search

andru-kun / wildrig-multi

multi algo miner for AMD, NVIDIA and Intel gpu's
251 stars 51 forks source link

挖掘 curvehash 算法 内核一直重启 ？？ #150

Open paopao-tang opened 2 years ago

paopao-tang commented 2 years ago

大概十几分钟重启一次 正常吗 是内核问题 还是 网络 矿池 等问题？？

andru-kun commented 2 years ago

I don't understand Chinese, so I hope Google Translate made a good translation for me. So, stability of AMD gpu mining curvehash depends on core voltage, you need to find stable values. For NVIDIA - don't use any overclocking(OC), it will constantly crash the gpu.

paopao-tang commented 2 years ago

The video memory of the NVIDIA GPU core I used for frequency down running is 0. I tested several mineral pools. After running for more than ten minutes, the core will be restarted

andru-kun commented 2 years ago

You need to try first stock clocks(for core and memory). And use power limit if needed.

paopao-tang commented 2 years ago

Video memory 0, core 0, adjust power consumption to control? Hive OS I use

paopao-tang commented 2 years ago

您需要试用第一个存量时钟（用于和内存）。并需要在内存时使用功率限制。 QQ图片20220818184713 QQ图片20220818184657

andru-kun commented 2 years ago

Oh, man, this is not a gpu crash. This is too many rejects, because pool uses too low difficulty at start and your ping is too high. Try parameter --max-rejects 32, maybe that will help

paopao-tang commented 2 years ago

QQ图片20220818190042 QQ图片20220818190132 About 15 minutes, several machines almost restart at the same time, and the running time of the kernel is the same