely that the program has reached it's most optimal potential already. Yet... they persist that there will always be another which will outshine the previous versions. And so on, so forth. Based on my experience with updates this isn't always the case. Examples being updates with Windows 10 operating systems. On rare occasions it makes thing so much worse! Or updates to your cellphone operating system. Sometimes drivers or bios for hardware can get botched and mess something up. All because people were crying for an update. Bottom line, leave good enough alone. There are only so many ways to make a wheel. Seriously, Claymore is making a fortune based on the work that he already did. What else is he obligated to do? Whether he improves his software or not he's making pretty good money from our 1% devfee's and he doesn't even have to log on to play Tech Support. He can retire... or as Digital_Seytan asked... die (let's hope not) and still have a very nice and somewhat stable income.
|
|
|
|
iDisturbia
Jr. Member
Offline
Activity: 84
Merit: 6
|
|
April 27, 2020, 11:36:51 AM |
|
If it ain’t broken ...!! Exactly. In my opinion it's because previous updates yielded better results so people like him are hoping for a huge improvement each time and can never get enough. Yet they don't understand that there will always be limitations which are mostly due to the drivers or the equipment itself. Not everything relies on the developer. If it was up to them they would make my GTX 1070's hash at 100MH/s with only 50w from the wall. But it's simply not possible. Yet some people think that with each update it might bring it ever so closer to being a possibility. It's very likely that the program has reached it's most optimal potential already. Yet... they persist that there will always be another which will outshine the previous versions. And so on, so forth. Based on my experience with updates this isn't always the case. Examples being updates with Windows 10 operating systems. On rare occasions it makes thing so much worse! Or updates to your cellphone operating system. Sometimes drivers or bios for hardware can get botched and mess something up. All because people were crying for an update. Bottom line, leave good enough alone. There are only so many ways to make a wheel. Seriously, Claymore is making a fortune based on the work that he already did. What else is he obligated to do? Whether he improves his software or not he's making pretty good money from our 1% devfee's and he doesn't even have to log on to play Tech Support. He can retire... or as Digital_Seytan asked... die (let's hope not) and still have a very nice and somewhat stable income. Well said and 100% agreed with everything there. I am not sure why some folks are a little concerned with 1% dev fees. I like how Claymore (and others) remind us .. “if you don’t like it, don’t use it”! Simple. Write your own mining code and save your 1%. But, the fact is many of us are not capable of doing that! So, it’s a fair business! My problem however is with those wanna-be coders who have copied from certain open-source APIs / codes, closed source theirs and charge 4 times more - and you can’t even keep your rigs up and running for an hour straight!! Those folks never see a penny (??) from me no matter how much I love certain coins. Instead, I buy them directly!! Both Claymore and Phoenix deserve to enjoy early retirement!
|
|
|
|
|