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

[hesido]

information is gold.
Reading this thread you get information. 

Some people would still miss it, but this could be in the original post:
"Some users report higher hash-rates with RX580's on CN-Heavy algorithm when running GPU-Z in the background. Mileage may vary."

[1679594367]

1679594367

[1679594367]

1679594367

[RuMiner]

Hello! I'm using both Claymore and SRBMiner rigs, so I've made a small tool to monitor all of them in one Claymore's interface. Here it is: https://bitcointalk.org/index.php?topic=4324950

[doktor83]

information is gold.
Reading this thread you get information. 

Some people would still miss it, but this could be in the original post:
"Some users report higher hash-rates with RX580's on CN-Heavy algorithm when running GPU-Z in the background. Mileage may vary."

I will put this right away in the tips section on first page

[doktor83]

Hello! I'm using both Claymore and SRBMiner rigs, so I've made a small tool to monitor all of them in one Claymore's interface. Here it is: https://bitcointalk.org/index.php?topic=4324950

Very nice, thanks for this 

[LeonManveli]

does any one have a amd 390?   im mining haven  and im seeing around 675-690hashs which seems low,   my 580 with 2 threads is getting 1000-1100 on it.

maybe i don't have the config set up right for it.

usely the 390 beats the 580 in everything  and usely comes really close too the 580 in normal crytonight.   everything else the 390 edges it out.

I have Asus R9 390 and 730~800h/s. I'm mining Haven & Loki.. settings is Core v=-100mv, power limit =0, GPU clock=1100~1150mhz, Memory clock=1500. miner: SRBminer  v1.5.5, intensity=34, double Threads=True. Good Luck.. 

[rig1313]

why r9 390 hashrate is too low comapring to rx 480 8gb?

in ethereum r9 390 does 33 mh/s but in heavy algo it is too slow

this card must do over 1000 h/s in heavy algo

i think that something is wrong

[dingdongtobias]

why r9 390 hashrate is too low comapring to rx 480 8gb?

in ethereum r9 390 does 33 mh/s but in heavy algo it is too slow

this card must do over 1000 h/s in heavy algo

i think that something is wrong

you are right, i think you should make a miner that can do that.

[wgestickley]

@Doc

I have a very strange Problem with Intebsity and the Screchtpad error!

When I use all 11 GPU in the Miner together, than I am able to set Intensity for an RX570 Card to 61, but when I run this card alone i can’t go higher like 51....

Also when I put more RX570 together it’s not possible!

My RIG

1x rx560
7x rx570
3xrx580

Any ideas?

[hamesh]

V1.5.5 (read more)
- No need for all that stuff in start.bat anymore, because watchdog can now restart the whole miner process without it
- Fixed kernels 3 & 4 for heavy algo
- Fixed OCL binary caching for Vega cards
- Added last job received info in stats and API
- Added 'job_timeout' parameter to pools config, it reconnects to pool if no job received for 'job_timeout' time (in seconds), default is 5 min
- Pool reload now checks if current pool is first in reloaded pools list, if not, it switches to first pool

Found a bug.  Watchdog was unable to restart the miner when GPU hung up.  Error returned was could not find file - was clear it couldn't handle a space in a folder name in the path to SRBMiner-CN.exe  May want to either encapsulate the path or use the 8.3 folder name.  I moved it into folder with no spaces, but means won't restart if put it in Program Files or anywhere else with a space.

[doktor83]

V1.5.5 (read more)
- No need for all that stuff in start.bat anymore, because watchdog can now restart the whole miner process without it
- Fixed kernels 3 & 4 for heavy algo
- Fixed OCL binary caching for Vega cards
- Added last job received info in stats and API
- Added 'job_timeout' parameter to pools config, it reconnects to pool if no job received for 'job_timeout' time (in seconds), default is 5 min
- Pool reload now checks if current pool is first in reloaded pools list, if not, it switches to first pool

Found a bug.  Watchdog was unable to restart the miner when GPU hung up.  Error returned was could not find file - was clear it couldn't handle a space in a folder name in the path to SRBMiner-CN.exe  May want to either encapsulate the path or use the 8.3 folder name.  I moved it into folder with no spaces, but means won't restart if put it in Program Files or anywhere else with a space.

true, thanks for the report, i will fix it.

[Larvitar]

is there a "ideal scenario" to run GPU-Z? I noticed a hashrate drop while running.

For who had success doing this: what you did? What version? Running gpu-z.exe -minimized? How can I minetize the procedure?

[hesido]

is there a "ideal scenario" to run GPU-Z? I noticed a hashrate drop while running.

For who had success doing this: what you did? What version? Running gpu-z.exe -minimized? How can I minetize the procedure?

In my personal experience, I had same cards with same mem type and same OC settings, yet some did 950 h/s, some did, 980, some did 1020. When I shut down the miner and I restarted, this time a card would randomly work at 950, although it could be 1030 the previous run. By chance, I read here that someone had their hash rate increase when GPU-Z was turned on and he looped through each card.

In my case, no such thing is necessary. I just fire it up, and it "fixes" hash rates in 1-2 minutes. It fixes the hash rate fall from switching pools in a few minutes in the same manner.

I have 13 cards running, win 10, adrenaline drivers. Maybe it works for setups with many cards, maybe not. Maybe it works with adrenaline drivers, maybe not I can't be sure, but if you are suffering from random hashrates from the same card, this may work for you.

[RuMiner]

@doktor83, today I saw this:

Code:

[2018-05-29 12:00:03] Connected to devfee pool
[2018-05-29 12:00:08] Pool sent a new job (ID: 8286)
[2018-05-29 12:00:51] Pool sent a new job (ID: 8287)
[2018-05-29 12:01:10] Could not login to devfee pool
[2018-05-29 12:01:10] Now mining for user

I have no blocking rules, so you should possibly check your devfee credentials in source

[doktor83]

@doktor83, today I saw this:

Code:

[2018-05-29 12:00:03] Connected to devfee pool
[2018-05-29 12:00:08] Pool sent a new job (ID: 8286)
[2018-05-29 12:00:51] Pool sent a new job (ID: 8287)
[2018-05-29 12:01:10] Could not login to devfee pool
[2018-05-29 12:01:10] Now mining for user

I have no blocking rules, so you should possibly check your devfee credentials in source

could you tell me which algo is this?

[666damien666]

Hello, I have the problem in between, the miner complained that he overheated and then want to restart. then the mining rig crashes. I have 5 Vega 64 in the rig and use version 1.5.5.1, do you have an idea? What's weird is that I have a second rig with 3 x Vega 64 and this is crashes at the same time. Sometimes the whole works without problems 3 days...

does it say "critical shutdown, temperature reach 90C" ?? i had same issue when running SRB 1.5.1 on windows 1803 update. So when i rollback to 1709 windows 10, everything look fine.

have you checked, maybe it really reached 90c ? 

No, it doesnt reached 90c, my GPU is 45c

[666damien666]

my rig have a strange pattern, every 48 hours nontsop mining will get 90 Celcius shutdown , while my GPU not showing any burned effect , anyone know how to fixed this ?

i'm using SRB latest version.

had same issue when running SRB 1.5.1 on windows 1803 update. So when i rollback to 1709 windows 10, everything look fine.

So, if you updated to 1803 probably that's causing the problem, or worse if your gpu's are really getting to 90c.

By default if not set "shutdown_temperature", is set to 90c. I will just disable this default setup, so if "shutdown_temperature" is not explicitly set in config, it wont be enabled and no more problems on 1803.

i not set the "shutdown_temperature" and the rig make a reboot and his freeze, i have Windows Version 1709 :9

[doktor83]

V1.5.6

- Added support for Cryptonight B2N
- Added parameter "min_rig_speed" in config file
- Added parameter "main_pool_reconnect" in config file
- Added optional parameter --resetfans in .bat
- "shutdown_temperature" parameter now must be set to a value > 0 to get activated
- Added 1 min avg hashrate display
- Fixed a bug in watchdog restarting process
- Added support for CPU builtin OpenCL GPU (just for fun)


+ Cryptonight B2N is for Bitcoin2Network coin
+ "min_rig_speed" parameter defines the average hashrate we want to maintain. If 5 min average hashrate is less than the value we defined, restart miner. This option expects you to enter a value in H/S
+  "main_pool_reconnect" minimum value is 180 seconds (3 min), defines how often should miner try to reconnect back to main pool if not mining on it currently. If not set default value is 10 minutes.
It is activated only when it can't connect to main pool from the reconnecting process, so if user changes pool by pressing P, it won't try to connect back to main pool, because user changed the pool himself.
+ Removed auto fan reset on miner exit and put new parameter in bat : --resetfans, which if used resets fans to default settings on miner exit
+ CPU builtin OpenCL GPU

Tested and works on Intel HD 3000 builtin GPU

If user wants to use it has to put --usecpuopencl in bat, and define cpu_conf in config:

"cpu_conf":
[
   { "intensity" : 1, "worksize" : 1, "threads" : 1}
]

you can use "platform" : number , if not used, autodetected

This option is here more mostly for fun, as it hashes slower than a CPU miner would.
The OpenCL kernel is totally unoptimised, so maybe in the future if i get very motivated, i will look into speeding it up, if it's achiveable.

But hey, it's available, you can use it if you want

Tested and works on Intel HD 3000 builtin GPU

[maximumgame]

my rig have a strange pattern, every 48 hours nontsop mining will get 90 Celcius shutdown , while my GPU not showing any burned effect , anyone know how to fixed this ?

i'm using SRB latest version.

had same issue when running SRB 1.5.1 on windows 1803 update. So when i rollback to 1709 windows 10, everything look fine.

So, if you updated to 1803 probably that's causing the problem, or worse if your gpu's are really getting to 90c.

By default if not set "shutdown_temperature", is set to 90c. I will just disable this default setup, so if "shutdown_temperature" is not explicitly set in config, it wont be enabled and no more problems on 1803.

i not set the "shutdown_temperature" and the rig make a reboot and his freeze, i have Windows Version 1709 :9

Same for me. Set "shutdown_temperature" or not in config file the result same. My rig alway auto shutdown. My card very cool at 50c. I will test new ver 1.5.6 today

[doktor83]

my rig have a strange pattern, every 48 hours nontsop mining will get 90 Celcius shutdown , while my GPU not showing any burned effect , anyone know how to fixed this ?

i'm using SRB latest version.

had same issue when running SRB 1.5.1 on windows 1803 update. So when i rollback to 1709 windows 10, everything look fine.

So, if you updated to 1803 probably that's causing the problem, or worse if your gpu's are really getting to 90c.

By default if not set "shutdown_temperature", is set to 90c. I will just disable this default setup, so if "shutdown_temperature" is not explicitly set in config, it wont be enabled and no more problems on 1803.

i not set the "shutdown_temperature" and the rig make a reboot and his freeze, i have Windows Version 1709 :9

Same for me. Set "shutdown_temperature" or not in config file the result same. My rig alway auto shutdown. My card very cool at 50c. I will test new ver 1.5.6 today

Its not set automatically to 90c anymore, so you shouldnt have a prpblem anymore. Report please 

[maximumgame]

my rig have a strange pattern, every 48 hours nontsop mining will get 90 Celcius shutdown , while my GPU not showing any burned effect , anyone know how to fixed this ?

i'm using SRB latest version.

had same issue when running SRB 1.5.1 on windows 1803 update. So when i rollback to 1709 windows 10, everything look fine.

So, if you updated to 1803 probably that's causing the problem, or worse if your gpu's are really getting to 90c.

By default if not set "shutdown_temperature", is set to 90c. I will just disable this default setup, so if "shutdown_temperature" is not explicitly set in config, it wont be enabled and no more problems on 1803.

i not set the "shutdown_temperature" and the rig make a reboot and his freeze, i have Windows Version 1709 :9

Same for me. Set "shutdown_temperature" or not in config file the result same. My rig alway auto shutdown. My card very cool at 50c. I will test new ver 1.5.6 today

Its not set automatically to 90c anymore, so you shouldnt have a prpblem anymore. Report please 

just run 5 minuties in 1.5.6, when one of gpu hash rate = 0 cause  maybe i OC too hight your tool make my rig hang remotely, cant use teamview and vnc. i am going back home to connect my rig to monitor and use mouse to close tool and run again . Can you fix pls ?