Topic: PhoenixMiner 6.2c: fastest Ethereum/Ethash miner with lowest devfee (Win/Linux) - page 37. (Read 784950 times)

If you are running windows than most likely it is the vram that windows reserves.

Linux doesnt reserve vram for system. 

I have the same problem, anyone found solution ?

If you configured the bat file correctly which means just changing your wallet to yours which you say you have done then it takes some time to see balance in your wallet,you can see your balance in the pool where you are mining,just paste your wallet address in the pool website and you can see your stats there,it take some time to update the stats from one to a few hours.You can find everything there regarding your mining stats.

I have a rig running 4 x RX6800XT under PM 5.7b and it is ok but not yet perfect but acceptable for now much better than the week of downtime i had getting there. See my recent posts for the way to go to mine RX6000 series under Linux it doesnt appear to be a PM problem but but rather an opencl driver issue related to ROCM? and correct kernels to consider.

I am using Manjero at present but currently building a ubuntu image in the hopes of resolving stale share issue with better ROCM opencl drivers, but so far below is an example of performance under linux for these cards and I will say i like the RX6000 even if the prices are jacked up they run quiet with there massive heatsinks and even without fans on for periods very nice for primary indoor heating source so thats a massive saving on not having to turn the heating on thru winter and all is considered in the Return of investment calculation although i do mourn for my RVII as they where special.
Right now I put £35 a week electric in and get out £70 plus free heating. Bring on the depths of winter I am ready.


 
Eth: Mining ETH on eth-eu1.nanopool.org:9999 for 3:51
Available GPUs for mining:
GPU1: AMD Radeon RX 6800 XT (pcie 3), OpenCL 2.0, 16 GB VRAM, 72 CUs
GPU2: AMD Radeon RX 6800 XT (pcie 7), OpenCL 2.0, 16 GB VRAM, 72 CUs
GPU3: AMD Radeon RX 6800 XT (pcie 10), OpenCL 2.0, 16 GB VRAM, 72 CUs
GPU4: AMD Radeon RX 6800 XT (pcie 13), OpenCL 2.0, 16 GB VRAM, 72 CUs
GPU1: 47C 49% 154W, GPU2: 50C 49% 156W, GPU3: 50C 49% 158W, GPU4: 45C 49% 159W
GPU1: cclock 1200 MHz, cvddc 868 mV, mclock 1074 MHz, Tj 53C, Tmem 66C, 420 kH/J
GPU2: cclock 1200 MHz, cvddc 868 mV, mclock 1074 MHz, Tj 58C, Tmem 68C, 408 kH/J
GPU3: cclock 1200 MHz, cvddc 862 mV, mclock 1074 MHz, Tj 57C, Tmem 70C, 409 kH/J
GPU4: cclock 1200 MHz, cvddc 868 mV, mclock 1074 MHz, Tj 54C, Tmem 62C, 407 kH/J
GPUs power: 627.0 W; total power: 827.0 W; cost: 4.57 USD/day; 312 kH/J
Current -gt 15,15,15,15
Eth: Accepted shares 397 (91 stales), rejected shares 0 (0 stales)
Eth: Incorrect shares 2 (0.50%), est. stales percentage 22.92%
Eth: Maximum difficulty of found share: 8659.5 GH (!)
Eth: Average speed (5 min): 243.789 MH/s
Eth: Effective speed: 232.74 MH/s; at pool: 232.74 MH/s

Eth speed: 250.439 MH/s, shares: 397/0/2, time: 4:47
GPUs: 1: 64.730 MH/s (101/1) 2: 56.211 MH/s (98) 3: 64.730 MH/s (105/1) 4: 64.769 MH/s (93)

read page 1 of this thread slowly and carefully especially the miner paramters list also available to you if you PhoenixMiner -h from cmd line. It may take time to understand hbut you have two files of concern, config and epools. as far as i know and in my experience the wallet address is placed into the epools file and there are examples for it on most major pools.

any news on when next version gonna be?

because my 6600xt is not recognised yet by PM v5.7b.

is there a chance the hashrate will be better on new version that recognised it or is it gonna be the same?

because i can't use all that AMD settings for my 6600xt.

hi so i just set up my miner and im mining i put my adress in the .bat and i didnt notice any difference in in my ETH balance how many shares do i need to see a difference or did i do something wrong? or are some wallets not compatible?

If you have enough patience could show up someone with more knowledge to resolve your issue, but until then you should help yourself as most of us do by teaching our self with reading new information and try doing little changes until succeed.
What I can see in your log file you have a driver issue. Look at lines 4 and 16 from your file.

On the 1st page of this tread, devs stated they do not recommend using drivers higher than 21.8.1


I would recommend you to install and clean up your system from the drivers and after a restart, do a clean installation of any driver of your choice between 20.4.x and 21.8.1. So far for me, the best for mining is 21.3.1 except if you have RX 6600 XT cards in your rig.
If drivers does not help, try lower overclock settings until get it work.

A very basic question about watch dog. Is it automatically enabled if I don't specify it in config file? My understanding is that if GPU freeze it will try to restart the miner. Am I correct?

Yesterday, I experienced 'Bug Report', which is an AMD term when a GPU caused driver timeout. The miner paused with something like memory dump displayed on the session. I noticed it about 45 minutes later and manually restarted miner. I wonder whether this situation is supposed to be covered by watch dog or not.

The OC for that particular GPU could work couple of days without issue. May be I need to lower it a little bit. But if it can automatically recovered, it would be great.

Why my tahiti 280x is not supported in phoenix miner 5.7b When trying to mining etchash, ethereum classic... I tried in hive OS phoenix miner 5.7b with driver amd 20.40

i'm using windows 10 with nvidia driver 471.68.
i added -cvddc -400 to miner, but it do not work. card still mining on default vcore.
mining on gtx 1660s, and run phoenix as administrator.

If there are other newish or semi literate none windows miners here still perplexed with the RX6000 series not working check the link out at end of this post and note the following info regarding compatible Kernels which is info hard to come by ayy....


ROCM

Prerequisites

The AMD ROCm platform is designed to support the following operating systems:




Note: Ubuntu versions lower than 18 are no longer supported.

Note: AMD ROCm only supports Long Term Support (LTS) versions of Ubuntu. Versions other than LTS may work with ROCm, however, they are not officially supported.


source

https://rocmdocs.amd.com/en/latest/Installation_Guide/Installation-Guide.html

Info discovered on link below.

https://community.amd.com/t5/drivers-software/opencl-not-working-with-rx-6700-xt/td-p/469244

I haven't been around much, mostly because things had been working well for me until two days ago with Radeon Software 21.9.1 where PhoenixMiner broke again.   I'm not on the Linux side, so can't tell you the specifics, but the big problem comes from driver versions that clash with PhoenixMiner.   Things work, you install new video drivers, and PM stops working.   Go back to previous video drivers, and it works again.   AMD has been updating and changing the drivers to add features and improve performance of the Radeon 6000 series(I have a 6800 so also know the Big Navi pains of PhoenixMiner breaking).   As long as you keep your drivers to something supported by the miner, you are fine.   You can add the manual -clkernel 0 to get around THAT problem, but it is less than ideal.

Yes, there are very few posts by anyone from PhoenixMiner, and it is annoying, to the point where I and others are tempted to change for a miner that actually has SOME presence to address questions/problems when they come up.   The real key is that many are busy at this time of the year, and we don't have the time to tinker with a new miner to get it working, at least as well as PM does when it works.

With all my RX580 8GB on Win10:

PhoenixMiner.exe -rvram -1-rxboost 1 -clKernel 3 -eres 1 -> No hashrate
PhoenixMiner.exe -rvram -1-rxboost 1 -clKernel 3 -eres 0 -> 32.6 MH/s
PhoenixMiner.exe -rvram -1-rxboost 1 -clKernel 1 -eres 5 -> 32.3 MH/s


If -clKernel 3 can't use on next DAG epoch?

I appreciate that post because i know mining is an investment and steep learning curve even if it is very frustrating especially when intermittent problems arise.

My own step this week from Vega 2 to big navi has perplexed me for days till just now actually but ladies and gents we are hashing RX6800XT on this rig after all but ubuntu is gone.

I have just solved my problem by jumping to a new distro based on Arch linux, before today i was a ubuntu man and have all progress today to thank for the info found in the post below copied here for any others who wish to mine a RX6000 series card in linux. Its a good place to start and this works i just done it though now need to go and unlock the kernel for tweaking the amdgpu pro but can report 60Mhs out of the box and the temp reported for the RX6800XT like this is low at 47C though she is pulling 220W for that so uvolt etc commencing in 321 kernel willing.....



https://community.amd.com/t5/drivers-software/6900xt-and-opencl-on-linux/td-p/439474/page/2

This might sound harsh but here it is.
Sometimes users here can help you, sometimes not. Sometimes they want to help, sometimes they don't.
Phoenix people rarely post here anymore so you'll get very little or none from developers, they now depend on users
to do help and troubleshooting.

If you have no experience in mining and no ability to think on your own and do sequential logical troubleshooting and testing
and you are relying on others to kindly guide you through all your issues step by step you are probably in the wrong business.
Mining is NOT a plug and play project, it doesn't work that way. There are hundreds of combinations and setting that you will
have to do on your own. Sorry if the truth stings a bit.

I can see it, this is no where going forum, but my problem is wired. PhoenixMiner 5.6b everything works, on PhoenixMiner 5.7b one GPU doesn't work, I'm noob, so I don't know much about setting up the files, that's why I put the log up to see if someone from here help brother out.

For your problem the answer was given i see to read the manual and check your config file. I suggest you remove all but one GPU's from the rig and get that one working before adding the second, third, fourth.. etc.  Do it incrementally updating your config file each time until you understand and become comfortable with the requirements of the config commands for a stable system and to prevent actual GPU damage due to errors in config scripts such as setting a volt to high and pushing a card to hot so it expires permanently as it can happen with incorrect config variables stated especially in a mixed gpu rig.

For eg a 6 gpu rig should have 6 separate statments for each variable such as cvvdc etc or just one for all GPU's (unwise due to hardware tolerances as each GPU is its own beast but this can be a good debug test or place to start from if learning however know your GPU limits and never exceed the lowest threshold or smoke and fire then bitching on here of course haha)

Theres no help here by looks of it, just people raging and judging about other people who post fishing links and click on fishing links. Dont click on fishing links people its on you if you want to get PM downloaded then its post one for the link and nothing else in this thread.

Even PhoenixMiner appears to be making false statements here regarding the functionality of its latest release under Linux because the opencl drivers / kernels appear to be well and truly fubar for RX6000 series cards at present which is on AMD i bet lol. which is all very nice and im sure they (PM) are busy coding to correct or side step it rather than spend time here with us guys lol.

I am now considering Hive rather than waste more time trying to discover which kernel driver combo work or not for myself i refuse windows on principles but it may come to it at this rate, this is not what an open platform is meant to be there is not enough info about PM under any platform and its requirements so i am losing faith now throw us a bone please

So is there any help on these forums? Or you just list your inquiry and forgot about it?