Closed DaanVanYperen closed 9 years ago
Error in the benchmark, so i excluded it consider the release is defunct anyway.
On Thu, Sep 25, 2014 at 10:00 AM, Daan van Yperen notifications@github.com wrote:
What's going on with the Ashley insert/remove benchmarks? I'm assuming this is a benchmark bug?
— Reply to this email directly or view it on GitHub https://github.com/junkdog/entity-system-benchmarks/issues/9.
Hmmm maybe replace the benchmark images on the main artemis-odb README.md for one without Ashley 1.2.0, it's unintentionally disingenuous now. I'll add a quick notice to the readme.
Ah, now - I misread you (didn't get any sleep last night) - the ashley 1.0.1 pooling tests were off. The insert/remove performance degradation in 1.0.1 vs 1.2.0 comes from changing from an IntMap to Array. I tweeted siondream about it though, asking if the test was conforming intended ashley usage. Haven't heard back from him yet.
On Thu, Sep 25, 2014 at 10:12 AM, Daan van Yperen notifications@github.com wrote:
Hmmm maybe replace the benchmark images on the main artemis-odb README.md for one without Ashley 1.2.0, it's unintentionally disingenuous now. I'll add a quick notice to the readme.
— Reply to this email directly or view it on GitHub https://github.com/junkdog/entity-system-benchmarks/issues/9#issuecomment-56787269 .
The bar is there, it just starts out really low - 72.88 at 1024 entiteis, goes down to 1.05 at 65k entities.
Ugh! Would be nice to help them fix it. You don't seem to sleep anymore anyway!
Yup, definitely something that should be improved. If you're up for it, have a look at the benchmark code - just to make sure i haven't done anything stupid.
Profiling probably beats proofreading code like this. ;)
What's going on with the Ashley 1.2.0 insert/remove benchmarks? I'm assuming this is a benchmark bug?