Errors, bugs, questions - page 78

 
Malaxit:

Hello, answer my question:

Why can't i download mt5, i download the installation file, start it halfway, it stops, skips to another access point, etc. At the end there is an error-

-read failed [12002]

this is the story constantly no matter how many times i reinstall the terminal or windows but now the third day it fails

maybe not enough hard drive space? or no admin rights under windows
 
maryan.dirtyn:

please tell me why this construction always returns 0

and how do I know how many minutes it is now?

Something like this... :)

int Minute()
{
MqlDateTime tm;
TimeCurrent(tm);
return(tm.min);
}

or like this

int TimeMinute(datetime date)
{
MqlDateTime tm;
TimeToStruct(date,tm);
return(tm.min);
}

PS

And this is the most reading country... :(

Of course I'm sorry, but have you tried reading the article - Migration from MQL4 to MQL5?

 

Hello. When using fast optimization with the genetic algorithm, the tester somehow strangely calculates the required number of steps. When setting the parameters, the maximum number of them corresponding to the sequential search is calculated by multiplication:

However, when the test starts, the progress bar signals a higher number:

Since genetic selection considers a subset of the initial set of choices, it would seem that the steps should be at least no more.

1) maybe I misunderstand and one pass with a set of parameters is not one step (then what?)

2) maybe, indicator by mistake shows some maximal number of steps up to which very big sets of variants are cut down (by memory - approximately to such volume), but here it turns out vice versa as if....

3) the genetic algorithm is not simple and contains recalculations (populations are populated with copies, individuals that have passed into daughter populations are recalculated, ....)

Генетические алгоритмы - это просто!
Генетические алгоритмы - это просто!
  • 2010.05.25
  • Andrey Dik
  • www.mql5.com
В статье автор расскажет об эволюционных вычислениях с использованием генетического алгоритма собственной реализации. Будет показано на примерах функционирование алгоритма, даны практические рекомендации по его использованию.
 
I think I have it figured out - individuals in new populations and new individuals are not recalculated, and the results are searched among those obtained earlier, like
2010.08.04 13:43:44 Tester Genetic pass (96, 190) found in cache with result 43395.57
 

The genetic algorithm has its own expected initial training plan for the calculations. Of course, this plan is only a rough estimate, and the real values are obtained at the end.

In this case, when the number of complete overshoots is known to be less than the expected plan, we will adjust the expected number of passes.

Thank you for finding the inaccuracy!

 

There's also a question about the "ErrorDescription" library.

In my build 298 the editor doesn't know (line 43)

TRADE_RETCODE_REQUEST_CONNECTION

It was discussed here https://www.mql5.com/ru/forum/973/5862#comment_5862 that there was no such error in the help. Now in the help there is

10031

TRADE_RETCODE_CONNECTION

with the same number and it is recognised by the editor.

Question - what does the trade server currently return in this situation (and is 10031 even there) and will there then be a new version of the library?

 
k47:

There's also a question about the "ErrorDescription" library.

In my build 298 the editor doesn't know (line 43)

Better upgrade to build 302 and try again.
 

What should the user do in a situation where all work of the EA stops after the loss of connection to the server?

Does the terminal provide for broadcasting any signals in case the EA stops working?

Files:
example.txt  93 kb
 

What is the Projects folder for ?

What is the Project button in the MetaEditor Navigator for?
 
Yedelkin:

What should the user do in a situation where all work of the EA stops after the loss of connection to the server?

Does the terminal provide for broadcasting any signals in case the EA stops working?

So far there are no alarms, we will think about it.
Reason: