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

[sharix80]

It's fixed, testing new version atm if all good i will upload it today

Good news. Thank you for your efficient work.

[1632642766]

1632642766

[1632642766]

1632642766

[1632642766]

1632642766

[1632642766]

1632642766

[doktor83]

V1.7.5
- Added parameter --setup in cmd, which creates basic config files for user in interactive mode
- Added 'old_mode' parameter to config outside gpu_conf
- Added notification if using old kernel
- "worker" set in pools.txt now works for webchain too
- Fixed a few things regarding webchain mining


+ Added parameter --setup that creates basic configuration files for the user in an interactive mode. This is a quick way for beginners to set up everything and start mining.
Also added guided-setup.bat in the package that runs the --setup mode.

+ 'old_mode' was a gpu_conf only parameter until now

+ Webchain is a weird coin
To set the worker-id , use the 'worker' parameter in pools.txt file

[abg00]

V1.7.5

+ Webchain is a weird coin
To set the worker-id , use the 'worker' parameter in pools.txt file

worker work
but no shares on webchain. hashrate on pool very low

[abg00]

1.7.5 bit more on 570\8gb card on tube algo

[MadSciencist]

I haven't mine cryptonight coins for long time. I wanted to switch for haven, but I'm getting only 400h per card (rx580 8GB).
I can set intensity to 270 and get over 1200 per card but SRBMiner is not submitting any hashes then. Anything over 50 intensity is not submitting any hashes (tried worksize 4, 8, 16, 32).
Any ideas?

[zeef]

Same problem reported here since v1.7.1 or more the vega rig became unstable.  My rx580 rig have no problems with v1.7.4.
Vega 64 rig with v 1.7.4 get ramdom restarts, and i increase the mv in all gpus, lower intensity , and same problem!!

Driver version, algo, config file?

Does the miner restart, and work ok after that, or system freezes?
Can you describe ? Maybe i can find out what is happening

It was all ok on 1.7.0 ?

Also how often is this happening?
Is there any specific case you maybe noticed?

Hi,

For me, yes all same version, algo and config only change version. Occured when i migrate from v1.7.0 to v1.7.2, v1.7.3 and v1.7.4.
For me occurs 4-6 hours. System freezes.

I notice that in v1.7.0 the auto intensity is set to 60 in my vega 64 setup, in version 1.7.4 the auto set to 56.
If i put 56 or 60 in v1.7.0 all ok, no crashes for days..

Regards

[carlosmonaco]

Same problem reported here since v1.7.1 or more the vega rig became unstable.  My rx580 rig have no problems with v1.7.4.
Vega 64 rig with v 1.7.4 get ramdom restarts, and i increase the mv in all gpus, lower intensity , and same problem!!

Driver version, algo, config file?

Does the miner restart, and work ok after that, or system freezes?
Can you describe ? Maybe i can find out what is happening

It was all ok on 1.7.0 ?

Also how often is this happening?
Is there any specific case you maybe noticed?

For me I get some card hashing speed at 0h/s ( sometimes all together ), then miners restart by itself but failed to start , the system freeze ( restart needed) . I get that error more often after 1.7.0.
I post the log just above.
Windows 10 , Xmr ( V8) , 1.6.1.
PS : I tried to increase mV and lower clock speed and mem speed a lot but I still get this error since srb 1.7.1 on all rigs.
( No computing error before the crash , 1/24h for some)

[abg00]

I haven't mine cryptonight coins for long time. I wanted to switch for haven, but I'm getting only 400h per card (rx580 8GB).
I can set intensity to 270 and get over 1200 per card but SRBMiner is not submitting any hashes then. Anything over 50 intensity is not submitting any hashes (tried worksize 4, 8, 16, 32).
Any ideas?

write config here
compute mode on for this card?
i=57-59 for 8gb card

[NCarter84]

Dok - if we have all our stuff setup through the config-heavy, config-normalv8, config-mox, etc -- will we need to do anything new?

I have all my config files and pool files setup etc.

[doktor83]

Dok - if we have all our stuff setup through the config-heavy, config-normalv8, config-mox, etc -- will we need to do anything new?

I have all my config files and pool files setup etc.

Well you need to have a start.bat for every config you have.
Or what are you asking ?

[carlosmonaco]

Same problem reported here since v1.7.1 or more the vega rig became unstable.  My rx580 rig have no problems with v1.7.4.
Vega 64 rig with v 1.7.4 get ramdom restarts, and i increase the mv in all gpus, lower intensity , and same problem!!

Im running a Vega Rig, but has no such issues. Which algo are you having these issues?

Hello,
Reference card with blower ? Can you share your settings ?
Best regards

[MadSciencist]

I haven't mine cryptonight coins for long time. I wanted to switch for haven, but I'm getting only 400h per card (rx580 8GB).
I can set intensity to 270 and get over 1200 per card but SRBMiner is not submitting any hashes then. Anything over 50 intensity is not submitting any hashes (tried worksize 4, 8, 16, 32).
Any ideas?

write config here
compute mode on for this card?
i=57-59 for 8gb card

I'm getting 300-400h with intensity set that low. Yes, compute is turned on.

{ "id" : 0, "intensity" : 50, "worksize" : 16, "threads" : 2},
For all cards. It gives me 300-400h.

[shera037]

for nordmann666

pool.anyg.rs:3333 50k static diff

[WhyMe]

Hi guys,
anyone has problems with benchmark options in 1.7.5 ?
Here it simply doesn't work.

Just running benchmark-algos.bat

[2018-12-30 13:13:04] Miner version: 1.7.5
[2018-12-30 13:13:04] Windows version: 10.0 build 17134
[2018-12-30 13:13:04] Video driver version: 24.20.11021.1000
[2018-12-30 13:13:04] AMD Platform ID: 0
[2018-12-30 13:13:04] Created OCL context
[2018-12-30 13:13:04] Using fragments 1 for DeviceID 0 (Thread 0)
[2018-12-30 13:13:04] Using heavy_mode 1 for DeviceID 0 (Thread 0)
[2018-12-30 13:13:05] Created OCL command queue for DeviceID 0 (Thread 0)
[2018-12-30 13:13:05] Created OCL input buffer for DeviceID 0 (Thread 0)
[2018-12-30 13:13:05] Loading [haven] kernel for DEVICE BUS_ID[3] ...
[2018-12-30 13:13:05] ctx->Program for DeviceID 0 (Thread 0) loaded
[2018-12-30 13:13:05] Using fragments 1 for DeviceID 0 (Thread 1)
[2018-12-30 13:13:05] Using heavy_mode 1 for DeviceID 0 (Thread 1)
[2018-12-30 13:13:05] Created OCL command queue for DeviceID 0 (Thread 1)
[2018-12-30 13:13:05] Created OCL input buffer for DeviceID 0 (Thread 1)
[2018-12-30 13:13:05] Loading [haven] kernel for DEVICE BUS_ID[3] ...
[2018-12-30 13:13:05] ctx->Program for DeviceID 0 (Thread 1) loaded
[2018-12-30 13:13:05] ADL initialised successfully!

And then it stops.

Same result on my single VEGA or on a 12 RX570 rig.

No problem with 1.7.2

--runbenchmark seems to do the same not working result

Tested on various algos, it compiles kernels then stops.

Any idea ? Thanks

[doktor83]

Hi guys,
anyone has problems with benchmark options in 1.7.5 ?
Here it simply doesn't work.

Just running benchmark-algos.bat

[2018-12-30 13:13:04] Miner version: 1.7.5
[2018-12-30 13:13:04] Windows version: 10.0 build 17134
[2018-12-30 13:13:04] Video driver version: 24.20.11021.1000
[2018-12-30 13:13:04] AMD Platform ID: 0
[2018-12-30 13:13:04] Created OCL context
[2018-12-30 13:13:04] Using fragments 1 for DeviceID 0 (Thread 0)
[2018-12-30 13:13:04] Using heavy_mode 1 for DeviceID 0 (Thread 0)
[2018-12-30 13:13:05] Created OCL command queue for DeviceID 0 (Thread 0)
[2018-12-30 13:13:05] Created OCL input buffer for DeviceID 0 (Thread 0)
[2018-12-30 13:13:05] Loading [haven] kernel for DEVICE BUS_ID[3] ...
[2018-12-30 13:13:05] ctx->Program for DeviceID 0 (Thread 0) loaded
[2018-12-30 13:13:05] Using fragments 1 for DeviceID 0 (Thread 1)
[2018-12-30 13:13:05] Using heavy_mode 1 for DeviceID 0 (Thread 1)
[2018-12-30 13:13:05] Created OCL command queue for DeviceID 0 (Thread 1)
[2018-12-30 13:13:05] Created OCL input buffer for DeviceID 0 (Thread 1)
[2018-12-30 13:13:05] Loading [haven] kernel for DEVICE BUS_ID[3] ...
[2018-12-30 13:13:05] ctx->Program for DeviceID 0 (Thread 1) loaded
[2018-12-30 13:13:05] ADL initialised successfully!

And then it stops.

Same result on my single VEGA or on a 12 RX570 rig.

No problem with 1.7.2

--runbenchmark seems to do the same not working result

Tested on various algos, it compiles kernels then stops.

Any idea ? Thanks

I will check it, maybe i screwed something 

[abg00]

I'm getting 300-400h with intensity set that low. Yes, compute is turned on.

{ "id" : 0, "intensity" : 50, "worksize" : 16, "threads" : 2},
For all cards. It gives me 300-400h.

check compute mode
and use  "intensity" : 58, "worksize" : 8

also this trouble possible if use afterburner and temp limit low

[symplink]

for nordmann666

pool.anyg.rs:3333 50k static diff

some 100k port?

[shera037]

for nordmann666

pool.anyg.rs:3333 50k static diff

some 100k port?

pool.anyg.rs:6666 120k static diff

pool.anyg.rs:7777 160k static diff

[doktor83]

Webchain is a screwed coin, your pool reported hashrate will be much lower than miners local hashing speed, so don't be surprised.
Just sayin 

[symplink]

We know, basically was a cpu coin. But there pool is ok, not restarting gpus or rig.