Benchmarks are static

[ryanholmdahl]

Currently, our benchmarks have to be calculated elsewhere and then passed to the script via a json file. Ideally, we can capture this dynamically from the miner processes, and can use the rate of shares instead (since this can vary based on the quality of the pool server, and that's how you get paid).

[ParkerBeck]

This will be broken into 2 steps

• [ ] Create a python script that runs the benchmark and stores the hashrates into a text file.

• [ ] Modify the python script to immediately read and store the hashrates for coins we care about for easier parsing.

[ryanholmdahl]

We discussed a bit offline; my impression was (and tell me if I'm wrong)

• [ ] Create a script that parses miner outputs as shares*difficulty/hour, extensible for new miners
• [ ] Create a static benchmarking script that runs every pool-miner-coin combo and stores it in a text file
• [ ] Add dynamic benchmarking to read shares*difficulty/hour from current mining and use to update saved benchmarks

It's probably worth looking into if it's even worth looking at shares vs hashrate, or if pools are pretty consistent in giving the same number of difficulty-shares for the same hashrate.