Topic: SRBMiner Cryptonight AMD GPU Miner V1.9.3 - native algo switching - page 210. (Read 237263 times)

Windows 10 Pro (1709)
AMD Drivers 18.3.4

Config :

{
/* Type can be : normal, normalv7, lite, litev7, heavy, ipbc  */
"cryptonight_type" : "haven",

/* Intensity 0-> auto intensity, or value from 1-300  */
"intensity" : 0,

/* To use 2 threads per card set double_threads to true  */
"double_threads" : true,
}

Log :

[2018-07-01 20:54:27] Miner version: 1.6.1
[2018-07-01 20:54:28] AMD Platform ID: 0
[2018-07-01 20:54:28] AMD platform FOUND
[2018-07-01 20:54:28] Found 6 AMD devices
[2018-07-01 20:54:28] GPU0: Radeon RX 580 Series [Ellesmere] [8192 MB][Intensity 51.0][W: 8][T: 2][K: 1][BUS: 3]
[2018-07-01 20:54:28] GPU1: Radeon RX 580 Series [Ellesmere] [8192 MB][Intensity 51.0][W: 8][T: 2][K: 1][BUS: 5]
[2018-07-01 20:54:28] GPU2: Radeon RX 580 Series [Ellesmere] [8192 MB][Intensity 51.0][W: 8][T: 2][K: 1][BUS: 1]
[2018-07-01 20:54:28] GPU3: Radeon RX 580 Series [Ellesmere] [8192 MB][Intensity 51.0][W: 8][T: 2][K: 1][BUS: 2]
[2018-07-01 20:54:28] GPU4: Radeon RX 580 Series [Ellesmere] [8192 MB][Intensity 51.0][W: 8][T: 2][K: 1][BUS: 6]
[2018-07-01 20:54:28] GPU5: Radeon RX 580 Series [Ellesmere] [8192 MB][Intensity 51.0][W: 8][T: 2][K: 1][BUS: 4]
[2018-07-01 20:54:28] ADL is enabled
[2018-07-01 20:54:28] Cryptonight Haven mode enabled
[2018-07-01 20:54:28] DevFee pools loaded
[2018-07-01 20:54:28] DevFee pool SET
[2018-07-01 20:54:28] DevFee address SET
[2018-07-01 20:54:28] Starting init of mining threads
[2018-07-01 20:54:28] Compiling kernel for DEVICE BUS ID 3 ...
[2018-07-01 20:54:46] Loading kernel for DEVICE BUS ID 3 ...
[2018-07-01 20:54:47] Loading kernel for DEVICE BUS ID 5 ...
[2018-07-01 20:54:48] Loading kernel for DEVICE BUS ID 5 ...
[2018-07-01 20:54:50] Loading kernel for DEVICE BUS ID 1 ...
[2018-07-01 20:54:51] Loading kernel for DEVICE BUS ID 1 ...
[2018-07-01 20:54:52] Loading kernel for DEVICE BUS ID 2 ...
[2018-07-01 20:54:53] Loading kernel for DEVICE BUS ID 2 ...
[2018-07-01 20:54:55] Loading kernel for DEVICE BUS ID 6 ...
[2018-07-01 20:54:56] Loading kernel for DEVICE BUS ID 6 ...
[2018-07-01 20:54:57] Loading kernel for DEVICE BUS ID 4 ...
[2018-07-01 20:54:58] Loading kernel for DEVICE BUS ID 4 ...
[2018-07-01 20:54:59] ADL initialised successfully!
[2018-07-01 20:54:59] Finished init of mining threads
[2018-07-01 20:54:59] Main pool: haven.miner.rocks:4005
[2018-07-01 20:54:59] Gpu watchdog started
[2018-07-01 20:54:59] API disabled
[2018-07-01 20:54:59] json_send: {"method":"login","params":{"login":"hvi1aCqoAZF19J8pijvqnrUkeAeP8Rvr4XyfDMGJcarhbL15KgYKM1hN7kiHMu3fer5k8JJ8YRLKCah DKFgLFgJMYAfnJHXgHSJ5DKiNBvfPa.150000","pass":"w=ROGUE1","agent":"SRBMiner Cryptonight AMD GPU miner/1.6.1","rigid":""},"id":1}
[2018-07-01 20:54:59] json_receive: {"id":1,"jsonrpc":"2.0","error":null,"result":{"id":"852671046905748","job":{"blob":"0303e0c6e4d905a32e31781e4376180d21c487d484c9050d142e92873de727e342319454c658e70 000000091040058cfa39eb12f19ee8b41815fef19d51d991149d237adcd55c50b6b85e106","job_id":"236919075889465","target":"d96f0000"},"status":"OK"}}
[2018-07-01 20:54:59] Connected to haven.miner.rocks:4005
[2018-07-01 20:55:00] sock_ready: User logged in
[2018-07-01 20:55:00] pool_have_job: Pool sent a new job (ID: 236919075889465)
[2018-07-01 20:55:11] json_receive: {"jsonrpc":"2.0","method":"job","params":{"blob":"0303e0c6e4d905a32e31781e4376180d21c487d484c9050d142e92873de727e342319454c658e70 0000000d67680d5be05fa737cdc2d13ae3919bcb835d948037987bf788ce1f6b944f58206","job_id":"937923625437735","target":"d96f0000"}}
[2018-07-01 20:55:11] pool_have_job: Pool sent a new job (ID: 937923625437735)


Then the miner crashes

You right, my bad. My .bat doesn't use the good directory

That is not 1.6.1. Download 1.6.1 from here : http://www.srbminer.com/downloads/SRBMiner-CN-V1-6-1.zip


When using an unkown algo, this is what you get on 1.6.1 :
cryptonight_type can be : [normal, normalv7, lite, litev7, heavy, bittubev2, artocash, alloy, marketcash, b2n, stellitev4, haven, fast]

How can we test this with testnet pool ?
My 1.6.1 doesn't seems recognize "cryptonight_type" : "bittubev2",

{
/* Type can be : normal, normalv7, lite, litev7, heavy, ipbc, artocash, alloy, marketcash, b2n  */
"cryptonight_type" : "bittubev2",
"intensity" : 0,
"double_threads" : true
}

Miner says :
cryptonight_type can be : [normal, normalv7, lite, litev7, heavy, ipbc, artocash, alloy]

windows version, driver version, config file and log would be more helpful.

Same problem here.
1.6.0 works on Haven, 1.6.1 not.

It seems there's an issue with Haven Algorithm and V1.6.1

The miner crashes.

Tested on 2 rigs, same problem

No issue with V1.6.0

BITTUBE FORK INCOMING (block 110.000)

Get ready (v1.6.1) :

Example for Vega cards

config.txt :

{
"cryptonight_type" : "bittubev2",
"intensity" : 56,
"double_threads" : true
}

pools.txt :

{
"pools" :
[
   {"pool" : "mining.bit.tube:15555",
    "wallet" : "your-wallet",
    "password" : "x"
    }
]
}

config.txt :

{
"cryptonight_type" : "stellitev4",
"intensity" : 0,
"double_threads" : true
}

pools.txt :

{
"pools" :
[
   {
   "pool" : "communitypool.stellite.cash:6688",
    "wallet" : "your-wallet",
    "password" : "x"
    }
]
}

Those questions can never be prevented, no matter what i do to the miner.
All i can say is that the miner is not showing fake or higher values just to make the miner happy.
I am sure everyone knows what is the best for him, and if it's switching to another mining software, i am ok with that, i can't force anyone to use SRBMiner

Doctor, maybe you add in statistics effective speed (accepted by pool)? To prevent questions about differenece between miner speed and pools speed.

https://github.com/xmrig/xmrig-proxy/blob/dev/doc/STRATUM_EXT.md#14-algorithm-names-and-variants

C'mon dok.. DO IT
You can...

I hope you are cooking up something back there )

guys, help me to configurate SRB 1.6.1. for StelliteV4
I always have crushes miner

what are your fine tune values?
if you increase the intensity too much then your hashrate might fluctuate a lot and the miner will become slower overall, despite showing higher hashrates.

My 8x RX470/570 rig is showing 7400h/s hash rate, but unfortunately, i guess that SRBMiner is not showing the right value, because i mined for 24h and the pool show about 6700h/s.
I tested 3x different pools and coins, looks like my fine tune with intensity doesn't work in the real world, only show increased numbers at the miner.

Dok, do you know why rig with 13 GPU (RX570) gets stuck on "loading kernel" all the time?

I need to restart the rig 3 or 5 times (sometimes even more) until I can finally load all kernels...

Tried some more algos and seems to be the same issue for Haven also. The miner crashes when starting to send result back to pool. Also tried my r9 290s and r9 380s separately and they both crash with these algos. No problems with regular v7 or heavy.

Can anyone with these cards try if their miner also crashes when mining stellite or haven?

im waiting too

I have been developing this after forking from sebseb's work, added pushbullet notifications / dropdown for easy worker selection / market value (tradeogre only, so far, altex will come next) / coin switching for same algo coins.

The advantage of using this coin switch, even manually, is that there's no hashrate drop between pools/coins.

Currently the auto coin switching is dumb, is based on market value, but you'll be able to fine tune the selection process by using your speculative expectations: I'll be adding "Ignore market value above" "Ignore market value below" and "Override market value" options in GUI for easy and dynamic setup so you can mine based on difficulty of coin without being affected by fluctuating markets.

Give it a go from https://github.com/hesido/cryptonote-switching-relay-proxy

Planned features:

• Fine tune auto-switch logic
• Wait for round end before switch (to not lose shares in longer rounds in pplns, and ethical mining for PPS :=) )
• Aggregate notifications

Fantasy feature:
If SRB miner supports coin switching by XMR-Proxy standards, do different algo coin switching.

It would be a nice "selling" point for SRB miner, wouldn't it

Notes:
The proxy is tested to work with XMR-Stak, and should be able to work with other miners too.

In your miner, you just point to localhost or the ip that you host the proxy but in case of bugs or connection errors, make sure you define backup pools in your miner also )

Use at your own risk, monitor the software often, this is Work In Progess software.

I took the time and tested latest STAK 2.4.5, to see is hashrate on pool = hashrate in miner like you say.


https://i.imgur.com/oo57Vi3.jpg


As you can see on the image, STAK running for more than 6 hours.
Miner shows 1775hs as average for 15m.
Pool shows 1674hs as average for 6 hours.

If my math is correct : 1775 - 1674 = 101 hs difference.
If i had 6 vegas that is : 6 * 101hs = 606 hs dissapeared somewhere.

Look at the graph, that is what is called luck.
Sometimes 50 found shares, then only 6.

What i want to show you with this is that you can never have the same hashrate ALL THE TIME in your miner as on your pool.
Sometimes it will be less, sometimes more, and that is totally normal.