I've been fiddling with JSE again!
Uhoh!!
Sidetracked!!!

-=-=-

I managed to find a subtle way of rejigging the For loops, though it's not been quite the result I was hoping for.
Still, any increase in speed is a good increase in speed, right!?

In most places, you really can't tell the amount of work that went into this, but if you look specifically for the improvement, you can kinda tell it's there.

In this very specific test, I managed to get the end time down from about 7500 milliseconds to about 5200 milliseconds, which I think we can all agree, is quite an improvement.
Unfortunately, that's only really noticeable if you're "just" doing a bunch of nested for-loops, and are specifically looking for that.

Under real-world conditions, the difference is actually pretty negligible.
The Optimisationalism 4 test is pretty much identical to what it was a few days ago, showing just how very little the improvement has made in the grand scheme of things.
But the tweak has at least allowed me to rethink the way that the parser works.
A few more nips and tucks in that same style, and I think I can manage a bit more of an improvement, though I guess that depends on finding the little threads that I can pull apart to fix it up in that way.

I did consider a complete rewrite in how the parser works, but I imagine that probably wouldn't improve too much in the end, since I've already squeezed a ton of optimisation out of the current version. I reckon a complete rewrite would probably end up slower, until after I've spend 3+ years tweaking that one, too!

So... No, I think I'll leave it as is, and continue to nip and tuck what I've got.

... Right!?
Hmm..


"Cartoon @Derek man sitting on a giant mushroom, in a beautiful fantasy mystical forest, fairies, stream, bridge," by #ArtFlow.ai

Not very Greenie today. I do hope ArtFlow's not "lost" what Greenie is.. Hmm..