Topic: SRBMiner Cryptonight AMD GPU Miner V1.9.3 - native algo switching - page 3. (Read 237261 times)

Hey, how do i run a specific gpu alone or how do i disable certain gpus at command line?

I'm glad it helped you reduce your power consumption

your 11.5 HR per W is good. Not best  but good.

Best score for samsung is 12,6+HR per wat
for hynix is 11,9 HR/W

I dont have FE.
MY PPT table work with 64 -64LE . FE i dont now. Mybe FE wont biger SOC.


what is the difference in HR between LE and FE.
LE 2300 FE 2350?

Hi Guys!

So now I have tried something again, I have taken over the PowerTableSettings of LIVADA, so it definitely reduces the power consumption significantly. I changed it a bit, too.

I will definitely continue to work with SRB Miner and will continue to test the setup with my Vega64 rig.

Here are my values and settings:

SRB Miner 1.9.3.
Windows 10 (1803)
Adrenaline Driver 18.6.1
ASUS B250 Mining Expert
8x Vega64 - WATER cooled // 4x Vega64 FE
2x ENERMAX - Platimax 1700watts

For algorithm: MONERO/XMR - V4R

The achieved values are currently between: 2300 - 2350 H/s per card @tweak-profile "10"

Total power consumption at: ~ 2410 watts (/ 12 = 200.83 watts per card)
The system runs completely stable, the FE graphics card crashed at 1450Mhz / 866mV - with TWEAK profile "10"
So I reduced my FE cards to 1408Mhz / 866mV and is now stable.
The water-cooled cards are stable at 1450Mhz / 866mV

Here are the following links:

PowerTable-RegFile:
https://drive.google.com/file/d/15Gs-D58J9ZZ1aM0jhmjC7_G3Ud61f541/view?usp=sharing

SRB-Values:
https://drive.google.com/file/d/1Pnqsmdcdq4I7fI8IKC-hmu53FFIdYfox/view?usp=sharing

OverdrivnTool-Values:
https://drive.google.com/file/d/1SVWcCbzj-EFcxsMWihxf6dI_UjBUDufV/view?usp=sharing

Config-Settings:
https://drive.google.com/file/d/1g8v1RNF0ZS9d_bo1AQRzxY_wJz3csAqK/view?usp=sharing

I have already written a post where the cards to ~2450 H/s create, but was not paid attention to the power consumption!
I've been trying to achieve maximum performance here.

Here are the settings:

Topic SRB high performance: (without thinking about power consumption)
https://drive.google.com/file/d/1AiuqbS6gbPW6IsT7p8QmgB6GrfIphHzB/view?usp=sharing

OverdrivnTool-Values:
https://drive.google.com/file/d/1ydngQu040cwMsyW8gdJVLLJI52gH8fIj/view?usp=sharing

You can still try between: 1680Mhz/1000 or 1690Mhz/1000.

I am at your disposal for questions.

greetings
Patrick  

Hi DOKTOR. you chg your IP?

we'll just see that. After fulfilling the DOKTOR wish for everyone to switch from v8 algo to new algo(RX), the doctor is free to close his miner as planned. The SRB miner only supports CN algo, which will all be unprofitable for mining with GPU after the implementation of the RX algorithm. TRM have identical problem, but TRM at least supports other algorithms, not just CN.

me too.
i like to use a miner that gives less and consumes more electricity like SRBMINER because i'm very smart  



you meant a message like this?



After a post like this, REAL DOCTORS can't help it

Agree with you !

Dude i registered myself in this forum just because of you.. only to tell you this.. You are a trash as person and a frustrated person. I just feel pity of you.. and your trash talk, is the reason i will not change miner.

What is the point of showing 3 minute hashrate? We all know that TRM can be faster, but many support Doktor for he's hard work and low fees.
And about your earlier messages on last pages - you really do need to see a doctor, but not the one in this thread.

CONCEAL - CCX - COMPARE
2*vega56+1vega64
identical setup and amd mem tweak.
srbminer  tweak is disable

SRBMINER- 12740HR (595W)
http://imgbox.com/D8LxVHDa
vega56-4200hr
vega64-4400hr

TRM - 13560HR (602W)
http://imgbox.com/HbjsmnsO
vega56-4400hr
vega64-4740hr

can you tell him that he is not talking to me but the doctor just talked to me when i tested him all versions from 1.4.5 to 1.8.0

then he talked too much with the brick, so he became a brick himself

http://imgbox.com/AvptwgFw
http://imgbox.com/STmR2tyD
http://imgbox.com/ylFzQwWd
http://imgbox.com/nBwHxXRo
http://imgbox.com/u664nqAG
http://imgbox.com/tmbfAOih
http://imgbox.com/lEtGdwBA
http://imgbox.com/NTdbzs71
http://imgbox.com/OoGD7WLI
http://imgbox.com/fFI6v9x9
http://imgbox.com/FOlTtRj0

you lier. you not think im trool before:


You only use other multyacount DOKTOR.

first i think you only nob what use rig with 6 gpu in idle 300W but now i see you only retard with mental problem.
if admin ban  noob with mental problem like you all be happy

and doktor ignore me affter LIE all day and not find my 10 post where i say TRM is better minner.

to be continued...

Hi @doktor83 !

I have try to increase virtual memory in Windows as you say. I have 3 Sapphire RX Vega 64 Nitro+ (Use in SRB Miner) and 1 Sapphire Radeon HD 7950 (Use in XMR Stak)
I have upgrade the memory like this : 8192*4 (because I have 4 cards) + 8192 = 40.960 Mo
But nothing change, the error is always here..

P.S : I have this option enable beacause of XMR Stak :
http://www.sqlcoffee.com/images/Tuning05-01.PNG

What can I do now ? What do you need ?


Do you know why every card can't handle the same profil ? I don't understand.. This is the same GPU, same manufacturer, same conception.. There is a bug with my card ? Do I have to return the card in warranty ?
I can't go up to tweak 8 with this card while others can handle tweak 10 with 0 problem.. And this is exactly the same GPUs

i have time and read all last page.
@livada i first thinked you troller but i see now you real have mental problem, no i am sorry for you and we that must read messages on this forum from you it be best for everyone if admins ban you.

@doktor83 you ignore him no need show or prove him anything. he no understand because you say good he is a brick no reasone to talking to a brick.

Click here to download from Mega    
Click here to download from Google Drive

you can't  download? any other site?

DOKTOR- Im wait...

You find my 10 -20 or 40 post where i say TRM is BEST or BETTER?
Just find 10 not 36...

Now you ignore me?

You lie yesterday all day.

You not find my 10 post in 364 page? and who spam your thread?

Only  you.

Patrick,

Could you share the whole config I have to use in OverdriveNTool for getting 2450 H/s


Thx @doktor83, I'll try this !

If you get a "Error CL_MEM_OBJECT_ALLOCATION_FAILURE" try increase virtual memory in windows.
Go with number of GPU's * 8gb + 8 , for example 6gpu * 8gb + 8 = 56gb virtual memory.

Regarding the tweaking profiles, you must find the right profile for your weaker gpu's not every card can handle the same profile number.

You can do that in gpu_conf, for example :

Some ideas :
- if using tweaking profiles, try a lower profile for those gpus
- try lowering intensity for those gpu's
- try lowering memory clock for those gpu's
- if voltage too low, try increasing it
- i don't understand the error message, but it looks to me that miner can't restart because either the .exe has being removed by AV, or it doesn't like the robertÖ in the path.

The screen flicker and memory clocks reset is because of video driver crash.

Can you post your config file ?

Hi Patrick and @doktor83,

About my config :

I live in France
I use : https://minexmr.com/
I have 3 Sapphire RX Vegas 64 Nitro+
I'm on Windows 1903 x64 with 32Go RAM - 256Go SSD
I use Adrenalin 18.6.1 driver

There is my config in SRB Miner :

start.bat :

setx GPU_MAX_HEAP_SIZE 100
setx GPU_MAX_USE_SYNC_OBJECTS 1
setx GPU_SINGLE_ALLOC_PERCENT 100
setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_MAX_SINGLE_ALLOC_PERCENT 100
@echo off
cd %~dp0

cls
for /f "tokens=2 delims==" %%a in ('wmic OS Get localdatetime /value') do set "dt=%%a"
set LOGTIME=%dt:~0,4%_%dt:~4,2%_%dt:~6,2%_%dt:~8,2%_%dt:~10,2%_%dt:~12,2%
set LOGTIME=%LOGTIME: =%
set LOGTIME=%LOGTIME:,=.%.txt

REM LOGGING ENABLED
SRBMiner-CN.exe --config Config\config-normalv4.txt --pools pools.txt --logfile %LOGTIME%

pools.txt :

{
"pools" :
[
   {
      "pool" : "pool.minexmr.com:7777",
      "wallet" : "4897Wc5pexE25twCvQ1bpyBEyMZCrdeARSFqfYQBtPfSD6FZrCW6L6zfAB6PnhJJdaVi2Q3r1AgcViB PrMoZGqRj6hhKLUC.pc-mining-1+272000",
      "password" : "x"
   },
   {
      "pool" : "fr.minexmr.com:7777",
      "wallet" : "4897Wc5pexE25twCvQ1bpyBEyMZCrdeARSFqfYQBtPfSD6FZrCW6L6zfAB6PnhJJdaVi2Q3r1AgcViB PrMoZGqRj6hhKLUC.pc-mining-1+272000",
      "password" : "x"
   }
]
}

config-normalv4.txt :

{
"cryptonight_type" : "normalv4",
"intensity" : 0,
"double_threads" : true,
"target_temperature" : 55,
"gpu_conf" :
[
   { "id" : 0, "intensity" : 124, "worksize" : 16, "tweak_profile" : 10, "threads" : 2},
   { "id" : 1, "intensity" : 124, "worksize" : 16, "tweak_profile" : 10, "threads" : 2},
   { "id" : 2, "intensity" : 124, "worksize" : 16, "tweak_profile" : 10, "threads" : 2}
]
}

Sometimes, I have this error (@doktor83 any idea ?) :

https://drive.google.com/open?id=1E0W1XSJ0ktY61I35wBWLL1KO6FqZ6PdW

It appear when I start to mine with tweak to 10 then I close the miner (or it restart because of GPU crash) then I restart the miner.
Any idea ?

SO I have 2 problems..

- One of my GPU can't mine with tweak profil 10 (invalid shares appears.. and not on the others GPU (WHY ??)) This is exacly the same GPU

- If I use tweak profil 10 I have that problem when the miner has restarted :
https://drive.google.com/open?id=1E0W1XSJ0ktY61I35wBWLL1KO6FqZ6PdW

Thank you very much ! i will try it