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

[livada]

I have been doing some testing on one of my rigs before changing over all of them to SRB.  The initial output I am getting is higher than what I have been getting with Cast, but I did notice something on one of my multi-card rigs.  The hash rate started around 12.5k (with 6 cards) and then it dropped down to all except 2 of the cards doing below 2k each.  I checked the temperature and everything was fine.  I also didn't notice anything else strange.  I stopped the miner and restarted it.  They all came back to the normal numbers.  Any idea what could have caused the issue?

Startup numbers:

Code:

[2018-08-05 18:43:27] GPU0 :    2086 H/s [T: 48c, RPM: 2312, CC: 1408 MHz, MC: 1100 MHz][BUS:7]
[2018-08-05 18:43:27] GPU1 :    2088 H/s [T: 47c, RPM: 2302, CC: 1408 MHz, MC: 1100 MHz][BUS:21]
[2018-08-05 18:43:27] GPU2 :    2086 H/s [T: 49c, RPM: 2333, CC: 1408 MHz, MC: 1100 MHz][BUS:10]
[2018-08-05 18:43:27] GPU3 :    2084 H/s [T: 48c, RPM: 2351, CC: 1408 MHz, MC: 1100 MHz][BUS:3]
[2018-08-05 18:43:27] GPU4 :    2089 H/s [T: 47c, RPM: 2373, CC: 1408 MHz, MC: 1100 MHz][BUS:18]
[2018-08-05 18:43:27] GPU5 :    2087 H/s [T: 44c, RPM: 2332, CC: 1408 MHz, MC: 1100 MHz][BUS:13]
[2018-08-05 18:43:27] Total:    12520 H/s

try my settings for 2.1kh/s

overdrive p7: 1502mhz @ 950mv
              p3 mem: 1100 @ 950mv
              power target 50%
note: my cards are samsung memory


srb config is: { "id" : 0, "intensity" : 120, "worksize" : 8, "threads" : 2},

for each card.

this is vega 56 or 64? bios? and how much electricity use ? 950mv is to much.

vega 64.

950mv can be lowerd depending silicon lotery ;-p

i find 950mv the most stable with no stales or rejects and never any miner crash.

u can tune it depending ur cards and i would suggest to test each card with a lower voltage before putting them all lower than 950mv

my vega64LE use 900mv 1492GPU aand 1095mem . HR 2090.
If i put 950mv and1500GPU/1100mem  i have 2110HR but use 6-8W more from the wall and temp go up 1-2C.
And this is ok if you have free or cheap electricity.
but 20HR give only 0,3$ per month and use 5KW per month(0,5$)

[1675458385]

1675458385

[1675458385]

1675458385

[1675458385]

1675458385

[1675458385]

1675458385

[ikyo]

I have been doing some testing on one of my rigs before changing over all of them to SRB.  The initial output I am getting is higher than what I have been getting with Cast, but I did notice something on one of my multi-card rigs.  The hash rate started around 12.5k (with 6 cards) and then it dropped down to all except 2 of the cards doing below 2k each.  I checked the temperature and everything was fine.  I also didn't notice anything else strange.  I stopped the miner and restarted it.  They all came back to the normal numbers.  Any idea what could have caused the issue?

Startup numbers:

Code:

[2018-08-05 18:43:27] GPU0 :    2086 H/s [T: 48c, RPM: 2312, CC: 1408 MHz, MC: 1100 MHz][BUS:7]
[2018-08-05 18:43:27] GPU1 :    2088 H/s [T: 47c, RPM: 2302, CC: 1408 MHz, MC: 1100 MHz][BUS:21]
[2018-08-05 18:43:27] GPU2 :    2086 H/s [T: 49c, RPM: 2333, CC: 1408 MHz, MC: 1100 MHz][BUS:10]
[2018-08-05 18:43:27] GPU3 :    2084 H/s [T: 48c, RPM: 2351, CC: 1408 MHz, MC: 1100 MHz][BUS:3]
[2018-08-05 18:43:27] GPU4 :    2089 H/s [T: 47c, RPM: 2373, CC: 1408 MHz, MC: 1100 MHz][BUS:18]
[2018-08-05 18:43:27] GPU5 :    2087 H/s [T: 44c, RPM: 2332, CC: 1408 MHz, MC: 1100 MHz][BUS:13]
[2018-08-05 18:43:27] Total:    12520 H/s

try my settings for 2.1kh/s

overdrive p7: 1502mhz @ 950mv
              p3 mem: 1100 @ 950mv
              power target 50%
note: my cards are samsung memory


srb config is: { "id" : 0, "intensity" : 120, "worksize" : 8, "threads" : 2},

for each card.

this is vega 56 or 64? bios? and how much electricity use ? 950mv is to much.

vega 64.

950mv can be lowerd depending silicon lotery ;-p

i find 950mv the most stable with no stales or rejects and never any miner crash.

u can tune it depending ur cards and i would suggest to test each card with a lower voltage before putting them all lower than 950mv

my vega64LE use 900mv 1492GPU aand 1095mem . HR 2090.
If i put 950mv and1500GPU/1100mem  i have 2110HR but use 6-8W more from the wall and temp go up 1-2C.
And this is ok if you have free or cheap electricity.
but 20HR give only 0,3$ per month and use 5KW per month(0,5$)

I am using 1408/925 and 1100/905 for those numbers.  The electricity here isn't all that cheap so the small increase probably isn't worth it.  I also want to keep the temps as low as possible.  I have a dedicated AC unit in that room so the less I can tax it the better.

[livada]

I have been doing some testing on one of my rigs before changing over all of them to SRB.  The initial output I am getting is higher than what I have been getting with Cast, but I did notice something on one of my multi-card rigs.  The hash rate started around 12.5k (with 6 cards) and then it dropped down to all except 2 of the cards doing below 2k each.  I checked the temperature and everything was fine.  I also didn't notice anything else strange.  I stopped the miner and restarted it.  They all came back to the normal numbers.  Any idea what could have caused the issue?

Startup numbers:

Code:

[2018-08-05 18:43:27] GPU0 :    2086 H/s [T: 48c, RPM: 2312, CC: 1408 MHz, MC: 1100 MHz][BUS:7]
[2018-08-05 18:43:27] GPU1 :    2088 H/s [T: 47c, RPM: 2302, CC: 1408 MHz, MC: 1100 MHz][BUS:21]
[2018-08-05 18:43:27] GPU2 :    2086 H/s [T: 49c, RPM: 2333, CC: 1408 MHz, MC: 1100 MHz][BUS:10]
[2018-08-05 18:43:27] GPU3 :    2084 H/s [T: 48c, RPM: 2351, CC: 1408 MHz, MC: 1100 MHz][BUS:3]
[2018-08-05 18:43:27] GPU4 :    2089 H/s [T: 47c, RPM: 2373, CC: 1408 MHz, MC: 1100 MHz][BUS:18]
[2018-08-05 18:43:27] GPU5 :    2087 H/s [T: 44c, RPM: 2332, CC: 1408 MHz, MC: 1100 MHz][BUS:13]
[2018-08-05 18:43:27] Total:    12520 H/s

try my settings for 2.1kh/s

overdrive p7: 1502mhz @ 950mv
              p3 mem: 1100 @ 950mv
              power target 50%
note: my cards are samsung memory


srb config is: { "id" : 0, "intensity" : 120, "worksize" : 8, "threads" : 2},

for each card.

this is vega 56 or 64? bios? and how much electricity use ? 950mv is to much.

vega 64.

950mv can be lowerd depending silicon lotery ;-p

i find 950mv the most stable with no stales or rejects and never any miner crash.

u can tune it depending ur cards and i would suggest to test each card with a lower voltage before putting them all lower than 950mv

my vega64LE use 900mv 1492GPU aand 1095mem . HR 2090.
If i put 950mv and1500GPU/1100mem  i have 2110HR but use 6-8W more from the wall and temp go up 1-2C.
And this is ok if you have free or cheap electricity.
but 20HR give only 0,3$ per month and use 5KW per month(0,5$)

I am using 1408/925 and 1100/905 for those numbers.  The electricity here isn't all that cheap so the small increase probably isn't worth it.  I also want to keep the temps as low as possible.  I have a dedicated AC unit in that room so the less I can tax it the better.

You have very god HR for 1408GPU. 1408mhz  work with 890mv.

[Ballscack]

ANYONE TRIED THE NEW AMD 18.8.1 DRIVER !?

[bruslie]

blockchain driver still less than 1.6.2

cards, algo, settings, difference?

yes sir, sadly  everything the same 100 %

[P00P135]

ANYONE TRIED THE NEW AMD 18.8.1 DRIVER !?

NO! YOU USE KNOWN WORKING DRIVERS NOT THE NEWEST

[Ballscack]

ANYONE TRIED THE NEW AMD 18.8.1 DRIVER !?

NO! YOU USE KNOWN WORKING DRIVERS NOT THE NEWEST

That is a bit a lame answer mate.... i ask who has tested the newest driver to see what happens..

[dopper]

if i use "restart_devices_on_startup" after devcon restart i get CL_OUT_OF_HOST_MEMORY when creating clCreateCommandQueue for DeviceID 0 (Thread 0) error

i've tried the other version of devcon and it didn't help
if i use "restart_devices_on_startup" i get an error, if i remove this string from config then miner starts as usual

doc can you look at this too please?

[SpceGhst]

ANYONE TRIED THE NEW AMD 18.8.1 DRIVER !?

I think you should try it out and let us know how it works!  I didn’t even know there was an 18.8.1 driver.  I’m a big believer in the “if it ain’t broke, don’t fix it” theory...

[hamesh]

Anyone have download for 1.6.4?  I missed it.  1.6.5 won't initialize cards on half my rigs.  Some issue due to compiling everything into the .exe maybe.

[shgth]

Ok so just installed new Radeon 18.8.1 drivers.

Was using 18.7.1 drivers before with SRBMiner 1.6.4 as I play games on this system as well and wanted to use latest drivers for that. 1.6.5 gives slower speed on that driver(which is obvious as it wasn't recommend to use with it).

So now with 18.8.1 SRBMiner 1.6.5 still gives me slow speed with cryptonight but with 1.6.4 I get +10H/s higher hashrate with RX580. No changes for Vega 56.

[doktor83]

if i use "restart_devices_on_startup" after devcon restart i get CL_OUT_OF_HOST_MEMORY when creating clCreateCommandQueue for DeviceID 0 (Thread 0) error

i've tried the other version of devcon and it didn't help
if i use "restart_devices_on_startup" i get an error, if i remove this string from config then miner starts as usual

doc can you look at this too please?

Yes, this will be fixed in the next version, "restart_devices_on_startup" will be removed from the config and will be callable from the .bat, this way it runs before all the init and works OK every time.

[doktor83]

ANYONE TRIED THE NEW AMD 18.8.1 DRIVER !?

I got 1550h/s on Haven algo Vega64. Was 1450 with SBR 1.6.4 So +100h/s. Will try 1.6.5 soon.

Same h/s with 1.6.5

Fast algo - no share found with 18.8.1 and SBR 1.6.5!

These new drivers are really against CN mining, first 18.7.1 and CNV7 no shares found, now 18.8.1 and Fast algo..

[doktor83]

Anyone have download for 1.6.4?  I missed it.  1.6.5 won't initialize cards on half my rigs.  Some issue due to compiling everything into the .exe maybe.

Nope, thats not the problem probably. But you can download older versions from Mega, link on the first page.

[feekiller]

Guys... if you don't like the way the dev fee works move to some other miner.
Dok just ignore everyone complaining about the dev fee.

DEVFEE = 0

Ver 1.6.4

Start.bat

start coin.bat
timeout 6970
taskkill SRBMiner-CN.exe
start start.bat



OR:
Start.bat

start "coin.bat"
timeout 6970
taskkill /F /IM "SRBMiner-CN.exe"
start "start.bat"

[dingdongtobias]

ANYONE TRIED THE NEW AMD 18.8.1 DRIVER !?

I got 1550h/s on Haven algo Vega64. Was 1450 with SBR 1.6.4 So +100h/s. Will try 1.6.5 soon.

Same h/s with 1.6.5

Fast algo - no share found with 18.8.1 and SBR 1.6.5!

These new drivers are really against CN mining, first 18.7.1 and CNV7 no shares found, now 18.8.1 and Fast algo..

Its SBR problems, not driver. Using CastXMR have no problems on Vega cards at all with 18.7.1 and 18.8.1. But Cast not stable for some of my rigs.

There is no SBR, only SRB, is hard to write good??

[abg00]

Guys... if you don't like the way the dev fee works move to some other miner.
Dok just ignore everyone complaining about the dev fee.

DEVFEE = 0

Ver 1.6.4

Start.bat

start coin.bat
timeout 6980
taskkill SRBMiner-CN.exe


OR:

start "coin.bat"
timeout 6980
taskkill /F /IM "SRBMiner-CN.exe"

restart same time as devfee

[doktor83]

Guys... if you don't like the way the dev fee works move to some other miner.
Dok just ignore everyone complaining about the dev fee.

DEVFEE = 0

Ver 1.6.4

Start.bat

start coin.bat
timeout 6980
taskkill SRBMiner-CN.exe


OR:

start "coin.bat"
timeout 6980
taskkill /F /IM "SRBMiner-CN.exe"

restart same time as devfee

restart then come here and complain how unstable and slow is the miner, and demand bug corrections etc.
Don't be surprised if in next versions you won't see when is the fee mined

[christiano88]

Guys... if you don't like the way the dev fee works move to some other miner.
Dok just ignore everyone complaining about the dev fee.

DEVFEE = 0

Ver 1.6.4

Start.bat

start coin.bat
timeout 6980
taskkill SRBMiner-CN.exe


OR:

start "coin.bat"
timeout 6980
taskkill /F /IM "SRBMiner-CN.exe"

restart same time as devfee

restart then come here and complain how unstable and slow is the miner, and demand bug corrections etc.
Don't be surprised if in next versions you won't see when is the fee mined

Don't listen to them, killing the miner and restarting it will only create less profit for the user so if anyone is stupid enough to do it, they are killing their own profits.

I have a request, I know there have been discussions recently, but I am one of those that use profit switchers, and I would like you to kindly make your devfee mine more often and for less seconds to keep at the same level. Hell, I could even be glad with 1% devfee. Right now I am using a cooldown of 2h on my profitswitcher so it doesn't mine devfee for every miner restart and reach 5%+ devfee, but I am missing opportunities to mine the most profitable coin (in 2h, difficulty adjusts). Please, please make it something like every 30 minutes at least (it can mine the first devfee in the first 15 minutes, like it is now). I couldn't care less if you make it not show in the screen...I know you're fair and trustworthy, but please take my feedback into consideration.

[livada]

Guys... if you don't like the way the dev fee works move to some other miner.
Dok just ignore everyone complaining about the dev fee.

DEVFEE = 0

Ver 1.6.4

Start.bat

start coin.bat
timeout 6980
taskkill SRBMiner-CN.exe


OR:

start "coin.bat"
timeout 6980
taskkill /F /IM "SRBMiner-CN.exe"

restart same time as devfee

restart then come here and complain how unstable and slow is the miner, and demand bug corrections etc.
Don't be surprised if in next versions you won't see when is the fee mined

Don't listen to them, killing the miner and restarting it will only create less profit for the user so if anyone is stupid enough to do it, they are killing their own profits.

I have a request, I know there have been discussions recently, but I am one of those that use profit switchers, and I would like you to kindly make your devfee mine more often and for less seconds to keep at the same level. Hell, I could even be glad with 1% devfee. Right now I am using a cooldown of 2h on my profitswitcher so it doesn't mine devfee for every miner restart and reach 5%+ devfee, but I am missing opportunities to mine the most profitable coin (in 2h, difficulty adjusts). Please, please make it something like every 30 minutes at least (it can mine the first devfee in the first 15 minutes, like it is now). I couldn't care less if you make it not show in the screen...I know you're fair and trustworthy, but please take my feedback into consideration.

You use your script (profitswitcher) or any other software for change coin? pls more information for profitswitcher