SRBMiner Cryptonight AMD GPU Miner V1.9.3 - native algo switching

[doktor83]

What intensity/worksize/threads combo are you recommending on RX 570's 4GB for the new v8?

try 57/16/2

[1675453430]

1675453430

[1675453430]

1675453430

[1675453430]

1675453430

[1675453430]

1675453430

[1675453430]

1675453430

[doktor83]

Nice Dok! Can confirm ~2035 H/s for a V64 on V8. ~2135 H/s on current V7. V8 Only works using your pre-compiled kernels though. Running 18.6.1 drivers which are recommended, strange that they can't compile for V8. Just wanted to pass the info along, thx for all your wonderful work, best CN miner in my book by a long shot!

Do you get an error when trying to compile, or ?

[doktor83]

Could anyone help me,when i start the miner with my r9 290 i get the following error:

[2018-10-09 14:48:01] Error CL_LINK_PROGRAM_FAILURE when calling clLinkProgram [ctx->Program] for DeviceID 0 (Thread 0)
[2018-10-09 14:48:01] Error CL_INVALID_PROGRAM when calling clGetProgramBuildInfo for length of build log output for DeviceID 0 (Thread 0)

On top where you have something like : GPU0: Radeon RX 580 Series [ellesmere] [8192 MB][Intensity 56.0][W: 8][T: 2][K: 1][BUS: 2]

What do you have, [K: 1] or [K: 2] ?

Also driver version?

[doktor83]

It looks like the 18.3.4 drivers dont like my implementaion of V8 with the default worksize of 16 - tested on Vega56.
This is why some of you get those 500-600-700hs speeds probably.
On 18.6.1 it is OK, i will also test 18.5.1 , 18.5.2

EDIT:

From 18.5.1 is OK.
I will check why on 18.3.4 isn't but it is not high priority.

Ok, so checked on 18.3.4 and it looks like its not my V8 implementation that is the problem, but something else.


Anyone out there can try 1.6.8 with Vega56 or 64 on Blockchain drivers?

[samet_9511]

is it important worksize in config ?

i tried 8 and 16 and 12
i think no any effect. any idea ?

[doktor83]

is it important worksize in config ?

i tried 8 and 16 and 12
i think no any effect. any idea ?

it is important.
Which algo?
Which gpu?

[ZaGMaN]

Hi doktor83
Starting with version 1.6.5, on the normalv7 algorithm for R7 370 uses kernel 2 and give me 360h/s
On version 1.6,4 and earlier, these cards work on kernel 1 and give me 440h/s
What to do when the algorithm changes to normalv8 ?

[doktor83]

Hi doktor83
Starting with version 1.6.5, on the normalv7 algorithm for R7 370 uses kernel 2 and give me 360h/s
On version 1.6,4 and earlier, these cards work on kernel 1 and give me 440h/s
What to do when the algorithm changes to normalv8 ?

you can give me access to a machine you have that card installed in, and i can try to rework kernel1 to work with R7 too

[jazz1984]

Anyone still mining with the older Radeon 7970/280X or the R9 290 GPUs?

Was wondering what config you guys are using and what stability you guys get with this software, I am looking for an alternative to the Claymore XMR miner which will become obsolete when XMR forks on Oct 18th.

I tried to run the srmbminer on my 7970 with settings --cgpuintencity 15 --cgpuworksize 8 -cgputhreads 2, at the stock frequencies (925/1375) I got 360H / s on v7, its less than old claymore 11.3 (450)... but clay not support v8

PS Guys when need to change algo for Monero mining? Any dates?

[sxemini]

is it important worksize in config ?

i tried 8 and 16 and 12
i think no any effect. any idea ?

it is important.
Which algo?
Which gpu?

For me too. I have a Rig with 6x RX550 4G. I mine Monero (CN V7) with 34 Intensity. 8 or 12 worksize change nothing in hashrate. I get arround 2900 hahes with this config.

[doktor83]

is it important worksize in config ?

i tried 8 and 16 and 12
i think no any effect. any idea ?

it is important.
Which algo?
Which gpu?

For me too. I have a Rig with 6x RX550 4G. I mine Monero (CN V7) with 34 Intensity. 8 or 12 worksize change nothing in hashrate. I get arround 2900 hahes with this config.

550 4g should be ok on V7 with : 27/12/2
550 4g should be ok on V8 with : 28/16/2

higher intensity doesnt mean more hash always

[samet_9511]

is it important worksize in config ?

i tried 8 and 16 and 12
i think no any effect. any idea ?

it is important.
Which algo?
Which gpu?

stellitev4

RX550 2GB
RX470 4GB
RX570 4GB

it will be wonderful if you write worksize values for the different graphic cards in readme.txt

[sxemini]

is it important worksize in config ?

i tried 8 and 16 and 12
i think no any effect. any idea ?

it is important.
Which algo?
Which gpu?

For me too. I have a Rig with 6x RX550 4G. I mine Monero (CN V7) with 34 Intensity. 8 or 12 worksize change nothing in hashrate. I get arround 2900 hahes with this config.

550 4g should be ok on V7 with : 27/12/2
550 4g should be ok on V8 with : 28/16/2

higher intensity doesnt mean more hash always

With 27 for example i get only 2800 and if i increase the intensity step by step, then the hashrate increase step by step. So 27 is not the best value for me.
Hashrate is the same with worksize 8 and worksize 12.

[doktor83]

is it important worksize in config ?

i tried 8 and 16 and 12
i think no any effect. any idea ?

it is important.
Which algo?
Which gpu?

stellitev4

RX550 2GB
RX470 4GB
RX570 4GB

it will be wonderful if you write worksize values for the different graphic cards in readme.txt

I can't  because with the same card something that works for Joe, does not mean it will work for you.
You cards maybe cant handle the intensity Joe's cards can.

[samet_9511]

i know how to set intensity and threads.
i need to know only worksize. I always set 8 to worksize.

I have a farm with 35 rigs: RX550 2GB, RX470 4GB and RX570 4GB

[doktor83]

i know how to set intensity and threads.
i need to know only worksize. I always set 8 to worksize.

I have a farm with 35 rigs: RX550 2GB, RX470 4GB and RX570 4GB

These are my conclusions from lot of testings i made :

Rx550 is always 8 or 12 for 'old' algos, V8 is 16, that you have to test because again (and it is interesting), some say they get better results with 8, but others with the same card say they get better results with 12.
Rx470 and Rx570 is always 8 for 'old' algos, V8 is 16.

[doktor83]

OK, so after testing with 18.3.4 and V8 here is the conclusion :

Don't use 18.3.4 if you want to mine V8!

It's something in the V8 code that driver doesn't like, and probably not something i can fix.
I will put this info on the first page.

[GKumaran]

is it important worksize in config ?

i tried 8 and 16 and 12
i think no any effect. any idea ?

it is important.
Which algo?
Which gpu?

stellitev4

RX550 2GB
RX470 4GB
RX570 4GB

it will be wonderful if you write worksize values for the different graphic cards in readme.txt

There is a rough template on the first page for some cards and algos

[doktor83]

is it important worksize in config ?

i tried 8 and 16 and 12
i think no any effect. any idea ?

it is important.
Which algo?
Which gpu?

stellitev4

RX550 2GB
RX470 4GB
RX570 4GB

it will be wonderful if you write worksize values for the different graphic cards in readme.txt

There is a rough template on the first page for some cards and algos

Also a google document maintained by users, where one can see tips for settings on various cards and algos :

https://docs.google.com/spreadsheets/d/1pH326fwnYWC5Jef2FdF4KnJ4CLj0cwDDtGjrlgekk1A/edit#gid=514064850

[StaL89]

have you tried auto intensity for 550?

also it would be nice if you could write your settings

no
for 2gb 1.6.8
550
   { "id" : 6, "intensity" : 27,  "worksize" : 12, "double_threads" : true},
 GPU6 :    504 H/s [T: 49c, RPM: 0, CC: 1050 MHz, MC: 1950 MHz][BUS:4]
560
   { "id" : 7, "intensity" : 27,  "worksize" : 12, "double_threads" : true},
GPU7 :    488 H/s [T: 53c, RPM: 2635, CC: 1050 MHz, MC: 1850 MHz][BUS:2]

for 550 4gb 1.6.7
   { "id" : 0, "intensity" : 54, "double_threads" : true, "worksize": 12},
 GPU0 :    416 H/s [T: 51c, RPM: 2833, CC: 1090 MHz, MC: 2050 MHz][BUS:1]
same 550\4 in 1.6.8 only 380-390 h\s

Haпишитe ecли нe cлoжнo кaкиe y вac 550? пaмять? и ecли oчeнь нe cлoжнo тaйминги)