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

[misterkit]

Ok who is 132.ip-142-44-240.net? I only see this when running srb...it does not show up on xmrstak and monero ocean. 

[ahsbqt]

your miner not working for r9 390/390x and fury... even old tahiti with awesome miner

This miner normal work with R9 270X. Therefore it will work this 390/390X, fury and tahiti.

I dont know about V1.6.9, but V1.6.8 does not work with R9 390 but it works with Fury(she with HBM memory! If someone worked on optimization for R9 FURY, she would have performed very well on all CN algo!

I have a couple of r9 390 msi gaming working just fine on v1.6.8, they work perfect

[Mashy81]

I have a lot of errors on algo v8, according to HWINFO - does it somehow affect the map? on her condition or better not to pay attention?

Lower clocks or increase voltage a bit. Ideally run with zero errors

[diansari]

cant mine algo Cryptonight - RED (MOX), no accept share.. just pool send job..??

my GPU RX550 4gb

[BS0D]

When using algoswitch on moneroocean with "min_rig_speed" set in each config*.txt, the miner doesn't change it after switching. It uses that parameter from the first started config (which is set in batch file).

[Bojcha]

Ok who is 132.ip-142-44-240.net? I only see this when running srb...it does not show up on xmrstak and monero ocean. 

That is US Department of Defense

[ayiphelmy]

your miner not working for r9 390/390x and fury... even old tahiti with awesome miner

This miner normal work with R9 270X. Therefore it will work this 390/390X, fury and tahiti.

i'm using srbminer with awesome miner... it's not run in 3 rigs with that gpu

[alikas74]

hi @doktor83

can you help with better settings for v8
rx 550 2gb sapphire elpida
rx 560 2gb sapphire elpida
intensity, worksize, fragments?

thank you.

[VulcanR]

Hi all,
After switching to adrenaline 18.5/18.6 my hash shows 370-390 GPUs asus rog rx570. On blockchain driver I got 800h, but miner crashes after 2-3 hours.
Intensity-35, worksize-16, threads-2
Using Srb 1.6.8. win 10 x64
GPU settings voltage- 100, core-1140,memory-1980.

[Velmerk]

Rig with 588+7*Vega56 works well on 1.6.8, HR~14000
after update to 1.6.9 hashrate ~13900, and its not stable.

[dingdongtobias]

Rig with 588+7*Vega56 works well on 1.6.8, HR~14000
after update to 1.6.9 hashrate ~13900, and its not stable.

1.6.9 faster on Vega56 & Vega64 then 1.6.8

[dingdongtobias]

Hi all,
After switching to adrenaline 18.5/18.6 my hash shows 370-390 GPUs asus rog rx570. On blockchain driver I got 800h, but miner crashes after 2-3 hours.
Intensity-35, worksize-16, threads-2
Using Srb 1.6.8. win 10 x64
GPU settings voltage- 100, core-1140,memory-1980.

use the --setcomputemode must turn on

[varusisog]

your miner not working for r9 390/390x and fury... even old tahiti with awesome miner

This miner normal work with R9 270X. Therefore it will work this 390/390X, fury and tahiti.

I dont know about V1.6.9, but V1.6.8 does not work with R9 390 but it works with Fury(she with HBM memory! If someone worked on optimization for R9 FURY, she would have performed very well on all CN algo!

I have a couple of r9 390 msi gaming working just fine on v1.6.8, they work perfect

What is the clock speed and the hash rate?

[VulcanR]

Hi all,
After switching to adrenaline 18.5/18.6 my hash shows 370-390 GPUs asus rog rx570. On blockchain driver I got 800h, but miner crashes after 2-3 hours.
Intensity-35, worksize-16, threads-2
Using Srb 1.6.8. win 10 x64
GPU settings voltage- 100, core-1140,memory-1980.

use the --setcomputemode must turn on

ty, but compute mode is already on. On 18.6 drivers I get hash 0 or 370-390h or it just freezes even with default settings. After each miner closing, amd settings tool restores wat usage.

Timings
For 1:1500 and above.
777000000000000022AA1C00315A6B3CA0550F15B68C1506006AE4007C041420CA8980A90200000 01712262B612B3715

For 2:1500 and above.
777000000000000022339D00CE516A3D9055111230CB4409004AE600740114206A8900A00200312 0150F292F94273116

[Velmerk]

Rig with 588+7*Vega56 works well on 1.6.8, HR~14000
after update to 1.6.9 hashrate ~13900, and its not stable.

1.6.9 faster on Vega56 & Vega64 then 1.6.8

Vega56 ref
1.6.8 auto Intensity = 120
1.6.9 auto Intensity = 112, HR little bit slowly, may be the same after tuning "fragments"...

[dingdongtobias]

Rig with 588+7*Vega56 works well on 1.6.8, HR~14000
after update to 1.6.9 hashrate ~13900, and its not stable.

1.6.9 faster on Vega56 & Vega64 then 1.6.8

Vega56 ref
1.6.8 auto Intensity = 120
1.6.9 auto Intensity = 112, HR little bit slowly, may be the same after tuning "fragments"...

intensiti 112 give smaller hash then intensity 120 ? VERY INTERESTING! 

[Uaciuganadu]

Hey Dok,

I have a strange issue where the miner crashes after multiple hours and this is the end of the log:

Code:

[2018-11-01 19:16:38] algo_switch: Pool requested algo switch to [normalv8]
[2018-11-01 19:16:38] Restarting miner with normalv8 algo
[2018-11-01 19:16:39] Miner version: 1.6.9
[2018-11-01 19:16:39] Windows version: 10.0 build 16299
[2018-11-01 19:16:39] Video driver version: 24.20.11021.1000
[2018-11-01 19:16:40] AMD Platform ID: 1
[2018-11-01 19:16:40] AMD platform FOUND
[2018-11-01 19:16:40] Found 3 AMD devices
[2018-11-01 19:16:40] CPU AES-NI: TRUE
[2018-11-01 19:16:40] GPU0: Radeon RX 580 Series [ellesmere] [8192 MB][I: 58.0][W: 8][T: 2][F: 4][K: 1][BUS: 6]
[2018-11-01 19:16:40] GPU1: Radeon RX Vega [gfx900] [8176 MB][I: 124.0][W: 16][T: 2][F: 4][K: 1][BUS: 9]
[2018-11-01 19:16:40] GPU2: Radeon RX Vega [gfx900] [8176 MB][I: 124.0][W: 16][T: 2][F: 4][K: 1][BUS: 3]
[2018-11-01 19:16:40] ADL is enabled
[2018-11-01 19:16:40] Algo switching mode enabled
[2018-11-01 19:16:40] Algo switch minimum time : 60 seconds
[2018-11-01 19:16:40] CryptonightV8 mode enabled
[2018-11-01 19:16:40] Starting init of mining threads
[2018-11-01 19:16:40] Created OCL context
[2018-11-01 19:16:40] Created OCL command queue for DeviceID 0 (Thread 0)
[2018-11-01 19:16:40] Created OCL input buffer for DeviceID 0 (Thread 0)
[2018-11-01 19:16:40] Created OCL output buffer for DeviceID 0 (Thread 0)
[2018-11-01 19:16:40] Loading normalv8 kernel for DEVICE BUS_ID[6] ...
[2018-11-01 19:16:40] ctx->Program for DeviceID 0 (Thread 0) loaded
[2018-11-01 19:16:40] Created OCL command queue for DeviceID 0 (Thread 1)
[2018-11-01 19:16:40] Created OCL input buffer for DeviceID 0 (Thread 1)
[2018-11-01 19:16:40] Created OCL output buffer for DeviceID 0 (Thread 1)
[2018-11-01 19:16:40] Loading normalv8 kernel for DEVICE BUS_ID[6] ...
[2018-11-01 19:16:40] ctx->Program for DeviceID 0 (Thread 1) loaded
[2018-11-01 19:16:41] Created OCL command queue for DeviceID 1 (Thread 2)
[2018-11-01 19:16:41] Created OCL input buffer for DeviceID 1 (Thread 2)
[2018-11-01 19:16:41] Created OCL output buffer for DeviceID 1 (Thread 2)
[2018-11-01 19:16:41] Loading normalv8 kernel for DEVICE BUS_ID[9] ...
[2018-11-01 19:16:41] ctx->Program for DeviceID 1 (Thread 2) loaded
[2018-11-01 19:16:41] Created OCL command queue for DeviceID 1 (Thread 3)
[2018-11-01 19:16:41] Created OCL input buffer for DeviceID 1 (Thread 3)
[2018-11-01 19:16:41] Created OCL output buffer for DeviceID 1 (Thread 3)
[2018-11-01 19:16:41] Loading normalv8 kernel for DEVICE BUS_ID[9] ...
[2018-11-01 19:16:41] ctx->Program for DeviceID 1 (Thread 3) loaded
[2018-11-01 19:16:41] Created OCL command queue for DeviceID 2 (Thread 4)
[2018-11-01 19:16:41] Created OCL input buffer for DeviceID 2 (Thread 4)
[2018-11-01 19:16:41] Created OCL output buffer for DeviceID 2 (Thread 4)
[2018-11-01 19:16:41] Loading normalv8 kernel for DEVICE BUS_ID[3] ...
[2018-11-01 19:16:41] ctx->Program for DeviceID 2 (Thread 4) loaded
[2018-11-01 19:16:41] Created OCL command queue for DeviceID 2 (Thread 5)
[2018-11-01 19:16:41] Created OCL input buffer for DeviceID 2 (Thread 5)
[2018-11-01 19:16:41] Created OCL output buffer for DeviceID 2 (Thread 5)
[2018-11-01 19:16:41] Loading normalv8 kernel for DEVICE BUS_ID[3] ...
[2018-11-01 19:16:41] ctx->Program for DeviceID 2 (Thread 5) loaded

[Kgonla]

For all of us who are having more or less H/s with 1.6.9 than with previous....
I think H/s shown by the program is inaccurate.

Examples:
Pressed H, shown 992 H/s
Pressed S, it says max H/s have been 985

Set worksize to 1
I press H all the time I want, always showing 0 H/s but getting some hashes
S says H/s 200+

So I think the H/s shown by the S command are OK but the ones shown by H or automatically are not.

[Velmerk]

Vega56 ref
1.6.8 auto Intensity = 120
1.6.9 auto Intensity = 112, HR little bit slowly, may be the same after tuning "fragments"...

intensiti 112 give smaller hash then intensity 120 ? VERY INTERESTING!  

Yes...VERY INTERESTING!
1.6.8 auto Intensity = 120, HR ~14000
1.6.9 auto Intensity = 112, HR ~13900
1.6.9 manual Intensity = 116, HR ~13700
1.6.9 manual Intensity = 120, HR ~13800
1.6.9 manual Intensity = 124, HR ~13600

[doktor83]

For all of us who are having more or less H/s with 1.6.9 than with previous....
I think H/s shown by the program is inaccurate.

Examples:
Pressed H, shown 992 H/s
Pressed S, it says max H/s have been 985

Set worksize to 1
I press H all the time I want, always showing 0 H/s but getting some hashes
S says H/s 200+

So I think the H/s shown by the S command are OK but the ones shown by H or automatically are not.

Max is 'refreshed' every 30 seconds. You probably did not wait 30 seconds to press 'S' after you pressed 'H' and saw 992hs.

'H' shows average hashrate for a very short period, like 8-10 seconds.
'S' shows average for much more, that's why you had there 200hs.