How to make Python run as fast (or faster) than Julia

[bartc]

On 23/02/2018 20:12, Chris Angelico wrote:
> On Sat, Feb 24, 2018 at 7:02 AM, bartc <bc at freeuk.com> wrote:

>> I don't know what point you're making here. Unless it's that no software is
>> worth writing unless it's done on a big scale with a huge team of people,
>> and is very well known.
>>
>> My own point should be clear:
>>
>>   Python is 10 times slower than a competitor = doesn't matter
>>   My language is 1.5 times slower than the big boys' = matters a great deal
> 
> Most performance is a tradeoff. Python has other advantages; what are
> yours? So far, all you've said is that you're "not too much worse".
> That's great for a toy language!

Ned Batchelder has banned me from mentioning any of my private work here.

So I'll keep it generic. Let's say the Tiny C compiler is not taken 
seriously because it might be a couple of times slower than gcc-O3, even 
thought it's 1% of the size and compiles 1000% as fast.

But the difference in runtime speed between Python and other dynamic 
languages, if you look at benchmarks doing actual work in the language, 
can be much greater than two times, yet that doesn't appear to matter.

Even defining how fast or slow Python actually is is fraught with 
complexities making it hard to establish just what its comparative speed is:

No one seems to be able to agree with what benchmarks ought to be used. 
Some don't seem to get benchmarks. Then it's question of whether you run 
CPython. Or PyPy. Or Ironpython or Jpython. Or Numba with @jit. Or 
Numpy. Or @cache or whatever it is. Or you run Cython.

Or maybe you can combine @jit, @cache, Cython, and PyPy in the same 
program; I don't know. (Have I ever mentioned Python is complicated?)

So, you were asking about the benefits of having a small, simple 
self-contained implementation of a language. I really couldn't tell you...


-- 
bartc