search

miniZ-miner / miniZ

37 stars 1 forks source link

miniZ v2.0c SOL rate decrease on FLUX #13

Open thaneb opened 1 year ago

thaneb commented 1 year ago

Hi there, I don't know if I'm the only one to see a SOL rate decrease with the latest v2.0c, but I prefer to post my observation in case of. My 2 rigs started to use the v2.0c after being updated to hiveos 0.6-220@230127, and if rtx3070, 3060ti and 2070super lost few 0.2/0.3 sol, my rtx 2060super lost 4.5/5 SOL ! at first i suspected an issue with the card, as it wasn't using the whole 125W power limit applied by hiveos, but after many tries, i just downgrade the miniz to v2.0b and "banco", i recovered instantaneously my full SOL rate. mining on FLUXpool. thanks for your great work anyway !

Miiichal commented 1 year ago

Some things happens, cause i have 9 rigs (6 cards inside each). All the same configuration,m all the same types of cards etc. Sometimes two rigs are mining Zhash and Zelhash around 54 Sols instead of 71. Its because of unfortunatelly set the power to extra low settings (around 98W) Its since the version 2.0c

miniZ-miner commented 1 year ago

Hi, thanks for the feedback. Could you check if this issue is fixed with latest version v2.0c3? Cheers

thaneb commented 1 year ago

Sorry for the delay. I checked the latest 2.0c3, in the first minutes of mining, my 2060s seems to reach its highest SOLrate, and then it falls, but it's better than with the previous 2.0c (now i got a stable 50, whereas it was dropping to 47 with 2.0c & 2.0c2). I also tried the --autoclocks for dual mining FLUX+ZIL (Fluxpool +EZILE; and remove all hiveos overclocking settings), here's what i got: **** miniZ v2.0c3 **** Number of miniZ CUDA>=[.] using driver CUDA[11.60] devices found: 4 Driver: 510.85 Algo: EQ[125,4] [smart-pers] Pool#0: user[FLUX_WALLET.Correct-worker-name-with-%worker_name%] server[asia-flux.fluxpools.net] port[7003] ssl[yes] pers[ZelProof] Pool#1: user[ZIL_WALLET.Correct-worker-name-with-%worker_name%] server[asia.ezil.me] port[24443] ssl[no] pers[zil] Telemetry: [http://localhost:20000] Logging:: file[/var/log/miner/miniz/miniz.log] period[10] delay[0] Temp. limit: [90°C] [INFO ] Mining fee set to 2.00% miniZ<125,4>[06:0:00.0: 7885]: Selecting GPU#3[3] Zotac GeForce RTX 2060 SUPER miniZ<125,4>[06:0:00.0: 7873]: Selecting GPU#2[2] GALAX RTX 2070 Super EX Gamer Black miniZ<125,4>[19:0:00.0: 7843]: Selecting GPU#1[1] ASUS GeForce RTX 3060 Ti miniZ<125,4>[19:0:00.0: 7817]: Selecting GPU#0[0] Zotac GeForce RTX 3070 [INFO ] Applying clocks to GPU[00] core= +240MHz mem= +0MHz power=215W kernel=19 [INFO ] Applying clocks to GPU[01] core= +210MHz mem=+2450MHz power=115W kernel=19 [INFO ] Applying clocks to GPU[02] core= +190MHz mem= -800MHz power=145W kernel= 6 [FATAL ] GPU[0]: CUDA error 77 'an illegal memory access was encountered' in line 334 [INFO ] QUIT miniZ /hive/miners/miniz/h-run.sh: line 20: 17279 Segmentation fault (core dumped) ./miniZ $(< ${MINER_NAME}.conf) --logfile $MINER_LOG_BASENAME.log --telemetry 127.0.0.1:${MINER_API_PORT} --gpu-line

miniz exited (exitcode=139), waiting to cooldown a bit

Manual overclocking I'm using (best solrate with 2.0b) are: Cards Core PL SOLrate (+/-) 3070 +170 150 77 3060ti +145 142 64 2070s +100 150 63 2060s +180 135 54 all temps are between 57 and 60 degrees.