|
Awesome Miner version 6.5.2
Features
- Access to miner power usage via the Customize Progress Field feature
- Added a new setting to keep older console output logs (generated when running mining software as Administrator)
Integration
- Added support for URL handlers to add External Miners and setting up mining software via web links
User interface
- Native overclocking controls available in a new tab GPU Clocking for the selected miner to give faster access to overclocking operaitons.
Mining software
- CryptoDredge 0.20.1, including support on Linux
- GMiner 1.45
- Nanominer 1.3.4
- TT-Miner 2.2.5
Corrections
- Correction to benchmark feature where it in some scenarios could stop running
- Correction to NbMiner implementation on Linux
- Correction to feature for setting Antminer default pools via the web interface when using the Antminer S9 March 2019 firmware
|
|
|
|
Hey Patrike,
Is there a way to set the Global Electricity Price (per kWh) via HTTP API?
Thank you!
There are no feature to set the value via API today. You can set the price via the Rules, where you can use the Action called Set Property. You can for example use the rules to define different electricity prices depending on time or day. Rule actions can also be triggered via the API if needed. I do understand this doesn't give as much flexibility as having an API method for setting an exact price from an external application. Please let me know if you still prefer the API method and I will look into supporting it when possible. Thanks! Hey Patrike, If you could add it to your ever expanding "to-do list" that would be awesome! That would help me a lot! Here in Texas we have the option to go with a variable rate electricity plan that changes the price per kWh every 5 minutes. I'm currently monitoring the real-time price and have different rules and profiles setup in AM for my miners. For example, if the price per kWh goes crazy, up to say $1.00 /kWh, then my mine basically shuts down until the price drops. If the price goes negative (which happens often at night), the miners will overclock and go nuts! One more thought that might open up a whole new world of possibilities for AM: What if you gave us the option to pass a variable into AM via API that could be used elsewhere in the application? Maybe just a few empty variable slots (Var1, Var2, Var3, etc.) then those variables could be used as "Triggers" inside a rule. Just a thought. Thanks again! The API method for electricity price should be possible to include in the near future as it's only a matter of exposing a feature already available. Thanks for the suggestion about variables, that would make sense as well. Another concept available already today via the triggers are Miner Tags. You can set Tags on miners via the API and you can have trigger looking at certain tags. |
|
|
|
I think this is a TRM issue about not being consistent in how it closes down when using the X-button. I do get a close down message on my system when I do this - while you don't.
It's quite easy to expose the power usage so I can do that in one of the next versions. You can expect the following to work once available:
status.KHashAvg * 1000 / Math.Max(miningHelper.GetPowerUsage(), 1), 2)
As the Power Usage might be 0 in the very beginning, using the Math.Max operation will ensure that we never go below '1' in assumed power usage.
When you press (X) you trigger a CTRL_CLOSE signal rather than CTRL_BREAK or CTRL_C. This one is a little special under Windows, it will send the signal to all processes connected to the console, then also proceed to kill the process(es) after some timeout. Only 95% sure about this, but I after a quick test it seems Windows actually proceeds to kill the process after the signal handler(s) for all processes has returned, not after a timeout. We have an async shutdown procedure, i.e. the signal handler just sets a flag and triggers a cond var, then returns, and other threads in the process take care of the shutdown. Unfortunately, Windows kills the process before those threads even have gotten started. Sometimes you'll see the first log message, other times you don't. We need to add a block in the Windows signal handler and don't return before the shutdown has completed, or you can send CTRL_BREAK/CTRL_C or char input 'q' to the process, all those fixes should have the effect of a proper shutdown taking place. -- K Many thanks for sharing the details about this close behavior. It's correct that Windows eventually will terminate the process after sending the Close signal, but the software receiving the signal is given some time to terminate in a nice manner. I've explored the CTRL-C approach a bit, and that would be possible for Awesome Miner to support if neededl. It's a bit more complicated as it's based on a concept where you release and attach console handlers, but should be fully possible. |
|
|
|
Thanks for the notice. There will soon be a new version with this one included. |
|
|
|
Thanks for the update. I will take another look at this.
It's basically the difficulty to set only specific properties on this Antminer web page - as the page itself requires you to construct about 15 parameters with correct values and if any value is blank it can affect other parameters. Did you have 1, 2 or 3 default pools defined before you run into this issue?
I see the btc.com app for Antminer setup works fine (it can change LPM,ELPM mode and on latest firmware even OC) so until we resolve this in AM this tool can be handy. Yes all pools previously where defined via SSH, and after new firmware via web it just the web tool for default mining pools fill's in this data at pool 3 if you left it blank (no need for 3 pools anyway). The part that I miss the most is ability to manually change mining pools on the fly or via profit switching rule. From this perspective on the latest firmware every time I would need to apply new default and reboot the miner  Thanks for the details. I just noticed that the Antminer S9 firmware from May 31st has a new concept for the low power mode settings as well. It's not configured the same was as the previous firmware. Update: I did identify a case with the Antminer S9 firmware from March where Awesome Miner didn't set the parameters correctly if you only had 1 or 2 default pools defined. I will correct that scenario. |
|
|
|
Hi Patrike,
the blockchain driver will be unrecognized again.
Although I don't recommend to use the AMD Blockchain Beta drivers in general, from what I can see you do get the AMD GPU's listed in Awesome Miner system tab even with this driver. The Native Overclocking will however not work. Can you please let me know if it's a specific AMD GPU I should verify with, where it's not working in your case? Thanks! |
|
|
|
This is great to see you that you keep replying every feedbacks from 2014 to till now! Sad for me that I haven't tried the Awesome Miner, but today in a gossip, One of my friends suggested me to look at the Awesome Miner! So, I am here to learn about this miner and I will let you know my feedback after using it!
Thanks! All feedback and feature suggestions are of course welcome. |
|
|
|
When I stop TeamRedMiner via Awesome Miner, the console window of TeamRedMiner says "Shutting down..." and then it will exit right away. Don't you get any console message at all?
Awesome Miner do have a behavior to kill the mining process if it didn't resond to the first request to shutdown. Maybe you can increase the time Awesome Miner will wait to see if TeamRedMiner just needed a bit more time to close down in a nice manner? Increase the value of: Options -> Mining Settings -> Wait for process before terminating.
Thanks for describing the case about dev fee as well. Awesome Miner doesn't have any feature today to do this on a mining software basis. Some mining software also have different dev fee depending on algorithm.
Yeah in my case, awesome miner does an insta-kill, i tried the timeout but no avail. I tried closing TRM with a ctrl+c, works normally, the thread shudown and miner closes I'm not able to reproduce the problem - although I do see that TeamRedMiner terminate very fast, it says "Shutting down" for a very show moment before doing so. Not easy to see that message because the process is so fast. The fact that TeamRedMiner terminates very fast in your case as welll indicates that it did respond to the friendly Window close message from Awesome Miner. If it didn't respond, the close process would have taken about 5 seconds (configurable) before Awesome Miner took the next action where the process is killed. To compare with SrbMiner, it says "Releasing GPU resources", but it takes 2 seconds so it's easier to see that message. Awesome Miner is first sending a friendly request to close the mining window to the mining software - and this is the same for all mining software. You can try the same behavior by closing the mining software with the X-button in the mining window. When you click X to close the window, it should be the same concept as when Awesome Miner requests to close the mining window. When I try this with TeamRedMiner and SrbMiner, I do get the exact same behavior as when Awesome Miner do it. Could you compare with this as well? I tried logging the exit process, and confirm that the TRM isnt shutting down normally. TRM shutdown works only with Ctrl+C and not when clicking close in the CMD windows. Feature Request : Allow Total miner power usage value as a property in the status object. https://support.awesomeminer.com/support/solutions/articles/35000086023-customize-progress-fieldTrying to bring this value to Progess Field "Efficiency: " + Math.Round(status.KHashAvg / {{{status.PowerUsage}}}, 2) +" KHash/Watt"Thanks !! I think this is a TRM issue about not being consistent in how it closes down when using the X-button. I do get a close down message on my system when I do this - while you don't. It's quite easy to expose the power usage so I can do that in one of the next versions. You can expect the following to work once available: status.KHashAvg * 1000 / Math.Max(miningHelper.GetPowerUsage(), 1), 2) As the Power Usage might be 0 in the very beginning, using the Math.Max operation will ensure that we never go below '1' in assumed power usage. |
|
|
|
Hi Patrike,
I've been away for a couple of weeks. Awesome Miner happily churned away without any issues. Love the program!
Just updated everything to the latest versions and ran into a small issue when benchmarking. It seems that the long term benchmarks (10 minutes, 20 minutes, 1 hour, etc) are quitting after a few minutes, similar to the "Fast" or "Precise" benchmarks.
The following is (I think) the relevant log entries:
Thanks,
...jim
EDIT: It also occurs on the "Fast" and "Precise" benchmarks, sometimes. Depends on how quickly the error occurs.
Thanks Jim - I think I understand the scenario and will make a correction. |
|
|
|
Thank you for adding SINOVATE X25X algorithm.
Keep up the good work, you are doing great Patrike.
Thanks for the nice feedback! |
|
|
|
Hi
It's currently only ASIC miners running the custom firmware AsicFW that can report power usage to Awesome Miner.
is there a way to see the total for each ASIC ?
I only see power for each board but no total for all three boards.
Thank you
Hi and thanks for your question. It should be possible to see the miner total power usage in the Performance column for each miner in the Miners tab. Can you please open the Options dialog, Profitability section and review the setting "Power usage mode". Can you please select the first or second item in the list here about "Actual usage"? |
|
|
|
Hey Patrik,
I think I already asked this sometime before, but couldnt show a picture and details because this happens very rare.
- One or more cards are crashed. The mining window is still open, but isnt mining anymore.
- In AM it still shows "mining", but without any informations. AM will not restart that miner.
- The "GPU" tab is empty, also the "Pools" tab.
Is it possible to setup a rule to catch this? Or maybe AM can handle this automatically? I already tried some rules, but none works. Neither a single rule, or combined rules. For example:
- The trigger "Detect Miner State -> Mining" doesnt work.
- "Detect Device Count -> 0" doesnt work.
- "Power Usage -> 0" doesnt work.
I think because there is no miner API established because of the crash, even if the mining window is still open?
Is the predefined rule "Offline detection" also unable to find this case? Next time this happens, could you please save the API report for me and I can review this scenario in more detail? You can generate the API report via the toolbar Tools -> API Report. Thanks! |
|
|
|
1) If I click "Upgrade Remote Agent" is upgrade previous version
2) Yes, but Claymore Old logs are not saved
Thanks for the updated details: 1) I would actually recommend a reboot of this remote computer to begin with. In case there are pending Windows Updates to be installed on next reboot, these can prevent other installers from running correctly. If this didn't solve the problem, please uninstall Remote Agent manually and then install it again using the links above. 2) Right now the log file will only contain the latest log of the most recently started Claymore process. Once you start the process the next time it will clear the log. This behavior cannot be changed in Awesome Miner right now. 2) How do I diagnose which graphics card has failed? if I do not see the old log, as the miner program automatically restarts?  3) How to configure so that when unsuccessful restarts of the miner program, awesome rebooted rig ? 2) I understand your point and I will add a new property for keeping these logs. This will make it possible to go back to older mining software log files and review what happend. 3) Please see "Example: Reboot GPU miner after several mining software crashes": https://support.awesomeminer.com/a/solutions/articles/35000085907-rules-detect-mining-issues-and-automate-tasks |
|
|
|
Think there is a bug on tool "default pools via web"
When ever i apply a new default pool on reboot of S9 it activates ELPM
The most recent development versions of Awesome Miner has some internal changes to how the default pools are set on the Antminer via the web interface. It should also be a bit better at avoiding these kind of side effects. Can you please try if version 6.4.4 works better? Got the latest version, still the same issue Also it looks like its filling pool 3 config file with some part of command  I guess there is no chance to implement for the latest s9 firmware change pool option in AM? Thanks for the update. I will take another look at this. It's basically the difficulty to set only specific properties on this Antminer web page - as the page itself requires you to construct about 15 parameters with correct values and if any value is blank it can affect other parameters. Did you have 1, 2 or 3 default pools defined before you run into this issue? |
|
|
|
Great update, I was wondering can you add feature to send keystrokes on console window on GPU miners?
(for example a Z command in claymor to initiate automatic drci)
This can come handy when doing diagnostics on rigs without a need to leave AM
Thanks for the suggestion - I've noted your request and will explore this. |
|
|
|
Hey Patrike, Is there a way to set the Global Electricity Price (per kWh) via HTTP API? Thank you!  There are no feature to set the value via API today. You can set the price via the Rules, where you can use the Action called Set Property. You can for example use the rules to define different electricity prices depending on time or day. Rule actions can also be triggered via the API if needed. I do understand this doesn't give as much flexibility as having an API method for setting an exact price from an external application. Please let me know if you still prefer the API method and I will look into supporting it when possible. Thanks! |
|
|
|
Thanks for the frequent updates. I'm sure that it is a lot of work and I hope that you're open to ideas for improvement.
I had idea that I had is very unlikely that you will (or even can) implement but I don't see why it would be impossible.
Is there any way that you can configure a tool that will "auto-tune" OC settings based on efficiency or hashpower or balanced between the two?
I've spent the last two days trying to fine tune my OC settings for my RX 570 cards running the latest Claymore and rxboost enabled. I found that I pretty much have to start over because the OC settings that worked for PhoenixMiner doesn't work well with Claymore. It would run for about half an hour and then reboot which needless to say is frustrating. Currently I have about 30 different test profiles and as I'm halfway down the list I'm only finding four to be capable of running for more than an hour.
If only there was a way that I could select if I want my rig to run for most optimal settings based on benchmarks. Basically, if it crashes then the benchmark automatically would lower the OC setting applied by a preset integral then try again. This would continue until a stable OC setting is found then prompt the user to save it as a new profile.
I know that it's a lot to ask to an already jam packed program but it would make some users lives a tad bit easier.
Thanks.
Thanks for the feature suggestion. It's always appreciated with these kind of requests and I do take note of all requests here. You are correct that this request may not be a quick feature to implement, but it's still interesting. I had a similar request from another user in the past as well. As Awesome Miner can do both clocking and measure power usage, it would be possible to implement a feature that balance the overclocking and power usage to get optimized settings. The complexity is to handle and recover from crashes if the overclocking is too high, where the computer might have to reboot and the benchmark should continue and adjust the clocking in a smart way. This feature will probably not be implemented in the near future, but I do like the concept and it's for sure a candidate for future improvements. |
|
|
|
|
Awesome Miner version 6.5.1
User interface
- GPU Clocking Profiles can be applied via the context menu in the miner list
- Display number of selected miners in the Miners tab
- Display number of selected miners and number of successful and failed operations in the Antminer Control dialog (for operations like Configure API Access and Default Pools)
Mining software
- Added NbMiner support on Linux
- CpuMiner-Opt 3.9.1.1
Corrections
- Corrected hashrate unit displayed in the Mining History dialog
|
|
|
|
When I stop TeamRedMiner via Awesome Miner, the console window of TeamRedMiner says "Shutting down..." and then it will exit right away. Don't you get any console message at all?
Awesome Miner do have a behavior to kill the mining process if it didn't resond to the first request to shutdown. Maybe you can increase the time Awesome Miner will wait to see if TeamRedMiner just needed a bit more time to close down in a nice manner? Increase the value of: Options -> Mining Settings -> Wait for process before terminating.
Thanks for describing the case about dev fee as well. Awesome Miner doesn't have any feature today to do this on a mining software basis. Some mining software also have different dev fee depending on algorithm.
Yeah in my case, awesome miner does an insta-kill, i tried the timeout but no avail. I tried closing TRM with a ctrl+c, works normally, the thread shudown and miner closes I'm not able to reproduce the problem - although I do see that TeamRedMiner terminate very fast, it says "Shutting down" for a very show moment before doing so. Not easy to see that message because the process is so fast. The fact that TeamRedMiner terminates very fast in your case as welll indicates that it did respond to the friendly Window close message from Awesome Miner. If it didn't respond, the close process would have taken about 5 seconds (configurable) before Awesome Miner took the next action where the process is killed. To compare with SrbMiner, it says "Releasing GPU resources", but it takes 2 seconds so it's easier to see that message. Awesome Miner is first sending a friendly request to close the mining window to the mining software - and this is the same for all mining software. You can try the same behavior by closing the mining software with the X-button in the mining window. When you click X to close the window, it should be the same concept as when Awesome Miner requests to close the mining window. When I try this with TeamRedMiner and SrbMiner, I do get the exact same behavior as when Awesome Miner do it. Could you compare with this as well? |
|
|
|
I will soon make a new release where I will include the latest CpuMiner-Opt as well. Thanks! Update: I noticed v3.9.1.1 was just released as well |
|
|
|
|
Show Posts
