As discussed prior (#5), Spark isn't quite complete as a statistical performance profiler. Indeed, it doesn't necessarily provide enough context on it's own to properly analyze the logging. Prior to Sponge API 8, Spark could be used in tandem with /timings for a complete trace of interpretational variables such as playercount, TPS, chunk, entity count, etc.

While Spark's display has vastly improved (thanks Luck!) since I made that initial first suggestion, it still lacks a lot that could palliate the lack of timings in Sponge moving forward. I've made a ticket discussing this in SpongePowered/Sponge#3649, but of course, this could also be solved by expanding the initial Spark use case.

The info available in spark is much better than it used to be!

What other metrics did you have in mind? Could you be specific?

Thanks for the prompt answer.

I'd mentioned this before, but seeing as Timings is being phased out by Sponge (API 8+), it would be interesting to see Spark address what currently would have required Timings to do. See https://spark.lucko.me/docs/misc/spark-vs-others#minecraft-timings, where you outline quite well why timings would have been preferred - accessibility, ease of display, as well as additional metrics.

Thanks - quite a few of the things there are out of scope for spark in my opinion.

spark is not intended to be a timings replacement, so striving for feature parity is not an aim of the project. However, I'm sitll open to specific suggestions that would fit in appropriately. :) feel free to open issues if you have any.