Alternative implementations of standard functions/approaches - page 5

 
fxsaber)
 
i-Series functions that have appeared recently are written in MQL? And when compiling, do they embed their MQL sources (like mqh) to compile more efficiently?
 
fxsaber:
The i-Series functions that have appeared recently are written in MQL? And when compiling, they embed their MQL-sources (like mqh) for more efficient compilation?
How is this possible without #include ?
What if there is a naming conflict?
 
Nikolai Semko:
How is this possible without #include ?
What if there is a conflict of names?

These are not fundamental features of the implementation at all.

 
fxsaber:
The i-Series functions that have appeared recently are written in MQL? And when compiling, do they embed their MQL sources (like mqh) to compile more efficiently?

No, of course not.

 
Rashid Umarov:

No, of course not.

Then please clarify these sentences

Forum on trading, automated trading systems and testing of trading strategies

Alternative implementations of standard functions/approaches

Renat Fatkhullin, 2018.06.09 17:10

We returned to the idea of embedding the mass of simple system functions natively in the resulting MQL5 code, which allowed us to speed them up a lot by using them in the global optimization of the resulting code.

It concerns functions like NormalizeDouble, string operations etc.

Will be available in next betas after next week's release.
 
it seems to be about native code execution at the instruction level of the processor
 
fxsaber:

Then please clarify these sentences

Not sure myself - let's wait for implementation, there will probably be an explanation.

 

I think adding these functions to Five is not a priority at all. Except for iBarShift, which is really indispensable, in terms of efficiency. For all the rest, the existing self-written analogues are not inferior in performance. So what is the point of wasting time and resources on this. After all, there are much more necessary things for programmers that are missing in the language itself.

 
Alexey Navoykov:

the available self-written equivalents are not inferior in performance.

Even the analogues vary greatly in terms of performance. The number of people who can write them or find them is less than a percentage of the users.

Reason: