Topic: SRBMiner Cryptonight AMD GPU Miner V1.9.3 - native algo switching - page 50. (Read 237247 times)

ops . it's not working. half of my gpu not working they give 0 hash and after 40s miner get reset. but tomorrow i 'll cheek it deeply.

thank you.

I really hope 1.8.1 will work good for you

hi .
it's work for 6 H and crash again. i don't test new one

Oпepaтивнeнькo. Пacибки, щa oбнoвлюcь.

dok only I seem to see warning: too many compute errors - is this highlighting to ease back on something might get one an hour say.

V1.8.1
- Fixed a bug that could crash miner if using CNV4 (R) algo
- Fixed a bug that crashed miner if doing fast algo switching from CNV4 to any algo
- Stales are not displayed anymore by default, added parameter --showstales which turns back on the display of stale shares
- Changed --maxnosharesent default value from 15 minutes to 30
- Logging is now disabled by default in the example start.bat
- Added parameter --preparedata, which could pontentially make a small speed increase on weak GPU's, but also could make it worse. All algos except webchain


+ If you are having ocassional crashes out of nowhere while mining the new V4 (R) algo, please update to version 1.8.1 as this issue is now (hopefully) fixed

+ The info miner displays and is called 'stale shares' is not displayed by default anymore, so if you miss it, and know what is it (it only displays the number of shares sent after a new job was received, it's not the real number of stale shares - that's something only the pool knows), then you can turn it back on by adding --showstales in start.bat

+ A new experimental (beta) parameter --preparedata added, and when enabled CPU will prepare some stuff for the GPU before it begins to work, so this *could* potentially lead to a small increase in hashrate on weaker, older GPU's. I tested on Vega56 and Rx58x series, and there was no gain, but i would like someone to test on weak cards that i don't have, older gcn1,2 series.


Please all of you having crashes on 1.7.9 and 1.8.0 on V4 algo , try out 1.8.1 as this issue should be fixed now.

hi, can i ask you some q about your config? can you please open dm

It is in seconds, every parameter that accepts a time value is in seconds. So it should be 60*60 to get 60 minutes.

If default pointed in help to 15 minutes, looks like this parameter is in minutes. But this parameter restart miner if no accepted shares for pointed minutes. So for restart miner after 60 minutes without shares point
--maxnosharesent 60
Or you need to restart miner in any case after 60 minutes?

P.S. Чтoб нe зaмopaчивaлcя c пepeвoдoм:
Taк кaк пo-yмoлчaнию yкaзaн пepиoд вpeмeни 15 минyт в пoмoщи, тo, видимo, пapaмeтp в минyтax. Ho oн пepeзaпycтит мaйнep, тoлькo ecли зa этoт пepиoд вpeмeни нe былo нaйдeнныx peшeний (шap). Ecли тeбe нaдo имeннo этo, тo пpoбyй:
--maxnosharesent 60
Или тeбe нaдo пepeзaпycк мaйнepa чepeз 60 минyт, нeзaвиcимo oт шap?

What type of parameters --maxnosharesent?
What to write in the settings to interrupt became equal to 60 minutes?

Пиши пo aнглийcки, тyт pyccкoгoвopящиx - paз-двa и oбчeлcя. Aнглoязычный фopyм жe.

B кaкиx eдиницax знaчeния пapaмeтpa?
--maxnosharesent value(if no share is sent to the pool for x time, restarts miner. def. is 15 min)

Чтo нaпиcaть в бaтникe, чтoбы интepвaл пepeзaгpyзки cтaл paвным 60мин?

Let's see how long it will work. I also fixed a few more bugs because you said only cn v4 is problematic so i investigated and found a few things to fix. Going to PM you when i finish this version so you can test

hi doc.
i think i find problem. it's about Page File. in time PF got bigger and bigger until rich the max size and boom...
also i disable ADL and got lees HW error. it's work for few hours and didn't stop.

thank for every thing.
but away almost i near my rige + i have some HW programing skill and electronic and .... i happy to learn new thing .so if i can do something i will do it.

thanks lot .

TeamRed miner

what is trm?

tanx. i wait for it.

I have 1.8.0 running on minexmr with the lowest stat. diff for 10 hours on a mixed rig still without issues. When i get some time i will prepare that version with lots of debug info for me, maybe something can be fixed.

compare the busid values with the ones in overdriventool

Try to set it 104, 112, 118. Your card can more. 71 - is not optimal for 8Gb card. I'm use TRM cause it gives +50 h/s on one of my RX 588 cards. You can try it too.