Custom Miner Integration

[ATrump]

PhoenixMiner updated to 3.5c

зальют феникс 3.5с ?

run command:
custom-get http://download.hiveos.farm/custom/phoenixminer-3.5_c.tar.gz -f

[ut9ea]

ничего не изменилось

[ATrump]

Ну чо-то поменяли. У меня стал писать температуру карты, которой раньше не писал.
Напомни, какие у тебя проблемы остались?

[ut9ea]

не запускается на 73 образе с драйверами 18.20
Unable to open log file /var/log/miner/phoenixminer/lastrun_noappend.log - 2
No OpenCL platforms found
No avaiable GPUs for mining. Please check your drivers and/or hardware.

custom phoenixminer exited, waiting to cooldown a bit

[ATrump]

так на 57м же запустили. Или гонешься за цифрамим в номерах версий?
У меня ваще 73й при загрузке уходит в кернелпаник.

[ATrump]

For fair testing should be done in parallel on two completely identical rigs to one pool for at least 24 hours

of course, we understand this

I’ve changed API port for Phoenix, so now it can be one rig with Phoenix and Claymore runned simultaneously.

[cooper]

с логами проблему исправили, а вот с остальным, похоже, нет. В 57 образе с апдейтом до 74го, феникс 3.5с работает.

[salvo2002]

Post there your driver version

yup, did that here https://bitcointalk.org/index.php?topic=2647654.msg45559767#msg45559767

[novashop]

I cound’t add custom miner for ETH

How can you do?

[salvo2002]

testing now! same config for each setup, 2x570s and 2x580s — I let them autotune hence why they are showing slightly different hashrates.

[ATrump]

Good!
Don’t forget to set different worker names.

[ATrump]

I`d updated the driver to amdgpu-pro 18.10. The number of stale shares decreased! Reported hashrate become a little higher.

[salvo2002]

Don’t forget to set different worker names.

oops! forgot to do that – fixed now.

updated the driver to amdgpu-pro 18.10

Thanks for reporting on the new drivers! I’ll update after this 24hr comparison run.

[salvo2002]

@ATrump

recommend instead of
echo "-mport $WEB_PORT" >> $CUSTOM_CONFIG_FILENAME
in h-config.sh

and
WEB_PORT=3335
in h-manifest.conf

just change global_config.txt to:

#DONT EDIT THIS FILE, IT WILL BE OVERWTITTEN AFTER UPDATE

#Management port, readonly
-mport 3335

#Log to the same file
-logfile /var/log/miner/phoenixminer/lastrun_noappend.log

the rest of the options in there are not implemented yet (HW control) or are not the correct syntax for Phoenix… -r in Phoenix is -rmode

#run reboot.sh on GPU hangs
-r 1

#Target temp
-tt 68

#stop GPU temperature
-tstop 82

-allpools 1

I am running without a -tt option and it works fine (using HiveOS autofan)

[ATrump]

Thanks, I’ll fix it in next release.

[HaloGenius]

T-Rex NVidia GPU CUDA 9.1/9.2 miner package

Custom miner: T-Rex

T-Rex v0.6.6 NVidia GPU CUDA miner which supported algorithms: balloon, bcd, bitcore, c11, hmq1725, hsr, lyra2z, phi, polytimos, renesis, skunk, sonoa, tribus, x16r, x16s, x17 Miner home: https://bitcointalk.org/index.php?topic=3299496.0 Custom miner integration details Package auto-detect installed NVidia driver version for running CUDA 9.1 or CUDA 9.2 optimized version of miner. Starting release 0.6.6 moved to JSON format of config file. Also added failserver support. Telnet API binding t…

[Maks72ru]

скажите при обновлении майнера достаточно ли изменить путь или надо удалять старый надо