|
Hi Patrike, can you please made reading "block reward" property more accurate. At the moment accuracy is only one digit after point. For Ethereum this accuracy is not more enough. Regards.
|
|
|
|
Hi
Does the latest version of awesome miner include the new dual mining capabilities of t-rex?
Like when you set up a managed miner can you select dual mining for t-rex instead of manually adding the dual mining arguments in the command line section?
Thanks
looks like this dual mining with t-rex is less profitable compare to 71%-81% standart LHR unlock. You will lost more profit if your second algo is power hungry algo like Kawpow or Octopus. After many tests lolminer 1.33 is a winner for me. tested with 3060 v. 1.0 |
|
|
|
Inn Options->Mining Settings there is an option to "Use Ctrl+C event for closing miner process" This does not seem to work anymore.
example T-rex : If you use any coreclock lock settings, they are not dropped. T-rex is not exited properly.
Instead of showing: Main loop finished. Cleaning up resources...
T-Rex finished.
It seems to just be killed. That means that when you fire up another miner , the core clock in t-rex still exists so oc in that application is all over the place.
If I remote inn to a miner and do Ctrl+C the core clock lock is dropped and the exit is done correctly.
Hope this is fixed easily, or maybe there is an workaround for this ?
Could this be related to that you run the T-Rex miner process as Administrator? In that case I think Windows will block you from sending Ctrl+C to the process. You are right, had to enable adminmode to turn on coreclock lock. Damn it. Well the hunt for a workaround begins... If your rig OS is Linux (dont know a Windows syntaxis), then you can simple use a "Batch commands" in Profit profile for each algo where you want set fixed core clocks. Work with any miners software. Commands like this: Before start: sudo nvidia-smi -lgc 1545,1545 After stoping: sudo nvidia-smi -rgc Enjoy |
|
|
|
Hi Patrike, can you pls. add a rule action where user can apply a set of active online services (or simple switch on/off some services). So user can apply a low energy algos set. For example: at the night, whenn solar panels not deliver energy. Regards.
You can make 2 profit profiles and can switch it via rule. Make 1 for high energy algos and 1 for low energy algos. But sure i am agree, switch off and on some online services via rule would be nice. you are right, thanks for idea |
|
|
|
|
Hi Patrike, can you pls. add a rule action where user can apply a set of active online services (or simple switch on/off some services). So user can apply a low energy algos set. For example: at the night, whenn solar panels not deliver energy. Regards.
|
|
|
|
Patrike, can you please implement a easy way to execute a linux commands from AM profit profile. At the moment i have a problem if i use miners command --lock-cclock After using this command and later after switching to other mining programm GPU core clock still have value from last --lock-cclock call. But not ervery mining programm have option to lock core clock. If once core clock was locked, i cant change values with simple overclocking. This state must be reseted. Only a way to use a "nvidia-smi -rgc" command before or after mining process. But at the moment AM not have possibility to run commands on Linux side without opening separate cmd process window. Even this command: plink -no-antispoof -ssh user@192.168.188.89 -pw myPassword sudo nvidia-smi -rgc (plink is a part of putty installation package) not work from "ProfitProfile->Batch commands" section, but work perfect from cmd window. I have try to execute a simple windows command: C:\Windows\System32\cmd.exe to test if something happens in ProfitProfile->Batch commands. Its happens nothing. CMD window will not be opened. For which commands type this "batch commands fields" is? Cant find documentation for that. Regards. Hi, Please note that the "Batch commands" for a Profit Profile are configured per Algorithm. It isn't applied for all algorithms. When I try it for a specific algorithm, I can simply enter "cmd.exe" and it will launch a command window when I start the miner with this specific algorithm. Can you please double check that it was configured for the correct algorithm? For nvidia-smi I think it relates to that it must be started with sudo, but your workaround should probably work as well given that it's actually executed for the specific algorithm. Of course i use "Batch commands" on algorithm level in "Profit profile". In my situation happens nothing after adding simple call of cmd.exe. Also nothing happens if i try to call cmd.exe from "Managed profit miner properties"->"Command Line"->"Execute additional commands" and nothing if i call from "Profit profile propertis"->"Batch commands"->"Execute additional commands" I dont know whats wrong with my AM. its looks so simple und logicaly to add this functionality, but for me simple not work.... And in log file nothing. |
|
|
|
Patrike, can you please implement a easy way to execute a linux commands from AM profit profile. At the moment i have a problem if i use miners command --lock-cclock After using this command and later after switching to other mining programm GPU core clock still have value from last --lock-cclock call. But not ervery mining programm have option to lock core clock. If once core clock was locked, i cant change values with simple overclocking. This state must be reseted. Only a way to use a "nvidia-smi -rgc" command before or after mining process. But at the moment AM not have possibility to run commands on Linux side without opening separate cmd process window. Even this command: plink -no-antispoof -ssh user@192.168.188.89 -pw myPassword sudo nvidia-smi -rgc (plink is a part of putty installation package) not work from "ProfitProfile->Batch commands" section, but work perfect from cmd window. I have try to execute a simple windows command: C:\Windows\System32\cmd.exe to test if something happens in ProfitProfile->Batch commands. Its happens nothing. CMD window will not be opened. For which commands type this "batch commands fields" is? Cant find documentation for that. Regards. |
|
|
|
Hi Patrike, if i try to download a custom version of MiniZ miner for Linux, the system create wrong file (?smd_process_download=1&download_id=4427_1) instead of normal .gz archive.
Reagards
All built-in versions of MiniZ will give proper ZIP/GZ filenames. If you point to a custom file on the MiniZ web site, they are actually named strangely like in your example. I assume the extract process might fail here. same error when i try download GMiner |
|
|
|
Hi Patrike, can you please improve a trigger (Device property -> Core clock)
The reaction of this triger is to quick. I try to catch errors of GPU by reading a core clock property, if equals 0 then i do rig reboot. But after starting new mining process not all cards is initialized and triger will be activaed. This offten happens with NBMiner. Maybe some delay can be implemented in this triger? Adjusting Options->Advance->Ignore rules on startups not help for this triger.
Regards.
Hello, Can you please try to combine it with the trigger "Mining uptime"? This trigger can be used to ensure the mining software has been running for a few minutes. You can have these two triggers (Device Property + Mining uptime) and "Match all" configured in your rule. thanks for idea, looks like this work now for normal rig. But this trigger combination not work if physical rig contains two or more virtual miners. In my scenario i have such rig. If one virtual rig perform switching, this action activate trigger and whole rig will be rebooted. How to solve this? Thanks for the update. So you have one Managed Miner and one Managed Profit Miner? In case of GPU issues, a reboot will affect both miners. Isn't a reboot what you want in case of GPU issues or what is the behavior you would like to see here? I have both virtual miners configured as Managed Profit Miner. I am understand that in case of GPU core clock fail, both miners will be rebooted. Problem is in profit switching event for this virtual miners. If one of this two miners perfom switching, the rule which compare core clock to null will be activated and perfom reboot like i define in rule action section. Looks like due switching the mining process will be stoped (what is of course normal), core clock go to null and trigger will be active. Maybe is a good idea to deaktivate trigger or reseting timer which count mining time due profit switching process? Regards |
|
|
|
|
Hi Patrike, if i try to download a custom version of MiniZ miner for Linux, the system create wrong file (?smd_process_download=1&download_id=4427_1) instead of normal .gz archive.
Reagards
|
|
|
|
Hi Patrike, can you please improve a trigger (Device property -> Core clock)
The reaction of this triger is to quick. I try to catch errors of GPU by reading a core clock property, if equals 0 then i do rig reboot. But after starting new mining process not all cards is initialized and triger will be activaed. This offten happens with NBMiner. Maybe some delay can be implemented in this triger? Adjusting Options->Advance->Ignore rules on startups not help for this triger.
Regards.
Hello, Can you please try to combine it with the trigger "Mining uptime"? This trigger can be used to ensure the mining software has been running for a few minutes. You can have these two triggers (Device Property + Mining uptime) and "Match all" configured in your rule. thanks for idea, looks like this work now for normal rig. But this trigger combination not work if physical rig contains two or more virtual miners. In my scenario i have such rig. If one virtual rig perform switching, this action activate trigger and whole rig will be rebooted. How to solve this? |
|
|
|
|
Hi Patrike, can you please improve a trigger (Device property -> Core clock)
The reaction of this triger is to quick. I try to catch errors of GPU by reading a core clock property, if equals 0 then i do rig reboot. But after starting new mining process not all cards is initialized and triger will be activaed. This offten happens with NBMiner. Maybe some delay can be implemented in this triger? Adjusting Options->Advance->Ignore rules on startups not help for this triger.
Regards.
|
|
|
|
|
Hi Patrike, can you pls. explain how and where can i see a GPU memory chip temperature in AM? Rigs on Linux (Hive) different cards 2080ti, 3060-3090. Regards
|
|
|
|
|
Hi Patrike, pls. enable ethash algo in MiniZ miner (last ver.)
No data output to the benchmark window if testing PhoenixMiner 5.6d
Autolykos is on the Nicehash now, need to be added to AM
Regards.
|
|
|
|
|
Hi Patrike, how to realise in AM fixed core clock on rig with Linux OS? RaveOS have this feature. This is very usefull for energy saving. And fixed core clock help against trottling on the cards with GDDR6X memory chips. Regards.
|
|
|
|
|
Hi Patrike, can you improve "Create notification" action and add a checkbox where user can define if notification will be with sound or soundless. Regards.
|
|
|
|
Hi Patrike, AM runing 5 days, computer RAM consumption is over 1.2 GB, i have restarted AM and now RAM consumption is only 210 MB.
AM 8.4.7, Win 10 pro, 20H2, build 19042.928, Windows Feature Experience Pack 120.2212.551.0, latest OS updates installed.
I see after AM restart, RAM consumption slowly growing.
Is there any memory leak in code? Regards.
Just checked my memory usage, AM is running for about a week since last reboot - 160 Mb. depend how many miners you use. After one day running, memory usage grow from 200 Mb till 530 Mb. This is most likely related to a known bug in the nVidia drivers. All application that uses nVidia NVLM to monitor the power usage (Awesome Miner, some mining software, ...) will end up with this memory issue. The issue didn't exist back in driver version 457.x, but it's present in 460.x and 461.x and possible more versions. Which version are you running today? Hmm, yes i use linux drivers on remote rigs 460.39 & 460.67. But on computer where AM runing, i use only CPU graphic and no other GPU installed. Not uderstand, how remote machine driver problem can consume AM computer memory. Rigs intself have no problems with this drivers, no memory leaks (at least i never got any messages depends memory on rigs). I control AM computer remotly over Microsoft remote desktop. After last AM restart (3 days ago), memory consumption now arround 960 MB. Need to do AM restart soon. But if you think, that is the NVIDIA drivers problem, i will wait till new driver versions. Regards. |
|
|
|
Hi Patrike, AM runing 5 days, computer RAM consumption is over 1.2 GB, i have restarted AM and now RAM consumption is only 210 MB.
AM 8.4.7, Win 10 pro, 20H2, build 19042.928, Windows Feature Experience Pack 120.2212.551.0, latest OS updates installed.
I see after AM restart, RAM consumption slowly growing.
Is there any memory leak in code? Regards.
Just checked my memory usage, AM is running for about a week since last reboot - 160 Mb. depend how many miners you use. After one day running, memory usage grow from 200 Mb till 530 Mb. |
|
|
|
|
Hi Patrike, AM runing 5 days, computer RAM consumption is over 1.2 GB, i have restarted AM and now RAM consumption is only 210 MB.
AM 8.4.7, Win 10 pro, 20H2, build 19042.928, Windows Feature Experience Pack 120.2212.551.0, latest OS updates installed.
I see after AM restart, RAM consumption slowly growing.
Is there any memory leak in code? Regards.
|
|
|
|
Hi Patrike, today i have created a rig with mixed cards, 4 x 3060ti + 2 x 3090
for this design i have used two miner objects, one with 4 x 3060ti and other with 2 x 3090. For every miner object i have selected only GPU IDs which belongs for 3060ti and 3090 of course. Every virtual rig have own profit profile. Problem is that the power usage for one of this rig (4x3060ti) will be displayed wrong. To much watts. I use for both virtual miners NBMiner. If i check it terminal output of NBMiner, the watt usage per card is ok (124watt on Etherium for 3060ti), but AM show wrong (874 w). In the same time, when both virual miners work, the miner object for 2x3090 show correct power usage. If i stop miner with 2x3090, then displayed in AM power usage for 4x 3060ti is to less as normal (350 w). Whats wrong? Regards.
Hi, 1) Please make sure you don't have any custom GPU mapping enabled (make sure the icon for "Map to system monitoring" is blue, not green). By not having a custom GPU mapping, Awesome Miner will try to map the specific GPUs reported by the mining software to the correct GPU 2) If you still experience this issue, can you please generate and send me an API report for this miner while running (via the toolbar Tools -> API Report). Thanks! Problem solved, the "Map to system monitoring" was a key. Big thanks. |
|
|
|
|
Show Posts
