PhoenixMiner 3.0c: fastest Ethereum/Ethash miner with lowest devfee (Windows)

[dohfish]

Phoenixminer,
My rig config: win 10 pro, 6 x msi 570 gaming x 4G, mining at Nanopool, mclock 2030, vcddc 870, cclock 1100, vcore 950, fan 80%.

Miners 2.6 and 2.7a both crash win, BSOD. Claymore doesn’t, but it resets the voltages and clocks after sometime after the decree sessions I guess. Now I’m running 2.7a with -mi 8. This reduces the local hashrate to around 173 from 175 of claymore.

Please look into this issue.

Thank you.

Try give them more voltage and test the miner first. If you can run for 24hrs stable, try decrease the voltage. For my experience, the claymore hardware control is not so stable. I am now using overdrivetools to control. It 's easy to use.

Then Phoenix draws more power than claymore with reduced hashrate. I’ll wait. Thanks for the tip.

What you are seeing is that your clocks are not stable, software access the systems in different ways which can make it behave differently.

2030 is per my experience quite high for a 570, I have only a single card (570) which are stable at 2030+ - The median is 2000 on memory for my 120 rx570's

Thanks a lot for the info on mem clocks. And as you’re running lots of rx 570, could you help me solve this problem with one of my cards: I set fan speed to 100% in afterburner 4.4.2 but claymore reports 90% and this number never goes higher, so the GPU hits 57C. I’m asking because the other 5 report higher 95-96% fan speed and are running cool at 50C.
Is this a hardware issue? I have reinstalled windows but the problem persists.

Thank you.

I cannot stress this enough - Do not use any other software for controlling your GPUs than OverDriveNtool - It is the ONLY stable software for controlling it.

But in either case, 57C is in nowhere near a problem, I run mine at 65-70C, you need to remember that a GPU is built for temperatures in the area of 90-100C.

Anything below 80 is perfectly safe imo.

[1534283705]

1534283705

[1534283705]

1534283705

[tenmoi]

Phoenixminer,
My rig config: win 10 pro, 6 x msi 570 gaming x 4G, mining at Nanopool, mclock 2030, vcddc 870, cclock 1100, vcore 950, fan 80%.

Miners 2.6 and 2.7a both crash win, BSOD. Claymore doesn’t, but it resets the voltages and clocks after sometime after the decree sessions I guess. Now I’m running 2.7a with -mi 8. This reduces the local hashrate to around 173 from 175 of claymore.

Please look into this issue.

Thank you.

Try give them more voltage and test the miner first. If you can run for 24hrs stable, try decrease the voltage. For my experience, the claymore hardware control is not so stable. I am now using overdrivetools to control. It 's easy to use.

Then Phoenix draws more power than claymore with reduced hashrate. I’ll wait. Thanks for the tip.

What you are seeing is that your clocks are not stable, software access the systems in different ways which can make it behave differently.

2030 is per my experience quite high for a 570, I have only a single card (570) which are stable at 2030+ - The median is 2000 on memory for my 120 rx570's

Thanks a lot for the info on mem clocks. And as you’re running lots of rx 570, could you help me solve this problem with one of my cards: I set fan speed to 100% in afterburner 4.4.2 but claymore reports 90% and this number never goes higher, so the GPU hits 57C. I’m asking because the other 5 report higher 95-96% fan speed and are running cool at 50C.
Is this a hardware issue? I have reinstalled windows but the problem persists.

Thank you.

I cannot stress this enough - Do not use any other software for controlling your GPUs than OverDriveNtool - It is the ONLY stable software for controlling it.

But in either case, 57C is in nowhere near a problem, I run mine at 65-70C, you need to remember that a GPU is built for temperatures in the area of 90-100C.

Anything below 80 is perfectly safe imo.

Thank you for the info. But we are stressing them 24/7 so I’m rather worried.

As for overdrive tool, do you know how to control the fan speed? How many rpms in overdriventool translate to 1% in afterburner?

[dohfish]

The RPMs depend on the model, so that would be trial and error

I have had my GPUs running at ~70C 24/7 for ages, my oldest 580 is 18 months old, have been on 100% ever since i got it - Dont worry about that.

I typically just set the max temp to 80 and let the software handle it on its own.

[ManDoneKTM]

Mine are running between 43-48 ºC , rx580
Also i have an r9 200 hd7970 , is running @59-61 ºC is a little bit hotter .

[cmd_whoami]

hi, i'm mining with 12 gpu rig.....

all my gpy working ok at 29 mh/s... but when i turn off or disconnect the screen....  mi hashrate on gpu 0  goes to 5 mh/s....

what is the problem?

can anyone help?

Is your monitor connected to the hdmi port on the GPU that drops speed when you turn it off ?

yes, connected on hdmi port on the GPU

@ KADOK29 - Have you tried turning on On Board Video in BIOS to use CPU graphics? You may also need a HDMI dummy plug. Newegg and Amazon have them. Other wise even when you use TeamViewer it will pull graphics from the 1st 16x slot. You can check by seeing if Intel HD Graphics is in device manager under Display Adapters.

[axe0911]

I was testing out the 2.7a version. Getting this error -

17581:20:02:46.344: main Increase the Windows page file size to at least 16 GB to avoid out of memory errors and unexpected crashes.

And the GPU OC gets removed. My page size has already been increased.

[ckai93]

a few hours back pool shows that rig went offline but the miner still submit shares.
i've restarted the miner and still submit shares as usual but pool still doesn't show that.
been in touch with ethermine support but they insist that it wasn't a server side fault.

still not sure if it was a pool or miner issue but till then, i'll continue using claymore.

Had the same issue connected to the asia1.ethermine.org. Looks like the difficulty had been raised on that pool too?

Switched to the us server for now and it’s reporting as normal.

Be good to know if you hear anything back!

guess i'm not the only victim here. are u using -proto 2 for stratum proxy?
here's the reply I got from ethermine:

Hi XXXXXXX,

It is normal that some of our servers do not respond to ping requests.

Our initial investigation resulted that you where a victim of a so called BGP hijacking attack. The attacker was able to spoof the IP addresses of our mining server and redirect your hashrate to his own stratum server. If you look at your mining software log you will see that your miner was disconnected from our pool at the time the issue started. When it reconnected it was redirected to the attackers server. This can be seen by the work packages provided by the stratum server. While our stratum has a fixed difficulty of 4000MH the server from the attacker sent out work packages with a difficulty of 10000MH. For example:

Real work package: New job #9591c733 from asia1.ethermine.org:14444; diff: 4000MH
Attacker work package: New job #9538fa6e from asia1.ethermine.org:14444; diff: 10000MH

Furthermore our work package includes 4 parameters while the work package of the attacker included only 3 parameters:

Real work package: Received: {"id":5,"jsonrpc":"2.0","result":["0x9591c733c9eccdc842b361e699142278cfba5de5a17bf4a49d8a36de62a23f89","0x4e99a30e99712c8c6e292fe7ba6b27a37c7ced12e2ec7862f31fb676724cb404","0x0112e0be826d694b2e62d01511f12a6061fbaec8bc02357593e70e52ba","0x4e0e58"]}
Attacker work package: Received: {"jsonrpc":"2.0","id":0,"result":["0x6facf1be3f19166290f14f39a8bef0ae7ec3f282fa7378f9f3a098f1bf1ce81f","0x4e99a30e99712c8c6e292fe7ba6b27a37c7ced12e2ec7862f31fb676724cb404","0x000000006df37f675ef6eadf5ab9a2072d44268d97df837e6748956e5c6c2116"]}

Unfortunately there is not much we can do from our side against such kind of attacks as the current Ethereum stratum protocol is neither encrypted nor authenticated. The only real solution to avoid this type of attack is to implement an encrypted and authenticated stratum protocol via SSL. We will be working together with the mining software developers to implement such a method.

We are very sorry for the inconvenience this incident has caused.

so yea, for people using -proto 2 for stratum server, I would be super cautious about it. But since I have yet to see this kind of problem with Claymore, I'll be sticking with them as PhoenixMiner actually gave me lower or equal hash rates as compared to Claymore; plus I'm not that heavy of a miner even if there's a 1 mh/s bump per card. 

[fadius]

I tested 2.7a yesterday on an RX580 8GB ... seemed to work well, very slightly faster (~ 0.2 MH/s) than claymore at same core/mem clocks/voltages.
I like it and would definitely considering switching to it from claymore.

1 issue I noticed is that when I setup the monitoring port, I'm able to see the console output by accessing http://IP_ADDRESS:PORT however a monitoring app I used (claymore monitor on Android) is no longer able to get the stats. If I understood correctly, claymore monitoring apps should still work ?

thanks!

[strike101]

somethings weird both my rigs 12xRx570 temps just went up to around 75-79c , started around 1-3 days ago

they were previously hovering around 62-69c for weeks

1100 - 850mv gpu clock , 1950 - 900mv mem clock  @ around 29.x mh/s


i didn't upgrade , change anything...

[Tranzius]

Can someone help me with this? What does it mean? Showing randomly in miner..

"CMD command miner_file rejected: CMD is in read only mode"

thank you.

[tenmoi]

The RPMs depend on the model, so that would be trial and error

I have had my GPUs running at ~70C 24/7 for ages, my oldest 580 is 18 months old, have been on 100% ever since i got it - Dont worry about that.

I typically just set the max temp to 80 and let the software handle it on its own.

Thank you very much. You have been very friendly and helpful.

[tenmoi]

somethings weird both my rigs 12xRx570 temps just went up to around 75-79c , started around 1-3 days ago

they were previously hovering around 62-69c for weeks

1100 - 850mv gpu clock , 1950 - 900mv mem clock  @ around 29.x mh/s


i didn't upgrade , change anything...

I know this is off topic but you can get 29.x at mem clock 1950. It’s awesome. My rx 570 gets 29.x at mclock 2030. what is your 570 model and where did you get the mod? Mine goes down 1MH for every 10 mv reduction in mem clock.

Thank you.

[Bigf00t]

Please make a Linux version

[junkstory]

Please make a Linux version

Why not just install Windows 10? Even with Claymore v11.0, my GTX1070 runs 2-3 MH/s slower on Linux, and also 2 degree hotter, primarily because there is no ability to undervolt and underclock the GPU core with nice tools like Afterburner. I absolutely do not understand the rhyme or reason of mining on Linux.

If you don't want to pay for the full Windows 10 license, just don't activate it - you won't miss out on anything.

[PhoenixMiner]

Phoenixminer,
My rig config: win 10 pro, 6 x msi 570 gaming x 4G, mining at Nanopool, mclock 2030, vcddc 870, cclock 1100, vcore 950, fan 80%.

Miners 2.6 and 2.7a both crash win, BSOD. Claymore doesn’t, but it resets the voltages and clocks after sometime after the decree sessions I guess. Now I’m running 2.7a with -mi 8. This reduces the local hashrate to around 173 from 175 of claymore.

   Hopefully this will be resolved once the hardware control options are supported by PhoenixMiner (in 2.7b). In our tests roughly the same clocks/voltages that work with Claymore should work with PhoenixMiner with small exceptions.

For a guy like me with a ton of miners I would really like to see a global update or a push feature to push an update from a main console.  Claymore has this issue as well so it's not just you all.  But for a guy that has 50-100 + miners man it would rock.  That is what makes linux so nice.  I'm half and half right now but might go full linux because of the console push features.

Hats off to you all though.  You are for sure getting better by far since 2.3!  Testing 2.6 on 10 machines again.

   This is a good idea but it also would be a dangerous way to sneak some malware via this vector. Perhaps when we implement fully encrypted (and authenticated) remote management API, it would be safe enough.

Hello,
need option for controlling fanspeed like claymore " -tt 60 -fanmin 75 " for example.

   Coming soon (a few days until release of 2.7b with all hardware control functionality).

Any update on the Linux version?

   Soon-ish (TM)  It will have to wait until the hardware control and dual mining are implemented. It would help if you tell us what distro of Linux you are using (or intend to use) and (if you are using AMD cards), which drivers, so that we now what to test first.

https://www.virustotal.com/#/file/237d6f3f781d9a4a718559c956d2da0bf81e82c374e7ce6dca44e397bcd404ff/detection

Detected 9 times for specifically Troj.Win64.Wdfload   

   Well, we were starting to worry that something happened to you - no posts stating for the Nth time the false positives of AV software  Seriously, use your brain please. We are busting our asses and using considerable time and expertise to ... install a shitty trojan or your mining rig?! This is like building a flying saucer in order to use for shoplifting. There is quite simple explanation for this - some of our anti-debug code has the same signature as part of this stupid trojan. As our anti-debug code is randomized and scrambled in each build, there is no way we would chase the exact few dozens of bytes for weeks, or to turn off our anti-debug measures just to get "clean" report from these AV programs.

a few hours back pool shows that rig went offline but the miner still submit shares.
i've restarted the miner and still submit shares as usual but pool still doesn't show that.
been in touch with ethermine support but they insist that it wasn't a server side fault.

still not sure if it was a pool or miner issue but till then, i'll continue using claymore.

   Please check in the log if the miner is not redirected by someone using the Claymore's remote debugging to mine for another address. As a rule, do not use the commnad-line option -cdm 2 (or -mport with positive port number) if your CDM port is accessible from outside your LAN, or specify a remote management password with -cdmpass and use something like OpenVPN to control PhoenixMiner outside of your LAN.

I have many Incorrect shares after switching DAG size.
Could you please add commands to restart miner when get more than xx incorrect shares?
I mine to nicehash and found this problem every time when switch DAG size.(if not switch DAG,they not have any incorrect share)
Rig Rx560 4Gb.
Thank you.

   This means that the new DAG contains errors. Probably this is result of overclocking or undervolting (especially on memory). If there is an error during DAG generation, there is no easy way to find it and it would create a log of "random" incorrect shares. To solve the problem, lower the memory overclock a bit. We are thinking about a better solution like checking the DAG for integrity or something like this but it would slow down the DAG generation considerably.

I was testing out the 2.7a version. Getting this error -

17581:20:02:46.344: main Increase the Windows page file size to at least 16 GB to avoid out of memory errors and unexpected crashes.

And the GPU OC gets removed. My page size has already been increased.

   This is just a warning that is shown if the total size of virtual memory (physical memory plus current page file size) is less than about 20 GB. The overclock drop is known problem with AMD drivers that randomly reset voltages and/or clocks on some of the cards. We will resolve the issue in 2.7b when we will support the hardware control options. We will be checking periodically if the settings are still in effect and set them again if they aren't.

[preda]

this miner is very very good i tried fo days and works perfect, just claymore has the dual mining that is useful especially with new algho added, saying this i hope you will develop dual mining in phoenix too would be wonderful and you will get so many people by your side using your miner! well done anyway

[PhoenixMiner]

a few hours back pool shows that rig went offline but the miner still submit shares.
i've restarted the miner and still submit shares as usual but pool still doesn't show that.
been in touch with ethermine support but they insist that it wasn't a server side fault.

still not sure if it was a pool or miner issue but till then, i'll continue using claymore.

  Wow, IP hijacking attack?! The -proto 2 or -proto 3 has nothing to do with this BTW. If the attacker has spoofed the IP address of ethermine, there is nothing we can do (and the same goes for Claymore). Until the stratum protocol is encrypted or at least authenticated, these things can happen and the only defense is to check your stats from time to time. Please note that IP hijacking happens not on your machine or your network but on the routers of your ISP or even some of the backbone routers of the Internet, so there is nothing you can do to mitigate it. Fortunately, these things are rare but they will be on the rise with the increasing popularity of the crypto.

I tested 2.7a yesterday on an RX580 8GB ... seemed to work well, very slightly faster (~ 0.2 MH/s) than claymore at same core/mem clocks/voltages.
I like it and would definitely considering switching to it from claymore.

1 issue I noticed is that when I setup the monitoring port, I'm able to see the console output by accessing http://IP_ADDRESS:PORT however a monitoring app I used (claymore monitor on Android) is no longer able to get the stats. If I understood correctly, claymore monitoring apps should still work ?

thanks!

   The remote JSON protocol is the same as the one in Claymore dual miner but the logs are obviously different (the one shown by http://IP_ADDRESS:PORT). Probably the tool you are trying to use is parsing the logs to gather additional information and is "confused" because it expects the logs to be the same as Claymore's character-by-character. This is beyond the scope of compatibility we are willing to provide as we want to add even more information in the console. Hopefully, with the increasing of the popularity of PhonixMiner, more tool creators will add support for our miner as well.

somethings weird both my rigs 12xRx570 temps just went up to around 75-79c , started around 1-3 days ago

they were previously hovering around 62-69c for weeks

1100 - 850mv gpu clock , 1950 - 900mv mem clock  @ around 29.x mh/s

i didn't upgrade , change anything...

   Check your OC settings (esp. core voltages) with something like GPU-z. The AMD drivers tend to "lose" some overclocking settings if the GPU is idle for some time but the Wattman still incorrectly shows the lower voltage. This will be fixed in 2.7b when we will finally provide hardware control setting integrated in the miner itself.

Can someone help me with this? What does it mean? Showing randomly in miner..

"CMD command miner_file rejected: CMD is in read only mode"

thank you.

   If you are not trying to upload epools.txt or config.txt files with the Claymore remote manager, then someone is trying to "hijack" your miner via the remote control port. Fortunately, you haven't enabled the full mode, so this is not possible but the miner warns you that someone tries to upload file via the remote management socket.

Please make a Linux version

   We will, definitely. It would take some time however, so bear with us.

[Tranzius]

Quote from: Tranzius on February 19, 2018, 08:32:47 PM
Can someone help me with this? What does it mean? Showing randomly in miner..

"CMD command miner_file rejected: CMD is in read only mode"

thank you.
   

If you are not trying to upload epools.txt or config.txt files with the Claymore remote manager, then someone is trying to "hijack" your miner via the remote control port. Fortunately, you haven't enabled the full mode, so this is not possible but the miner warns you that someone tries to upload file via the remote management socket.
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Thank you very much for your answer. I'm definitely not tryind to upload any files..

How can I get rid of this? Is this Windows related management port or miner socket?

[PhoenixMiner]

Quote from: Tranzius on February 19, 2018, 08:32:47 PM
Can someone help me with this? What does it mean? Showing randomly in miner..

"CMD command miner_file rejected: CMD is in read only mode"

thank you.
   

If you are not trying to upload epools.txt or config.txt files with the Claymore remote manager, then someone is trying to "hijack" your miner via the remote control port. Fortunately, you haven't enabled the full mode, so this is not possible but the miner warns you that someone tries to upload file via the remote management socket.
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Thank you very much for your answer. I'm definitely not tryind to upload any files..

How can I get rid of this? Is this Windows related management port or miner socket?

  It's a feature of PhoenixMiner. If you don't use remote monitoring at all, just add -cdm 0 to the command-line options and the remote monitoring/management port won't be opened at all.

[xinomahc]

any particular reason i cant seen to pull stats from the miner's api?

my code sample:

Code:

using (var socket = new Socket(AddressFamily.InterNetwork, SocketType.Stream, ProtocolType.Tcp))
            {
                socket.Connect("localhost", 3333);
                socket.Send(Encoding.ASCII.GetBytes("{\"id\":0,\"jsonrpc\":\"2.0\",\"method\":\"miner_getstat1\"}"));
                socket.Receive(receiveBytes, receiveBytes.Length, SocketFlags.None);
            }
            var resp = Encoding.ASCII.GetString(receiveBytes);

my code works flawlessly for claymore but fails for phoenix's api... socket.Receive just hangs forever.. any tips?