System specs: Ryzen 5 3600, RTX 2060 Super. All tests are run on 1 CPU core.
Benchmark results are ordered by filters per second (filters/s). Finding and counting the resultant possible answers after being given a guess and colours counts as 1 filter. (e.g., what and how many words can be the answer if we guess "raise" and see grey, green, yellow, grey, grey?)
| Implementation | filters/s | Seconds to suggest best word |
|---|---|---|
| My friend's | 1.3k | Too many |
| Naive (Python/NumPy) | 13k | Loads |
| Naive (C++) | 170k | A decent amount |
| Precomputed word-pair colours (Python/NumPy) | 250k | 10-13 |
| Precomputed word-pair colours (C++/CUDA) | 1.4B | 0.002 |
This method achieves the 3146121 filters to find the best word in the greedy expected information algorithm presented in this video in about 2ms, giving it an effective filtering speed of 1.3B-1.6B filters/s.