![]() ![]() (Even if this information can't be always collected, as for example Corsair Link supports a proprietary API to read the real PSU load, any approximation would be enough, plus some algorithms do report this on per-card basis. Moreover, I'm really begging you to add an option to specify a maximum power load, either per-card or in total, such that we could assert that we'd rather prefer the algorithms to shut down (or not run at all, potentially starting up slowly one-by-one to watch a safety margin) before this occurs. ![]() I'm aware that this can be probably overridden using some CLI parameters of the algorithms, yet I'm sure that from UX perspective it makes no sense to expect all your users to immediately have innate knowledge of which algorithms are the "exception" here, so I'd suggest to at least signal those "dangerous algorithms" which may change the power limit when running on default settings. This setting gets reset by some specific algorithms (seemingly randomly) when they're run on default settings, and it seems to especially occur when benchmarking, such that we never know beforehand when the reset is going to occur. When we lower the power limit (TDP) using a tool like MSI Afterburner in order to fit within dimensions of our power source, NiceHash constantly puts our hardware at risk of being destroyed, or in better case at least of having a long downtime due to PSU shutting down without auto-restart, due to a major negligence on your side, considering that lowering TDP is a pretty typical use-case which you really should have considered, as many cards just don't perform better and simply waste electricity & increase temperatures when running at 100%. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |