On 17/09/2022 08:54, Jacques Clavel via use-livecode wrote:
point 6. maybe potential for future performance gains :
- But today LCB is surprisingly less efficient (/3) than LCS (???)
Yes, at first glance it is surprising; typed variables etc. *should*
make it possible to be more efficient. And I think some day they will.
But for now, I think that isn't, and shouldn't be, a priority for LC Ltd.
The intended uses for LCB are to interface to other libraries/languages,
and to provide graphic / UI widgets - and performance is not critical to
either of those roles. If I have a LC app which is suffering
performance, I should first look at the algorithm, the data structures,
engine features I might be able to use, ...., separate process, ...
And if none of them fix it for me, then I'm probably going to want a
high-performance library in some other language (C, C++, even NumPY,
whatever) which allows it to be highly optimized.
My new slogan :
If I'm suffering for performance, I don't want a 10% improvement from
LCB, I want a 10x improvement from C.
Alex.
_______________________________________________
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode
