diff options
| -rw-r--r-- | Tasks.org | 6 |
1 files changed, 5 insertions, 1 deletions
@@ -115,7 +115,11 @@ We have the code to do all this end-to-end, and to run a 'brute force' for compa * TODO It's fucking broken We can pick the right option, but our cost models seem really weird. -For example, insert grows linearly for linked lists, but goes down to 0 for vec. +~insert~ seems like the main culprit: probably because we're only measuring long runs of insert. + +** TODO Merge observations from multiple benchmarks + +ie if we have two different ~insert/100 ...~ lines we should use both instead of only one * BLOCKED Benchmarks & Evaluation |