diff options
Diffstat (limited to 'TODO.txt')
| -rw-r--r-- | TODO.txt | 13 |
1 files changed, 7 insertions, 6 deletions
| @@ -1,18 +1,19 @@ | |||
| 1 | Bug in esep table generation | 1 | Bug in esep table generation |
| 2 | - add pre-computation of h48 coordinates at distance <=7? | ||
| 3 | - unit tests (but how do I test? just leave it there to check regressions) | ||
| 4 | - Add long-running test for h0k4 (maybe as a tool?) | ||
| 5 | - compute all tables for h<11 | 2 | - compute all tables for h<11 |
| 6 | - compute visited up to a fixed depth (7? 8?) | 3 | x compute visited up to a fixed depth 8 |
| 7 | - compute additional step (if needed) to fill <=base | 4 | - compute additional step (if needed) to fill <=base |
| 8 | - brute-force the last 2 steps (only 18 moves + 18*15 move pairs) | 5 | - brute-force the last 2 steps (only 18 moves + 18*15 move pairs, + sim) |
| 9 | - compare with known h0 results (from long-running test) | 6 | - compare with known h0 results (from long-running test) |
| 10 | - compute table for h=11 | 7 | - compute table for h=11 |
| 11 | - is is worth pre-computing stuff? | ||
| 12 | - can it be unified to the other computation, or is it much better | 8 | - can it be unified to the other computation, or is it much better |
| 13 | to do it ad hoc? | 9 | to do it ad hoc? |
| 10 | - Add long-running test for h0k4 (maybe as a tool?) | ||
| 11 | - tests for other sizes? | ||
| 14 | - optimize | 12 | - optimize |
| 15 | - use cached values for invcoord_esep? check if it is faster | 13 | - use cached values for invcoord_esep? check if it is faster |
| 14 | - Unify and improve performance of finding nasty sim | ||
| 15 | (right now we don't even use selfsim) | ||
| 16 | - parallelize with pthread | ||
| 16 | 17 | ||
| 17 | (OLD: | 18 | (OLD: |
| 18 | - Fails for UFRUFU, try command | 19 | - Fails for UFRUFU, try command |