PhoenixMiner 6.2c: fastest Ethereum/Ethash miner with lowest devfee (Win/Linux)

[PB_SilentMiner]

Hey All,

Just switched from Claymore 11.9, so hoping things are more stable now.

I'm running Windows 10 (1709) AMD RX480 Nitro+ 8Gb Samsung Cards using 18.2.1 driver and latest Phoenix Miner 3.0c

But am a little perplexed, when setting -tt 59 in the cmd line, it does not seem to adhere to it and most cards are over this setting around 64 with fanspeed around 40%


Why?

Instead of -tt 59, try -tt -50, it will set fixed fan speed to 50% and you will get lower temperatures.
You can also increase it if that doesn't give you gpu temperature you wanted.

According to the Readme -tt xx is Set fan control target temperature

From using Claymore this is just set the temp I want the GPU's to run at? I.E -tt 59 and the fan control will do the rest. Am I mistaken?

The setting you are referring to I think is -fanmin

Does not help with all this hot weather, I'm having to shutdown my miners as they can't cope.

[lesjokolat]

Anyone tried the new amd 18.8.1 driver yet ?

I have I updated to aug windows august update and AMD 18.8.1.

works fine

[deecept]

Hey All,

Just switched from Claymore 11.9, so hoping things are more stable now.

I'm running Windows 10 (1709) AMD RX480 Nitro+ 8Gb Samsung Cards using 18.2.1 driver and latest Phoenix Miner 3.0c

But am a little perplexed, when setting -tt 59 in the cmd line, it does not seem to adhere to it and most cards are over this setting around 64 with fanspeed around 40%


Why?

Instead of -tt 59, try -tt -50, it will set fixed fan speed to 50% and you will get lower temperatures.
You can also increase it if that doesn't give you gpu temperature you wanted.

According to the Readme -tt xx is Set fan control target temperature

From using Claymore this is just set the temp I want the GPU's to run at? I.E -tt 59 and the fan control will do the rest. Am I mistaken?

The setting you are referring to I think is -fanmin

Does not help with all this hot weather, I'm having to shutdown my miners as they can't cope.

-tt [number] = target temperature (auto adjusting fan speed to match temp)
-tt -[number] = fan speed (static fan speed)

[Metroid]

Hey All,

Just switched from Claymore 11.9, so hoping things are more stable now.

I'm running Windows 10 (1709) AMD RX480 Nitro+ 8Gb Samsung Cards using 18.2.1 driver and latest Phoenix Miner 3.0c

But am a little perplexed, when setting -tt 59 in the cmd line, it does not seem to adhere to it and most cards are over this setting around 64 with fanspeed around 40%


Why?

-tt 59 is the temperature it targets, for example if your fan rpm speed is 3000 and you dont set anything then it will be at 100% 3000 rpm to hit that temperature, if you want your fan with fixed speed then you must set -fanmax 59.

The problem about this miner is that it crashes a lot at the beginning, from 10 tries, 7 times it crashed and that is clean install with window s 1709, in claymore it never crashes, if you are not at home and it restarts and crashes you lose hours of mining. So the dev needs to fix this thing, if it does not crash at beginning then is very stable, need to get through the starting point, need a new version with the fix.

[crypper]

Hey All,

Just switched from Claymore 11.9, so hoping things are more stable now.

I'm running Windows 10 (1709) AMD RX480 Nitro+ 8Gb Samsung Cards using 18.2.1 driver and latest Phoenix Miner 3.0c

But am a little perplexed, when setting -tt 59 in the cmd line, it does not seem to adhere to it and most cards are over this setting around 64 with fanspeed around 40%


Why?

-tt 59 is the temperature it targets, for example if your fan rpm speed is 3000 and you dont set anything then it will be at 100% 3000 rpm to hit that temperature, if you want your fan with fixed speed then you must set -fanmax 59.

The problem about this miner is that it crashes a lot at the beginning, from 10 tries, 7 times it crashed and that is clean install with window s 1709, in claymore it never crashes, if you are not at home and it restarts and crashes you lose hours of mining. So the dev needs to fix this thing, if it does not crash at beginning then is very stable, need to get through the starting point, need a new version with the fix.

Well,
Think of it in this way: you are the only one (or one of a few) who has this issue. Most likely it is caused by your set up, rather than some bugs in the miner. It's a bit naive to expect phoenix to release a "a new version with the fix" when you are not even properly describing the problem: what is your setup? how many cards do you have? at what point the miner crashes? During the dag generation? do you have enough virtual memory?
Have you tried -gser, -altinit?

[Metroid]

Well,
Think of it in this way: you are the only one (or one of a few) who has this issue. Most likely it is caused by your set up, rather than some bugs in the miner. It's a bit naive to expect phoenix to release a "a new version with the fix" when you are not even properly describing the problem: what is your setup? how many cards do you have? at what point the miner crashes? During the dag generation? do you have enough virtual memory?
Have you tried -gser, -altinit?

The startup crash happens, on any number of gpus, tested on systems with  1 to 13 gpus and all give the error, virtual memory is set to 50gb, although i tried with no virtual memory and the problems still happen, gser and alinit set, problems still happen and the system is celeron 3900 ssd sandorfce based 128gb memory ram 8gb, log does not show much, here it is,

Code:

2018.08.07:05:58:55.281: main Phoenix Miner 3.0c Windows/msvc - Release
2018.08.07:05:58:55.281: main Cmd line: -epsw x -tt 60 -fanmax 85 -tmax 90 -altinit
2018.08.07:05:58:55.281: main No CUDA driver found
2018.08.07:05:59:00.088: main Available GPUs for mining:
2018.08.07:05:59:00.088: main GPU1: Radeon (TM) RX 470 Graphics (pcie 1), OpenCL 2.0, 4 GB VRAM, 32 CUs
2018.08.07:05:59:00.171: main ADL library initialized
2018.08.07:05:59:00.437: main Listening for CDM remote manager at port 3333 in full mode
2018.08.07:05:59:00.437: main Starting GPU mining
2018.08.07:05:59:00.438: main Matched GPU1 to ADL adapter index 35 (method 1)
2018.08.07:05:59:00.519: main GPU1: Created ADL monitor for adapter 35; overdrive version: 7
2018.08.07:05:59:00.519: main GPU1: using AMD driver ver 18.7.1

Also tried on different amd drivers, nothing at all, that is what the log shows. Previous release 2.9, i never had a startup crash and like i said is a windows 10 x64 1709 fresh install, no updates were downloaded, nothing was modified other than I disabled windows update.

[crypper]

Well,
Think of it in this way: you are the only one (or one of a few) who has this issue. Most likely it is caused by your set up, rather than some bugs in the miner. It's a bit naive to expect phoenix to release a "a new version with the fix" when you are not even properly describing the problem: what is your setup? how many cards do you have? at what point the miner crashes? During the dag generation? do you have enough virtual memory?
Have you tried -gser, -altinit?

The startup crash happens, on any number of gpus, tested on systems with  1 to 13 gpus and all give the error, virtual memory is set to 50gb, although i tried with no virtual memory and the problems still happen, gser and alinit set, problems still happen and the system is celeron 3900 ssd sandorfce based 128gb memory ram 8gb, log does not show much, here it is,

Code:

2018.08.07:05:58:55.281: main Phoenix Miner 3.0c Windows/msvc - Release
2018.08.07:05:58:55.281: main Cmd line: -epsw x -tt 60 -fanmax 85 -tmax 90 -altinit
2018.08.07:05:58:55.281: main No CUDA driver found
2018.08.07:05:59:00.088: main Available GPUs for mining:
2018.08.07:05:59:00.088: main GPU1: Radeon (TM) RX 470 Graphics (pcie 1), OpenCL 2.0, 4 GB VRAM, 32 CUs
2018.08.07:05:59:00.171: main ADL library initialized
2018.08.07:05:59:00.437: main Listening for CDM remote manager at port 3333 in full mode
2018.08.07:05:59:00.437: main Starting GPU mining
2018.08.07:05:59:00.438: main Matched GPU1 to ADL adapter index 35 (method 1)
2018.08.07:05:59:00.519: main GPU1: Created ADL monitor for adapter 35; overdrive version: 7
2018.08.07:05:59:00.519: main GPU1: using AMD driver ver 18.7.1

Also tried on different amd drivers, nothing at all, that is what the log shows. Previous release 2.9, i never had a startup crash and like i said is a windows 10 x64 1709 fresh install, no updates were downloaded, nothing was modified other than I disabled windows update.

Well, your cmd line is very strange: 1. it doesn't specify any pools or wallets; 2. it contained -epsw parameter that is claymore's parameter and not phoenix's. Can you post your bat file here and a screenshot or log from start of mining to the error. Also you say it returns you some error - what is it? it's not in your log.

Also now looking at your setup, I 'm wondering how  -eres parameter works. It says "-eres <n> Allocate DAG buffers big enough for n epochs ahead (default: 2) to". Does it mean it will allocate buffer of 2xDAG on a GPU? If so then your 4GB RX470 is too small for that. Anyone knows how -eres works?

[johnalim]

Interesting indeed.

Move the GPU with that error to a different slot with different cables, and see what GPU has the error. If the GPU you moved has the error again, I would assume it could be failing, it's the luck of the silicon lottery..

Sorry for the late reply. I tried to disable GPU 12 and the miner works with GPUs 1-11. Then I tried to disable GPU9 and re-enable GPU 12 and the miner also works with GPU12 now working and GPU9 idle. I swapped the cards a bit too just to rule out hardware.

It appears only 11 GPUs can run at any given time. So this confirms it is not a GPU hardware problem.

I raised the VRAM to min 40GB and max 56GB and all 12 GPUs were humming along just fine. Thank you PhoenixMiner devs for the tip.

[Metroid]

Well, your cmd line is very strange: 1. it doesn't specify any pools or wallets; 2. it contained -epsw parameter that is claymore's parameter and not phoenix's. Can you post your bat file here and a screenshot or log from start of mining to the error. Also you say it returns you some error - what is it? it's not in your log.

Also now looking at your setup, I 'm wondering how  -eres parameter works. It says "-eres <n> Allocate DAG buffers big enough for n epochs ahead (default: 2) to". Does it mean it will allocate buffer of 2xDAG on a GPU? If so then your 4GB RX470 is too small for that. Anyone knows how -eres works?

There is no hope, all other settings are default, so i have no idea really, also i have 5 types of motherboards and all of them show the same problem, my guess it lies within the network, somehow, pm detects some anomaly and crashes it, claymore might detect it and direct it forward, these things are too complex, too many things that cant be fixed, need to accept, although like i said, previous pm vesion worked without problems, pm dev said this version changed many things.

Well i'm not a miner, i just test things out from time to time.

[crypper]

Well, your cmd line is very strange: 1. it doesn't specify any pools or wallets; 2. it contained -epsw parameter that is claymore's parameter and not phoenix's. Can you post your bat file here and a screenshot or log from start of mining to the error. Also you say it returns you some error - what is it? it's not in your log.

Also now looking at your setup, I 'm wondering how  -eres parameter works. It says "-eres <n> Allocate DAG buffers big enough for n epochs ahead (default: 2) to". Does it mean it will allocate buffer of 2xDAG on a GPU? If so then your 4GB RX470 is too small for that. Anyone knows how -eres works?

There is no hope, all other settings are default, so i have no idea really, also i have 5 types of motherboards and all of them show the same problem, my guess it lies within the network, somehow, pm detects some anomaly and crashes it, claymore might detect it and direct it forward, these things are too complex, too many things that cant be fixed, need to accept, although like i said, previous pm vesion worked without problems, pm dev said this version changed many things.

Well i'm not a miner, i just test things out from time to time.

If you choose to accept - it's up to you, but i'm telling you, your bat file is wrong (you are mixing up commands from claymore and phoenix).
Another idea: auto-tune was introduced in 3.0 (if i'm not wrong). Try to by-pass it by specifying -gt parameter (e.g. -gt 15)

[ThePunisher49X]

Anyone tried the new amd 18.8.1 driver yet ?

I have I updated to aug windows august update and AMD 18.8.1.

works fine

Any increase in hashrate after the update to 18.8.1?

Still running the blockchain drivers from last year and if they are hashing at around the same then I wouldn't bother to update them since they are stable enough running for the past 6 months.

[lesjokolat]

Anyone tried the new amd 18.8.1 driver yet ?

I have I updated to aug windows august update and AMD 18.8.1.

works fine

Any increase in hashrate after the update to 18.8.1?

Still running the blockchain drivers from last year and if they are hashing at around the same then I wouldn't bother to update them since they are stable enough running for the past 6 months.

i ve let it run now almost 24hrs, am finding it actually a bit slower. I think the new AMD optimizations in the driver may have reduced some phoneix kernel speed.

[spykerhond]

This miner fucks out on cuda

Hey dude.
From your posts I am convinced you don't have the brain to handle something like a miner.
You come across as a moron, and I believe it.
You should not mine,  try ping-pong or something that doesn't require intelligence.
I seriously doubt you know which end of a video card to plug into a computer.

Hope these words weren't too big for you, I tried to keep it simple.
Have someone read this to you.

Sorry I offended you , was a bit tired after struggling with miner crashes right through the night. You seem very in love with the word "I". Just a word of caution , please do not adress anybody in person like you type on the keyboard, people can be very dangerous, better only on the net, much safer this way.

[Stihlus]

This miner fucks out on cuda

Hey dude.
From your posts I am convinced you don't have the brain to handle something like a miner.
You come across as a moron, and I believe it.
You should not mine,  try ping-pong or something that doesn't require intelligence.
I seriously doubt you know which end of a video card to plug into a computer.

Hope these words weren't too big for you, I tried to keep it simple.
Have someone read this to you.

Sorry I offended you , was a bit tired after struggling with miner crashes right through the night. You seem very in love with the word "I". Just a word of caution , please do not adress anybody in person like you type on the keyboard, people can be very dangerous, better only on the net, much safer this way.

Usually...almost on any forum / thread you want to find something helpful related to your issue, that's the idea. It's not the right place to argue/dispute such things.
In my opinion both of you start with your left leg, it's not fine when you start to swear/offend someone ... even if it's in real life or online
Just take chill and try to solve your issues with miner. I'm sure someone will help with the right info if you try to explain what's the issue, what error/behavior do you have with the miner.

Have a nice day !
Regards,

[rgzrgz]

"debugger detected" rx 570 and 580. win 10 ltsb x 64.
i try -amd

[mtechnolog]

Miner closes when it starts mining devfee. I'm using the latest version.

[bitt86]

Hello there, Phoenix makes me 189.8 from 188,5 mhs on my 6x rx580. Its also working on my 6x 1050ti 4gb making me more than 15mhs each. Im so glad from your work Dev. Keep your hard work on. Cheers!

[ojcrisanto]

This software gives me stable at 31.1 mh/s using rx 570 4gb cclock 1150 mclock 2150 mvddc 830 cvddc 850. Thank you dev for your hardwork.

[bitt86]

Hello there, Phoenix makes me 189.8 from 188,5 mhs on my 6x rx580. Its also working on my 6x 1050ti 4gb making me more than 15mhs each. Im so glad from your work Dev. Keep your hard work on. Cheers!

All clear, but my power drain from wall is 1130W :O ( 6x rx580 8gb ) Someone have similar situation ?

[sxemini]

@bitt86
How much mvddc, cvddc, cclock, mclock? Sorry you ask a question an give us no informations? How we should answer?