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

[sk0mi]

I am getting better hashrate with Claymore, because of rxboost option. Is it an equivalent to that option in phoenixminer??

you run claymore and just after it is loaded the straps you pause (right click, edit, mark), then you run phoenix normally, i gain 1 to 2 hashrate / graphics cards

First card is: RX580 PULSE OC
Other: RX580 Nitro+ SE

https://i.ibb.co/HF3ygxq/SCSC.png

Not only -rxboost, but you need to run -strap 1 command also.
Claymore can find which cards are you using and apply straps for them, if they are in the straps table.
For Polaris cards it’s really big gain.
My 6 x RX470 4GB rig goes from 177MH/s to 184.5-186 MH/s depending on MEM clock (2000-2050)

You can use AmdMemTweak to actually read timings and all the settings from Claymore while Claymore is running.
After that, you can apply those settings through AmdMemTweak with a batch file.
No need for Claymore after that.
AmdMemTweak tool is working good and it’s easier to read and apply straps which are already tested (through Claymore), than testing and making your own. But with AmdMemTweak you can try sqeezing even little more hashrate just with changing those reads you got from Claymore.
It’s a little boring to read and then apply settings for all the cards, so many typing, but once you do it, you won’t be needing Claymore.

For boosting Phoenix I used to run Claymore with -rxboost 1, -strap 1, -wd 0 and -mport 0 commands.
Didn’t pause miner with “mark” but I stopped the cards with typing 0,1...5 to disable all 6 cards.
After that just minimized Claymore and run Phoenix.
All the straps were applied to Phoenix and I was getting same hashrates.

Now I use AmdMemTweak, idk why, maybe because of my OCD and not wanting to run two miners at the same time...

Both, Claymore and AmdMemTweak boost solutions are working fine, you can try any of those.
Also driver can make a small change. 18.12.1.1 is proven to work best for my rigs.
Maybe you’ll get better results with different driver.

[sk0mi]

Which driver is best for rx 570 as it sometimes crashes with blue screen and erratic?

I have rx 470 and 18.12.1.1 is proven to be best for me.
Maybe your clocks or/and voltage are a bit high?
Check those also, might not be the driver but the OC...or even -gt if you set it too brutal

[ivanpop]

For anyone who cares the newest nvidia driver 456.38 increased my hashrate 2-3mhs after this driver update... might help others with 1070's.. but i have noticed on all my rigs their newest driver increased hashrate a couple mhs.

Doesn't increase anything...

[Tms88]

hi there how can we Improve this tweak =>


-mi 14
-openclLocalWork 64
-openclGlobalMultiplier 2048.


because it works but 30% SHARE Invalid.

[UselessGuru]

hi there how can we Improve this tweak =>


-mi 14
-openclLocalWork 64
-openclGlobalMultiplier 2048.


because it works but 30% SHARE Invalid.

Second that. That'd be awesome!

[ManuBBXX]

hi there how can we Improve this tweak =>


-mi 14
-openclLocalWork 64
-openclGlobalMultiplier 2048.


because it works but 30% SHARE Invalid.

Yes, but 30% invalid represent far less even with the  (so fake) hashrate better

[ivcelmik]

Hi all,

I'm getting this on Ubuntu 16.04 18.4 drivers any ideea ?

GPU1: Unable to reset core clocks and voltages - error 0
GPU2: Unable to reset core clocks and voltages - error 0
GPU3: Unable to reset core clocks and voltages - error 0
GPU4: Unable to reset core clocks and voltages - error 0
GPU5: Unable to reset core clocks and voltages - error 0
GPU6: Unable to reset core clocks and voltages - error 0
GPU7: Unable to reset core clocks and voltages - error 0
GPU8: Unable to reset core clocks and voltages - error 0
GPU9: Unable to reset core clocks and voltages - error 0


Thank you in advance,
Best Regards,

Any ideea someone how can i fix this?

Ty in advance

[UselessGuru]

Yes, but 30% invalid represent far less even with the  (so fake) hashrate better

So fake?
When I compare with what I get at the pool the reported numbers look reasonable.

[sk0mi]

epoch 374.. it seems this is the end for 4 Gb ?
my 470 got 1.5 MH/s

I have RX470 4GB cards also.
Try Team Red Miner.
No problem with 374, works like it’s 360
I was very pleased to see that  
Phoenix sometimes gets through, sometimes gives me same hashrates as yours at 374.
TRM keeps working like it’s nothing.
But I saw someone saying it can crash after some time.
I didn’t use it more than 2-3 hours, and during that time everything was OK.
Some mandatory commands in command line but was working just fine.
I’m back at Phoenix but it’s good to have an alternative.
Or just switch to Linux.

[O1eG]

epoch 374.. it seems this is the end for 4 Gb ?
my 470 got 1.5 MH/s

ETC ?
Also.
Waiting PhoenixMiner 5.2...
While you can switch PhoenixMiner to ETH.

[MineOrDie]

With the newest 456.38 nvidia drivers it gives me NVML errors all over the place.

[foucss]

Hello
I try to mine with my r9 290

Phoenix Miner 5.1c Linux/gcc - Release build
--------------------------------------------

Unknown OpenCL driver version! Hashrate and stale shares may suffer
OpenCL platform: OpenCL 2.0 AMD-APP (1800.11)
Available GPUs for mining:
GPU1: Hawaii (pcie 1), OpenCL 2.0, 4 GB VRAM, 40 CUs
GPU2: Hawaii (pcie 2), OpenCL 2.0, 4 GB VRAM, 40 CUs
GPU3: Hawaii (pcie 3), OpenCL 2.0, 4 GB VRAM, 40 CUs
GPU4: Hawaii (pcie 4), OpenCL 2.0, 4 GB VRAM, 40 CUs
GPU5: Hawaii (pcie 5), OpenCL 2.0, 4 GB VRAM, 40 CUs
Unable to initialize AMD hardware monitor
Eth: the pool list contains 1 pool (1 from command-line)
Eth: primary pool: eth-eu1.nanopool.org:9999
Starting GPU mining
Eth: Connecting to ethash pool eth-eu1.nanopool.org:9999 (proto: EthProxy)
Eth: Connected to ethash pool eth-eu1.nanopool.org:9999 (213.32.74.230)
Eth: New job #c1812d9e from eth-eu1.nanopool.org:9999; diff: 10000MH
GPU1: Starting up... (0)
GPU1: Generating ethash light cache for epoch #363
GPU2: Starting up... (0)
GPU3: Starting up... (0)
GPU4: Starting up... (0)
Listening for CDM remote manager at port 3333 in read-only mode
GPU5: Starting up... (0)
Eth: New job #fd47cf28 from eth-eu1.nanopool.org:9999; diff: 10000MH
Eth: New job #68348e60 from eth-eu1.nanopool.org:9999; diff: 10000MH
Light cache generated in 3.0 s (20.7 MB/s)
GPU2: Free VRAM: 3.953 GB; used: 0.007 GB
GPU2: Disabling DAG pre-allocation (not enough VRAM)
GPU2: Allocating DAG for epoch #363 (3.84) GB
GPU2: Allocating buffers failed with: clCreateBuffer (-61).
Fatal error detected. Restarting.
GPU3: Free VRAM: 3.953 GB; used: 0.007 GB
GPU3: Disabling DAG pre-allocation (not enough VRAM)
GPU3: Allocating DAG for epoch #363 (3.84) GB
GPU3: Allocating buffers failed with: clCreateBuffer (-61).
GPU4: Free VRAM: 3.953 GB; used: 0.007 GB
GPU4: Disabling DAG pre-allocation (not enough VRAM)
GPU4: Allocating DAG for epoch #363 (3.84) GB
GPU4: Allocating buffers failed with: clCreateBuffer (-61).
GPU5: Free VRAM: 3.953 GB; used: 0.007 GB
GPU5: Disabling DAG pre-allocation (not enough VRAM)
GPU5: Allocating DAG for epoch #363 (3.84) GB
GPU5: Allocating buffers failed with: clCreateBuffer (-61).
Eth: New job #790a11a8 from eth-eu1.nanopool.org:9999; diff: 10000MH
GPU1: Free VRAM: 3.953 GB; used: 0.007 GB
GPU1: Disabling DAG pre-allocation (not enough VRAM)
GPU1: Allocating DAG for epoch #363 (3.84) GB
GPU1: Allocating buffers failed with: clCreateBuffer (-61).
Eth: New job #b762d767 from eth-eu1.nanopool.org:9999; diff: 10000MH
Eth speed: 0.000 MH/s, shares: 0/0/0, time: 0:00
GPUs: 1: 0.000 MH/s (0) 2: 0.000 MH/s (0) 3: 0.000 MH/s (0) 4: 0.000 MH/s (0) 5: 0.000 MH/s (0)
Eth: New job #ac1c4301 from eth-eu1.nanopool.org:9999; diff: 10000MH

Free VRAM: 3.953 GB; used: 0.007 GB
So a 3.84 GB DAG should enter in the free VRAM.

I should work regarding the DAG size any advice?

Thank you

[kamy1352]

hi
Just used 5.1c
very small increase in hash
right now my hash rate of 2 card increased from 93.4 to 94.1
considering this is my test environment.

1,  Asus TUF EVO Gaming 5600xt on 40.5 MH/s 94-95W(GPU) (original Asus bios update...no boost only mem and GPU clock and voltage boost)
2 MSI AMD 5700 on 53.6MH/s 109-110W(GPU) (no bios change or update...only mem and GPU clock and voltage boost)


is it worth the risk to make change in bios or upgrade for more hash rate with these numbers??

considering my cards  memory work at 1860MHz (max. locked memory speed)

system is stable more or less ....but as I use it for web browsing and watching movies ,it sometimes crash unexpectedly..... not sure whats the reason...
also mu RX5700 is primary graphic card to work in my windows 10

[Ursul0]

Your number looks great, but Im not on those types of 5700xt.

5700 should be able to do about 57mh without the memory timings, so I don't think 52 is a "great looking number"

[GODLIKEE]

Your number looks great, but Im not on those types of 5700xt.

5700 should be able to do about 57mh without the memory timings, so I don't think 52 is a "great looking number"

please share the setting of a 5700 able to do 57mh without the memory timings or with the memory timings

[Ursul0]

Your number looks great, but Im not on those types of 5700xt.

5700 should be able to do about 57mh without the memory timings, so I don't think 52 is a "great looking number"

please share the setting of a 5700 able to do 57mh without the memory timings or with the memory timings

1) you flash bios mod ( populate higher mem frequencies from the strap of 1500)
2) set volts to 788-810mv, ~1360 core and 1860 mem

[abg00]

Hello,

I'm getting this error after 2 days of mining:

Eth speed: 311.939 MH/s, shares: 7024/0/0, time: 26:27
GPUs: 1: 51.948 MH/s (1216) 2: 52.040 MH/s (1175) 3: 52.040 MH/s (1161) 4: 51.885 MH/s (1182) 5: 52.022 MH/s (1164) 6: 52.004 MH/s (1126)
Eth: New job #92ed9bce from eu1.ethermine.org:4444; diff: 4000MH
Eth: New job #790af2c0 from eu1.ethermine.org:4444; diff: 4000MH
Eth: New job #a157ed03 from eu1.ethermine.org:4444; diff: 4000MH
0x00007FFE27409389 (0x0000022475BD94A8 0x000002247F867230 0x0000000000000000 0x00000000FFFFFFFE), aclWriteToMem() + 0x1550FC9 bytes(s)
0x00007FFE25C5823E (0x0000000000000000 0x000000FA0FFFFA50 0x000000FA0FFFFA79 0x000002247F867230), clGetPipeInfo() + 0x1D4CE bytes(s)
0x00007FFE25CE3DC9 (0x0000000000021180 0x000002247F867060 0x000000FA0FFFFA79 0x0000000000000020), clGetPipeInfo() + 0xA9059 bytes(s)
0x00007FFE25D0233E (0x0000000000000020 0x000002247F9E6D99 0x0000022470B9AC10 0x000000FA0FFFFBC0), clGetPipeInfo() + 0xC75CE bytes(s)
0x00007FFE25D025A1 (0x0000022476AE46E0 0x000002247F867088 0x000000FA0FFFFC70 0x0000022471062510), clGetPipeInfo() + 0xC7831 bytes(s)
0x00007FFE25CF4636 (0x000002247F867088 0x0000000000000000 0x000002247F9E7000 0x0000000000000000), clGetPipeInfo() + 0xB98C6 bytes(s)
0x00007FFE25C4A7A3 (0x0000022475BD9510 0x000002247F867060 0x00000224768FCAE8 0x00000224768FCAE8), clGetPipeInfo() + 0xFA33 bytes(s)
0x00007FFE25C4A89B (0x0000000000000000 0x0000022476CA11C0 0x0000022476CA11C0 0x0000000000000000), clGetPipeInfo() + 0xFB2B bytes(s)
0x00007FFE25C3B65E (0x0000022476CA11C0 0x0000000000000000 0x0000000000000000 0x0000000000000000), clGetPipeInfo() + 0x8EE bytes(s)
0x00007FFE25C58209 (0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000), clGetPipeInfo() + 0x1D499 bytes(s)
0x00007FFE49408364 (0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000), BaseThreadInitThunk() + 0x14 bytes(s)
0x00007FFE49F85E91 (0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000), RtlUserThreadStart() + 0x21 bytes(s)

C:\Users\a\Desktop\PhoenixMiner_5.1c_Windows>pause
Press any key to continue . . .

Could be a riser problem ?

I was receiving the same error when I had my RX 5700 XT GAMING OC on a test bench plugged directly into the MOBO, so I don't think it is a riser issue.  I can't confirm it, but it *seemed* to occur whenever GPU Memory Junction Temperature hit 110 degrees (add -hstats 2 to your bat file to display Tmem temp). I think 110 is when the card starts to throttle and that throws this error for whatever reason in PhoenixMiner.  I bumped my -fanmin up to 40 and now my tmem is running at a "cool" 102 degrees. Haven't had the aclwritetomem issue since.

I have this same issue with my RX 5700 XT gigabyte gaming oc (with igors lab mod, no mpt file loaded), I already increased my fan speed to 80% to see much lower memory junction temperature (after 80% fan speed, 80-85C). I'm still getting this errors and sometimes freezes my system. I don't think this is a riser problem. I need some help.

0x00007FFE27409389 (0x0000022475BD94A8 0x000002247F867230 0x0000000000000000 0x00000000FFFFFFFE), aclWriteToMem() + 0x1550FC9 bytes(s)
0x00007FFE25C5823E (0x0000000000000000 0x000000FA0FFFFA50 0x000000FA0FFFFA79 0x000002247F867230), clGetPipeInfo() + 0x1D4CE bytes(s)
0x00007FFE25CE3DC9 (0x0000000000021180 0x000002247F867060 0x000000FA0FFFFA79 0x0000000000000020), clGetPipeInfo() + 0xA9059 bytes(s)
0x00007FFE25D0233E (0x0000000000000020 0x000002247F9E6D99 0x0000022470B9AC10 0x000000FA0FFFFBC0), clGetPipeInfo() + 0xC75CE bytes(s)
0x00007FFE25D025A1 (0x0000022476AE46E0 0x000002247F867088 0x000000FA0FFFFC70 0x0000022471062510), clGetPipeInfo() + 0xC7831 bytes(s)
0x00007FFE25CF4636 (0x000002247F867088 0x0000000000000000 0x000002247F9E7000 0x0000000000000000), clGetPipeInfo() + 0xB98C6 bytes(s)
0x00007FFE25C4A7A3 (0x0000022475BD9510 0x000002247F867060 0x00000224768FCAE8 0x00000224768FCAE8), clGetPipeInfo() + 0xFA33 bytes(s)
0x00007FFE25C4A89B (0x0000000000000000 0x0000022476CA11C0 0x0000022476CA11C0 0x0000000000000000), clGetPipeInfo() + 0xFB2B bytes(s)
0x00007FFE25C3B65E (0x0000022476CA11C0 0x0000000000000000 0x0000000000000000 0x0000000000000000), clGetPipeInfo() + 0x8EE bytes(s)
0x00007FFE25C58209 (0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000), clGetPipeInfo() + 0x1D499 bytes(s)
0x00007FFE49408364 (0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000), BaseThreadInitThunk() + 0x14 bytes(s)
0x00007FFE49F85E91 (0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000), RtlUserThreadStart() + 0x21 bytes(s)

C:\Miners\PhoenixMiner_5.1c_Windows>pause
Press any key to continue . . .

I think I just solved my issue by downloading all the visual c++ redistributable runtimes (check the link below to download all the redist runtimes). My rig is running 26 hours straight now.

https://www.techpowerup.com/download/visual-c-redistributable-runtime-package-all-in-one/

I've had the same problem a few times already, but updated the driver to the latest version and also decreased slightly the memory and for now it seems OK.

same trouble
5.1c
20.8.1
13 x rx5600xt
all card and all riser ok
8gb ram
92gb swap

Eth: Share actual difficulty: 7740 MH
Eth: Share accepted in 38 ms
Eth: New job #a285d2e1 from ssl://eu1.ethermine.org:5555; diff: 4000MH
0x00007FF87E32E2B9 (0x0000023C1422D5B8 0x0000023C1861D9F0 0x0000000000000000 0x0000000000000000), aclWriteToMem() + 0x14E6779 bytes(s)
0x00007FF87CBC8AEE (0x0000023C17D48EF0 0x000000E44E5FF8E0 0x000000E44E5FF8F9 0x0000023C1861D9F0), clGetPipeInfo() + 0x2022E bytes(s)
0x00007FF87CC60546 (0x0000000000004900 0x0000023C1861D820 0x000000E44E5FF8F9 0x0000023C178825B0), clGetPipeInfo() + 0xB7C86 bytes(s)
0x00007FF87CC809CF (0x0000000000000020 0x0000023C17409B19 0x0000023C178825B0 0x000000E44E5FFA70), clGetPipeInfo() + 0xD810F bytes(s)
0x00007FF87CC80C65 (0x0000023C17882668 0x0000023C1861D838 0x000000E44E5FFAC9 0x00007FF800000000), clGetPipeInfo() + 0xD83A5 bytes(s)
0x00007FF87CC727A7 (0x00000243EFF560C0 0x0000000000000000 0x0000023C1861D838 0x0000023C1740A000), clGetPipeInfo() + 0xC9EE7 bytes(s)
0x00007FF87CBB8EDC (0x0000023C180B5858 0x0000023C1861D820 0x0000023C180B5858 0x0000023C17C753C0), clGetPipeInfo() + 0x1061C bytes(s)
0x00007FF87CBB8FCB (0x0000000000000000 0x0000000000000000 0x0000023C18718340 0x0000023C18718340), clGetPipeInfo() + 0x1070B bytes(s)
0x00007FF87CBA943A (0x0000023C18718340 0x0000000000000001 0x0000000000000000 0x0000000000000000), clGetPipeInfo() + 0xB7A bytes(s)
0x00007FF87CBC8AB9 (0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000), clGetPipeInfo() + 0x201F9 bytes(s)
0x00007FF8B1057974 (0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000), BaseThreadInitThunk() + 0x14 bytes(s)
0x00007FF8B334A271 (0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000), RtlUserThreadStart() + 0x21 bytes(s)

[ManuBBXX]

Your number looks great, but Im not on those types of 5700xt.

5700 should be able to do about 57mh without the memory timings, so I don't think 52 is a "great looking number"

please share the setting of a 5700 able to do 57mh without the memory timings or with the memory timings

1) you flash bios mod ( populate higher mem frequencies from the strap of 1500)
2) set volts to 788-810mv, ~1360 core and 1860 mem

bios mod is just setting the same timings from 1500mhz straps to other higher freqs ?
Or is it a bios with other improved settings ?

I did only that : https://mineshop.eu/2020/03/03/bios-modified-rx5700-how-to-edit-timings-10-speed-increase/

[adaseb]

bios mod is just setting the same timings from 1500mhz straps to other higher freqs ?
Or is it a bios with other improved settings ?

I did only that : https://mineshop.eu/2020/03/03/bios-modified-rx5700-how-to-edit-timings-10-speed-increase/

Its just to modify the higher straps with the 1500mhz strap. That's what most people do. However some people update the clock settings so they don't need to use an overclocking program like MSI afterburn on every start-up and update the voltage profiles so they don't have to use WattMan to undervolt on every start up.

So besides the straps for 1600Mhz+ and the core engine/memory clocks and voltage there isn't much to change. Changing anything else can damage the GPU and you shouldn't play around with random values. These days you don't need to modify even the straps, some miners do it for you on the fly. However those don't always work, like in my case so I had to mod the straps.

[scryptr]

FALSE HASHRATE--

This "trick" has been explored by other miners and presented on YouTube and in other blog posts.  The hashrate is inflated at the console, and results in a lower poolside hashrate.  In other words, the miner's console reports a larger hashrate number than is processed at the pool, as the number of invalid shares is also exceedingly high.

I tried many settings using this scheme, and the invalid shares produced subtracts more from the hash than is represented at the miner's console.  It does not work.

--scryptr

hi there how can we Improve this tweak =>


-mi 14
-openclLocalWork 64
-openclGlobalMultiplier 2048.


because it works but 30% SHARE Invalid.

Yes, but 30% invalid represent far less even with the  (so fake) hashrate better