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

[camchimonline]

Hi everybody,
How to fix compute errors?
Thanks

[UnclWish]

Hi everybody,
How to fix compute errors?
Thanks

Lower overclocking, rise voltage if undervolted. Use native bios.

[Kgonla]

Cool!
With new version my RX 580 4GB give me about 10 H/s more & my 580 8GB about 15 more than with previous version.
Not tested fragments & bralock yet, so maybe it can improve a bit more.

Ups! My first test gave me +10 H/s with 4 gb card but now is very unstable reaching up to -90 H/s than before, tested changing intensity but no change.
The 8 gb is running fine.

write algo
write config

no mindreaders!

Sorry, you are right.

Here are my tests after 40-50 tries:

Drivers: Modified blockchain drivers for W7, not using precompiled kernels from 1.6.9
Algo: V8
Config at this time:
"gpu_conf" :
[
   { "id" : 0, "intensity" : 100, "worksize" : 8, "threads" : 2, "fragments" : 16, "bralock" : 50}, RX 580 8gb, connected to screen
   { "id" : 1, "intensity" : 61, "worksize" : 8, "threads" : 2, "fragments" : 16}, RX 580 4gb, with riser

V 1.6.8

Same actual, above config, without the fragments and bralocks
8gb card, very unstable, giving an average of about 935 H/s
4gb card, very stable, giving 1002-1005 H/s

V 1.6.9

8gb card, after playing with fragments & bralock I get 975 quite stable, with some peaks and valleys, this is about +40 H/s from 1.6.8

4gb card, here are the problems:
Rarely, I start program & have 1012-1015 H/s it is +10 H/s
Other times it start at about 950 H/s (+- 10 or 20 H/s, sometimes up to -40 H/s), after 2 to 6-8 minutes at some point the hash rate increases reaching above 1010 & stays there for the time I tested
Often, it starts about 950, like explained above but never goes higher than 960-965 H/s
I tested intensities from 50 to 62 & the same happens, also played with fragments, memory speed & bralock but no changes

Thank you.

[jazz1984]

Hi guys. Use srb 1_6_9 on my 7970 and have a less hashrate compared to 1_6_8. The problem is that the value of "fragments"  does not change, it is always 4. Heres my config.txt:
{
"cryptonight_type" : "normalv8",
"intensity" : 57,
"worksize" : 16,
"double_threads" : false,
"fragments" : 8
}


Where is my mistake? Thank you.

[Fanpant]

Hi guys. Use srb 1_6_9 on my 7970 and have a less hashrate compared to 1_6_8. The problem is that the value of "fragments"  does not change, it is always 4. Heres my config.txt:
{
"cryptonight_type" : "normalv8",
"intensity" : 57,
"worksize" : 16,
"double_threads" : false,
"fragments" : 8
}


Where is my mistake? Thank you.

What is your hash rate for the 7970? What is the clock speed?

[doktor83]

Hi guys. Use srb 1_6_9 on my 7970 and have a less hashrate compared to 1_6_8. The problem is that the value of "fragments"  does not change, it is always 4. Heres my config.txt:
{
"cryptonight_type" : "normalv8",
"intensity" : 57,
"worksize" : 16,
"double_threads" : false,
"fragments" : 8
}


Where is my mistake? Thank you.

You can't define fragments there.
Do it like this :

Code:

"gpu_conf" : 
[ 
	{ "id" : 0, "intensity" : 57, "worksize" : 16, "threads" : 2, "fragments" : 8}
]

[jazz1984]

Hi guys. Use srb 1_6_9 on my 7970 and have a less hashrate compared to 1_6_8. The problem is that the value of "fragments"  does not change, it is always 4. Heres my config.txt:
{
"cryptonight_type" : "normalv8",
"intensity" : 57,
"worksize" : 16,
"double_threads" : false,
"fragments" : 8
}


Where is my mistake? Thank you.

What is your hash rate for the 7970? What is the clock speed?

Last stable results on v8 was 470-475 H/s. Driver 18.6.1, win 7 x64, srbminer_1_6_8 intencity 57, worksize 16, no doublethreads. Clocks 1100 core / 1500 memory.

[jazz1984]

Hi guys. Use srb 1_6_9 on my 7970 and have a less hashrate compared to 1_6_8. The problem is that the value of "fragments"  does not change, it is always 4. Heres my config.txt:
{
"cryptonight_type" : "normalv8",
"intensity" : 57,
"worksize" : 16,
"double_threads" : false,
"fragments" : 8
}


Where is my mistake? Thank you.

You can't define fragments there.
Do it like this :

Code:

"gpu_conf" : 
[ 
	{ "id" : 0, "intensity" : 57, "worksize" : 16, "threads" : 2, "fragments" : 8}
]

Thank you, now it works. As i understand value of fragments can be equal only to this values:0,1,2,4,8,16,32,64,128 ?

[doktor83]

Hi guys. Use srb 1_6_9 on my 7970 and have a less hashrate compared to 1_6_8. The problem is that the value of "fragments"  does not change, it is always 4. Heres my config.txt:
{
"cryptonight_type" : "normalv8",
"intensity" : 57,
"worksize" : 16,
"double_threads" : false,
"fragments" : 8
}


Where is my mistake? Thank you.

You can't define fragments there.
Do it like this :

Code:

"gpu_conf" : 
[ 
	{ "id" : 0, "intensity" : 57, "worksize" : 16, "threads" : 2, "fragments" : 8}
]

Thank you, now it works. As i understand value of fragments can be equal only to this values:0,1,2,4,8,16,32,64,128 ?

Yes, that's right.

You said you have less hashrate with 169 compared to 168. Is that what the exact same settings?
What numbers were you getting on 168, and now on 169 ?

[playfast]

Doktor, kudos on your fabulous miner and keep up the great work!
Question, is there a way to see which particular GPU found results/shares?

[nordmann666]

is there a more detailed log level possible?

miner stops mining until i see that there is no action since many minutes...have to press key and than i got error messages with many invalid shares following

[dingdongtobias]

is there a more detailed log level possible?

miner stops mining until i see that there is no action since many minutes...have to press key and than i got error messages with many invalid shares following

if win10 don't click in the miner window that halt miner execution, you have that problem 100%

[nordmann666]

is there a more detailed log level possible?

miner stops mining until i see that there is no action since many minutes...have to press key and than i got error messages with many invalid shares following

if win10 don't click in the miner window that halt miner execution, you have that problem 100%

it stopps without clicking in the cmd - and i disabled quick edit mode

[jazz1984]

Hi guys. Use srb 1_6_9 on my 7970 and have a less hashrate compared to 1_6_8. The problem is that the value of "fragments"  does not change, it is always 4. Heres my config.txt:
{
"cryptonight_type" : "normalv8",
"intensity" : 57,
"worksize" : 16,
"double_threads" : false,
"fragments" : 8
}


Where is my mistake? Thank you.

You can't define fragments there.
Do it like this :

Code:

"gpu_conf" : 
[ 
	{ "id" : 0, "intensity" : 57, "worksize" : 16, "threads" : 2, "fragments" : 8}
]

Thank you, now it works. As i understand value of fragments can be equal only to this values:0,1,2,4,8,16,32,64,128 ?

Yes, that's right.

You said you have less hashrate with 169 compared to 168. Is that what the exact same settings?
What numbers were you getting on 168, and now on 169 ?

Last stable results on v8 was 470-475 H/s. Driver 18.6.1, win 7 x64, srbminer_1_6_8 intencity 57, worksize 16, no doublethreads. Clocks 1100 core / 1500 memory.

With Srb_1_6_9 with same settings and drivers (only new version of miner) i got 465 H/s. The only i change the value of fragments. 1,2 shows very low hashrates 150-250, and the best for my card is "fragments" :16.

[UAGet]

RX 560 2Gb
1.6.8 - 526 h/s
1.6.9 - 490 h/s 

[doktor83]

RX 560 2Gb
1.6.8 - 526 h/s
1.6.9 - 490 h/s 

have no fear, just report here 

Anyone with 4gb 560 can report a change in hashrate?
I have only 4gb card and there nothing changed from 168->169 when testing.

[digitalunix]

doktor83 Maybe SRBMiner mining cryptonight-webchain soon?

[Kgonla]

RX 560 2Gb
1.6.8 - 526 h/s
1.6.9 - 490 h/s 

have no fear, just report here 

Anyone with 4gb 560 can report a change in hashrate?
I have only 4gb card and there nothing changed from 168->169 when testing.

You changed something that makes 8gb cards to do better hash rate but worst for 4gb & below, at least in some cases.

[Uaciuganadu]

V1.6.9
- Algo switching without miner restart
- Benchmark everything from algos.txt and just paste the hashrates (--benchmarkalgos)
- Added number of algo switches to API and stats
- Added cmd parameter --resetmineronalgoswitch which forces miner to reset on algo change
- Changed minimum --algoswitchmintime from 3 min to 30 sec, default is still 600 sec
- Fixed API pool stats when not connected to a pool
- Added cmd parameters --cjobtimeout, --cmaxdifficulty, --ckeepalive
- When critical temp is reached, ADL temp is now logged before shutdown
- Notification on too many compute errors (set with --gpuerrorsalert value)
- Added a performance tuning parameter 'fragments' for gpu_conf and cmd
- Fixed a bug that could crash miner on shutdown


+ Algo switch without miner restart is here finally. This is now the default mode. If you want to use the 'old fashioned' switch where the miner restarts itself use --resetmineronalgoswitch in .bat.
The only reason i can see to use the switching with restart is for Vegas, because they tend to drop hashrate on algo switch, and only --resetvega can help to regain full speed again.
Check out here how to setup algo switching

+ Some people were too lazy to benchmark algo by algo to get the 'hashrate' values for algos.txt, so there is a --benchmarkalgos now , that benchmarks the algos set in algos.txt, and creates benchmark_results.txt, from where you just have to take and paste the values into algos.txt

+ The --algoswitchmintime minimum value can now be 30 seconds, altough Monero Ocean really likes to disregard this setting

+ There is on-screen and log notification now when you have too much compute errors (default is 15). You can set this value with --gpuerrorsalert value

+ (re)added fragments parameter for some performance tuning. This parameter was set internally, but my tests show it can REALLY be different for even the same type of GPU, but different vendor.
So now everyone can find his 'optimal' value for this parameter. It differs for algos and also for the amount of VRAM, so a value that is good for 4GB cards maybe isn't the best value for 8GB cards.

Hey Dok,

Indeed on Vega the miner kinda brakes and can not handle the change without --resetmineronalgoswitch . This also affects the --benchmarkalgos as the function ignores the --resetmineronalgoswitch in my tests.

Is there a way for the above to get fixed or make the reset of the miner faster? Also can you make it so that the bench performs the restart?

[goran89]

RX 560 2Gb
1.6.8 - 526 h/s
1.6.9 - 490 h/s 

have no fear, just report here 

Anyone with 4gb 560 can report a change in hashrate?
I have only 4gb card and there nothing changed from 168->169 when testing.

You changed something that makes 8gb cards to do better hash rate but worst for 4gb & below, at least in some cases.

I have better hashrate on 4 gb cards with 1.6.9 on all algos.
3x RX570+ 2x RX 580 = 15 more hashes on v8
                                    35 more hashes on v7
                                    100 more hashes on fast