I'm wondering if Snap! could have a 'Run [] and [] at the same time' block. I would find this handy for Snap! and for me.
[poll type=regular results=always public=true chartType=bar]
- Vote
[/poll]
We have it! It's in the "Parallelization" library. (It's near the end of the library list, so you have to scroll down to see it.)
Oh, a library of blocks? Thats cool!
True.
This is no longer the case, since the library list was reorganized for 6.1. Parallelization is now in the first group of libraries (the ones I think should be primitives).
I understand the last group (hardware), but can you explain the other groups?
First group: Things that should really be primitives, like the old Tools library.
Second group: Jens's media computation libraries.
Third group: Extensions to the core language. Mostly Brian's, except for the https one.
The division isn't perfectly clean. For example, I think words and sentences should be primitive, but I know I'll never convince Jens. And bar charts is a modified version of Jens's Frequency Distribution Analysis library, vastly simplified for use in the BJC curriculum, not really something I'd call a core language feature.
Mainly the point is to get the first group up to the top, so you don't have to do archaeology on the big jumble we used to have to find the things you use every day. At some later time I may make subgroups within the groups, ordering by the main palette category they extend. For example, a lot of the first group are Control extensions; the colors library is mostly a Pen extension; bignums are an Operators extension. Etc.
Edit: I guess the first group is really "things Brian thinks should be primitives," because the things that Jens thinks should be primitives are primitives!
Edit2: Actually, could you all do me a favor? Take a few of your favorite projects, make a version that includes the bignums library, and time them with and without bignums enabled. One reason Jens doesn't want bignums just on all the time is that he says it'll slow down Snap!, so I'd like to know how much of a slowdown we're talking about. Thanks.
how did you do that?
He has the power!
Snap! Build Your Own Blocks There may be a subtle difference for one trial, but it adds up for multiple trials; my result was that BIGNUMS had the higher time. Another one: Snap! Build Your Own Blocks My results were very close (just off by two centiseconds, BIGNUMS on being the higher one)
That's not a surprise; the question is whether the difference matters. Thanks!
Edit: ... which it does, nearly twice as fast without bignums. :~(