[Development] [BB++] Now is 3.5x faster than Node.JS

[Phil Bouchard]

Thanks for the information; I already sent this information to Google, Uber
and NASA. I'm about to tell WebKit as well. We'll see what they have to
say.

As far as closures are concerned, there is a way to convert closures with
parameters and return values so I personally don't think it's worth paying
the price of a GC just for these features.


My 2 cents,
-Phil

Sylvain Pointeau <sylvain.pointeau at gmail.com> wrote:
> I would suggest you to make your language feature complete before making
> any benchmark.
> For your information, lambda and variable capture cannot be discussed, it
> is absolutely needed.
> 
> in my opinion, you totally under estimate the design of a language, please
> read on Stroustrup / c++ etc.
> 
> as a first language design attempt, I would personnaly implement an
> existing language that would fit the purpose. (kotlin, pascal?, etc)
> as a scripting language, it will be very hard to propose anything else than
> javascript, you can try of course but without a backing company, there is
> very little chance to succeed.
> 
> Best regards,
> Sylvain
> 
> Le dim. 30 juil. 2017 à 19:23, Phil Bouchard <philippeb8 at gmail.com> a
> écrit :
> 
>> On 07/21/2017 12:22 AM, Phil Bouchard wrote:
>>> Greetings,
>>> 
>>> I am now using C++14 standards, the error reporting is all compile-time
>>> now and I get a test that is 3.5 faster than Node.JS! You can see it by
>>> yourselves here:
>>> https://github.com/philippeb8/root_ptr/tree/bb++/bbpp2cpp
>> 
>> I just tried it on Windows 8.1 and the standard deviation is even
>> greater: 4.2x faster on Windows when it is 3.5x faster on Linux (2.40
>> GHz). This must be because Visual C++ generates slightly better code
>> than G++.
>> 
>> I'll generate documentation for your convenience soon.
>> 
>> 
>> -Phil
>> 
>> _______________________________________________
>> Development mailing list
>> Development at qt-project.org
>> http://lists.qt-project.org/mailman/listinfo/development
>> 
> 
>