Why is D slower than LuaJIT?

[Andrei Alexandrescu]

On 12/22/10 8:16 PM, Andrei Alexandrescu wrote:
> On 12/22/10 4:04 PM, Andreas Mayer wrote:
>> To see what performance advantage D would give me over using a
>> scripting language, I made a small benchmark. It consists of this code:
>>
>>> auto L = iota(0.0, 10000000.0);
>>> auto L2 = map!"a / 2"(L);
>>> auto L3 = map!"a + 2"(L2);
>>> auto V = reduce!"a + b"(L3);
>>
>> It runs in 281 ms on my computer.
>
> Thanks for posting the numbers. That's a long time, particularly
> considering that the two map instances don't do anything. So the bulk of
> the computation is:
>
> auto L = iota(0.0, 10000000.0);
> auto V = reduce!"a + b"(L3);

Oops, I was wrong. The two instances of map do something, I thought 
they're all applied to L when in fact they are chained. So my estimates 
are incorrect. At any rate, clearly iota incurs a 2x cost, which 
probably composes with other similar costs incurred by map.

Andrei