kingkilr

42 ms·

391.
NumPy in PyPy - status and roadmap (morepypy.blogspot.com)
44 points by kingkilr 15y ago|9 comments
392.
by kingkilr 15y ago
Only 1/3 faster? Unacceptable! I wonder just how IO bound it is since it's not using HDFS, it'd be nice to know what speedup we're doing on the CPU bound portion of the task.
393.
by kingkilr 15y ago
I don't know about implementations using PEGs, but I do know that the parser is possibly the least interesting thing about a virtual machine.
394.
by kingkilr 15y ago
We've started referring to it as the "RPython translation toolchain" which will hopefully reduce confusion.
395.
by kingkilr 15y ago
There are plans, but nothing yet, this reddit thread has some decent information: http://www.reddit.com/r/Python/comments/h0uuv/pypy_15_releas...
396.
by kingkilr 15y ago
Yes, unfortunately Christain couldn't make it to the sprint, we would have liked to have him there. However, Kristjian (sp?) from CCP who maintains Stackless was there.
397.
by kingkilr 15y ago
Future cloudy, ask again later :) Because of the way our JIT is written it would be entirely possible for us to branch for py3k and put JIT improvements on both with relatively little effort, it's likely that at least our next release will
398.
by kingkilr 15y ago
There's actually a sprint going on right now (where this release was done), and that was a topic of discussion. I don't think anyone has written up the conclusions yet, but estimate I heard was that it'd be about 1-month of person-work.
399.
by kingkilr 15y ago
1.5 should be a lot better on these, the Loop Invariant Code Motion really helps on very tight loops, as these benchmarks often have.
400.
by kingkilr 15y ago
I wish I could find the video the talk where we explained this in great detail, but basically: it's hard, it's been tried and failed (notably by Armin Rigo, one of the leads of PyPy). Adding a new GC to CPython is insanely hard, and manual
401.
by kingkilr 15y ago
IronPython actually beat us to 2.7, however I will note that it was pretty fast. We went from 2.5 to 2.7 support in under 12 months.
402.
by kingkilr 15y ago
That's because they can't, in the US trademarks are scoped to the domain they exist in.
403.
by kingkilr 16y ago
Well, there's Howard Gardner's theory of multiple intelligences, however it doesn't really give you a single number the way IQ does (how could it!). I'm reminded somewhat of the scene on the spaceship to Battle School from Ender's Shadow.
404.
by kingkilr 16y ago
Surprising today, or 100 years ago when people thought IQ tests measured some immutable property of the human brain :) today of course we know it's possible to change your IQ in all sorts of ways.
405.
by kingkilr 16y ago
That's how it works on my laptop... (running Narwal).
406.
by kingkilr 16y ago
It'll never stop being weird to pull up something like this and see myself :/
407.
by kingkilr 16y ago
I don't see why it's an issue that Object.create is used, as long as attributes are assigned in a consistent order the optimization should still hold, since each assignment (or read) site would be monomorphic with respect to the current map
408.
by kingkilr 16y ago
But if you try sometimes, you just might find, you get what you need!
409.
by kingkilr 16y ago
My guess is it's a universal binary.
410.
PyPy Tutorial Part 2: Adding a JIT (morepypy.blogspot.com)
6 points by kingkilr 16y ago|0 comments
411.
by kingkilr 16y ago
Well, some of us (the PyPy developers) are web developers, but I'll grant you I'm the exception, not the rule :)
412.
by kingkilr 16y ago
Maybe there's a space for "if you spent an average amount of time optimizing" :) For example my PyPy optimizations took maybe 4 hours total, with 0 time spent looking at assembly.
413.
by kingkilr 16y ago
Well, that's what the compiler is for! PyPy uses SSE under the hood for various float handing things.
414.
My experience with the computer language shootout(alexgaynor.net)
95 points by kingkilr 16y ago|59 comments
415.
by kingkilr 16y ago
It says to use the highest optimization level.
416.
by kingkilr 16y ago
Yes, our current release implements Python 2.5 completely, and our trunk implements 2.7. As for speed, I'll let: http://speed.pypy.org/ speak for itself.
417.
by kingkilr 16y ago
a) Unladen Swallow wasn't research, in fact it was explicitly "just engineering". b) I seriously doubt this goes anywhere, Python is simply not amenable to productive static analysis.
418.
by kingkilr 16y ago
This came up once before, here's what I had to say: http://news.ycombinator.com/item?id=2144142
419.
by kingkilr 16y ago
It's absolutely not true that escape analysis is implicit in tracing: https://bitbucket.org/pypy/extradoc/raw/63e4617062b2/talk/pe...
420.
by kingkilr 16y ago
I honestly don't know why it's not such a big deal for Lua, but given LuaJIT's performance, and knowing that it doesn't do escape analysis I know it must not be a big deal :) In Python it's a big deal because everything is boxed, and your

More ›