Errors, bugs, questions - page 461
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
When generating an EA it is not possible to add or change parameters (build 478) What is the problem?
When generating an EA, you can't add or change parameters (build 478) What's wrong?
It works fine for me. If this is not the case for you, write to the BOD with details
Good afternoon, could you please tell me if it is possible to get mouse coordinates to a script without connecting a DLL?
It works fine for me. If this is not the case for you, write to the EA with details
The MQL5 Wizard: Creating an Expert Advisor without programming has even a screenshot of what can be added.
Developers.
What is the fastest way to run code in Expert Advisor or in modules?
The matter is that the Expert Advisor that I am preparing to the Championship on Intel P4 3.0 had runtime of 7-8 minutes with 3600 lines (including comments and other "rubbish").
The test duration has almost halved after a bit more than 2000 strings were moved out of the EA to a separate file.
Question - Is it my luck to optimize everything (although I haven't changed anything), or it is the size and location of the code that determines the speed of execution?
When generating an EA, you can't add or change parameters (build 478) What's wrong?
I don't know about earlier, I have no respect for Wizard. Only used it a couple of times, not really bothering to add new parameters (fair enough assume I'll add everything I need myself as the play progresses).
Now, yes, adding a new one in the dialog box doesn't work...
Developers.
What is faster to run code in EA or scattered by modules?
The matter is that the Expert Advisor that I am preparing to the Championship on Intel P4 3.0 had runtime of 7-8 minutes with 3600 lines (including comments and other "rubbish").
The test duration has almost halved after a bit more than 2000 strings were moved out of the EA to a separate file.
Question: Was I so lucky to optimize everything (although I haven't changed anything) or does the size and location of the code affect the speed of execution?
There is no difference in speed resulting from moving a part of the code to separate files, since the entire program is actually assembled into a single unit.
Most likely, explicit or implicit optimization helped.
Why can't I add a new one? You could before and the MQL5 Wizard: Creating an Expert Advisor without programming even has a screenshot of what can be added
This article refers to the first version of MQL5 Wizard, while now there is a more functional second version, which has some differences.
When you create an Expert Advisor in generation mode, you can only follow a rigid script with predefined parameters.
In the template based EA creation mode there are no hard rules and you can easily add your own parameters.
Good afternoon, could you please tell me if it's possible to get mouse coordinates to a script without a DLL?
This article refers to the first version of MQL5 Wizard, while the more functional second version is working now, which has some differences.
When you create an EA in generation mode, you can only follow a rigid scenario with predefined parameters.
In the template EA creation mode there are no strict rules and you can easily add your own parameters.
I don't know, 519188 ms (the longest test) and 256718 ms are very different in my opinion.
Before changes the test was passed in 434234 ms, and by and large only a few classes were put into a separate file.
Maybe it's an optimization, but I honestly have nothing to do with it :)
I don't know, 519188ms (longest test) and 256718ms are very different in my opinion.
Before the changes the test was passed in 434234ms, and by and large only put a few classes into a separate file.
Maybe also optimization, but I honestly have nothing to do with it :)
Send to servicedesk two versions of the code to check? Without the code the answer is simple "everyone puts the code together in a single block, so there is no difference".
Maybe paging data or something else (like caching in a previously lifted agent) was affecting the results? Tests should be run several times after warming up to avoid simple errors.
Don't forget that the first test run is almost always longer due to the fact that the agent goes up into memory and builds its caches. Subsequent passes already work with a well-prepared agent.