optcarrot

I don't think it runs Optcarrot (the current most prominent Ruby benchmark) otherwise it'd be listed here https://github.com/mame/optcarrot. It doesn't have a lot of compatibility with standard Ruby yet according to https://eregon.me/blog/2020/06/27/ruby-spec-compatibility-re... which could be why it doesn't run it. But it's a really exciting project!

If you interested in seeing how the 3x3 initiative* has come along, here are the benchmarks so far: https://github.com/mame/optcarrot#readme Personally, I am very excited for this release. * Matz's goal to get Ruby 3 to be 3x faster than Ruby 2. -- @sosodev Thanks for the updated info!

A lot of focus has been on this benchmark https://github.com/mame/optcarrot/ It has seen huge performance improvements thanks to the mjit but not quite 3x yet.

https://github.com/mame/optcarrot Includes some rare, bonus ruby implementations too <3

In terms of pure Ruby performance, there's a semi-official non-trivial benchmark called optcarrot. Ruby 2.4 is 14.3x faster than Ruby 1.8.7 here: https://github.com/mame/optcarrot/blob/master/doc/benchmark.... Rails performance relative to JVM+Scala+Play is tricky to measure but the closest thing to a decent benchmark I know of is this: https://www.techempower.com/benchmarks/#section=data-r16&hw=... My own experience at ChartMogul with building production services in Ruby, Go and Rust is that the overall throughput of REST API type services increases a disappointing amount. With JRuby you get to retain all the benefits of the Rails ecosystem. Most of the Go and Scala community is driven by much larger companies who do open source libraries but they don't really build frameworks to help developers build software quickly.

Sinatra + Sequel is already very competitive in web performance with Go + Gin[1]. It's the Rails convenience stuff which slows things down massively. MJIT could probably bring Rails in line with Sinatra though. Between Ruby 1.8 and 2.5, performance has improved around 13x in tight loops[2]. The Rails performance issue has been massively overblown since 1.9 was released. Ruby 1.8 was a tree walking interpreter, so the move to a bytecode VM in 1.9 was a huge leap in performance. Twitter bailed to the JVM before moving to 1.9. A lot of those 10-100x performance differences to the JVM are gone thanks to the bytecode VM and generational GC. Bytecode VMs all have the same fundamental problem of instruction dispatch overhead, they're basically executing different C functions depending on input. Doing _anything_ to reduce this improves performanc

"optcarrot benchmark https://github.com/mame/optcarrot runs around 9x faster on JRuby+Truffle than it does on MRI" That's blazingly fast! The missing part now is C extension support and they seem confident to be able to deliver this. Color me impressed.

On the other hand, I am confident it's written in a much nicer way than most NES emulators (look at cpu.rb/ppu.rb). The goal of optcarrot is explicitly not to be the "practical"or "the fastest NES emulator". From https://github.com/mame/optcarrot > NOTE: We do not aim to create a practical NES emulator. There have been already many great emulators available. We recommend you use another emulator if you just want to play a game. JRuby+Truffle's performance is not bad compared to other languages: https://github.com/smarr/are-we-fast-yet#current-status