Open chuckremes opened 13 years ago
For an example of the problem, visit http://speed.rubyspec.org/comparison/ and click on the first benchmark to bring up a graph.
The main problem stems from the introduction of multiple environment. JRuby runs under the JVM. JDK7 is almost ready for release and offers many performance improvements. I want to be able to compare the same runtimes on JDK6 and JDK7 to see the improvement.
However, only JRuby runs in that environment while all of the other choices do not depend on any JDK at all. Unfortunately, the system tries to display results for rbx and regular ruby under the JDK6 environment. There will never be any data points for those runtimes under the JDK6 (or the future JDK7) environment. Therefore, speedcenter should be smart enough to drop those executable/environment combinations from the chart and legend.
This may be tricky/slow or both to implement. Will comment further on it as soon as I get a shot at it.
In the Comparisons tab, sometimes reading the chart is difficult when there are a lot of entries in the legend. It's hard to tell the difference between all the different shades of green and mapping them back to the bars.
So, I have a suggestion for improving the display.
If an executable/environment pair has no data point available, remove it from the legend and chart. This will rescale the data that is displayed so that it is larger and easier to read. Right now those "no data" spots cause the chart to rescale and is hard to read.