Per-Ake Minborg

`speedment-override.json`

We were talking about this the other day but no concrete plans yet. It is an important feature.

Hi and thanks for your question. In general, you should expect about the same execution time. In order to answer your question, I need to know what query you run...

They should execute in about the same time. If you run 100 iterations of each query, what is then the difference?

ok. That is a 50% increase in time which, in my opinion, is not satisfactory. I will reclassify this issue from `question` to `bug`. Speedment generates a bit more complicated...

ok. I guess you mean 50980 without Speedment and 53279 with Speedment? So, that means less than 5% increase which would be ok I think. Thanks for doing that extra...

Reproduced in `Issue720`

This is the current state running `Issue720`: ``` cnt = 246 speedmentElapsed = 42468519480 sqlElapsed = 15074681435 ratio = 2.817208420829226 ```

We have improved speed since this issue was recorded but there is more to do before this issue can be closed.

By default, the C2 JIT compiler kicks in after some 10,000 invocations of a method. This is the likely cause of the improvements for Speedment.