[jholman]

Yeah. I dunno about "most prevalent", though. TFA contained two "lame excuses", both of which are things I believed to be causes of slowness in dynamic languages, and now I have to reconsider.

    dynamic typing prevents type-based optimization
    monkey patching prevents optimization

I think the most common complaint I hear about GC is not that it affects computational throughput, but that it affects _predictability_ of computational throughput. One maybe doesn't care in scientific computing, but game developers are always going on about how they can't use a GC language because a stall mid-frame will knock them over 16ms/frame or 33ms/frame, which for console certification is a project-killer.

[cheald]

It's worth noting that WoW uses Lua; at some point Blizzard switched from Lua 5.0 (with used a stop-the-world GC) to 5.1 (which introduced an incremental GC) specifically because of this problem. Before, UI code could generate excess tables, kick in the GC, and dramatically impact your framerate. The incremental GC significantly helped this, since it permits the GC to be run over multiple frames, reducing or even eliminating the perceptible impact on the user experience.