When does it make sense to keep part of the robot code in an indicator? - page 2

New comment
 
Andrei01:

I do, but I've noticed that many non beginners use indicators, even though they seem to be more troublesome than useful.

I am wondering if there is a reason for that. ))



I can't understand smiling and cheerfulness... the question is serious, but the "smile" somehow stops you from expressing your opinion... Besides, you don't show what YOU have done on the subject... enter the discussion "from scratch" - I do not want to. You do not show developments in search on the issue, though the themes in this direction was a lot. Give references to the search is even uncomfortable after so many "stars under the avatar ... i still don't understand... the topic is for "talk" or for "get a solution".
 
DDFedor:

I can't understand smiling and cheerfulness... the question is serious, but the "smile" somehow stops you from expressing your opinion... Besides, you don't show what YOU have done on the subject... enter the discussion "from scratch" - I do not want to. You do not show developments in search on the issue, though the themes in this direction was a lot. Give references to the search is even uncomfortable after so many "stars under the avatar ... i still don't understand... the topic is for "talk" or for "get a solution".

I apologise, but unfortunately I didn't understand the connection between smiling and seriousness in the context of this discussion. I kind of reacted quite seriously to serious information on the subject so I did not give any reason to think the discussion was not serious. And I smiled just because the subject is serious and may be even painful for some people, since the use of indicators may be a matter of some kind of "faith/religion". If you are interested in concrete solutions for some problem I can share my thoughts, although I have not heard anything specific about it from you (but I do not exclude that I do not understand your idea correctly).

 
If you change the title of the topic to "when it makes sense to do indicator calculations in an EA", the answer is simple - none.
 
Integer:
If you change the title of the topic to "when it makes sense to do indicator calculations in an EA", the answer is simple - none.
Or maybe it would be more exact - "in which cases it makes sense to do the calculation of some part of the Expert Advisor logic in the indicator"?
 
Andrei01:
Or to be more precise - "In which cases does it make sense to calculate some part of the Expert Advisor's logic in the indicator"?


The meaning of "when it makes sense to calculate some part of the Expert Advisor's logic in the indicator" is the same as in the current topic title "when it makes sense to keep the part of the robot's code in the indicator?

 
Integer:


The phrase "In which cases does it make sense to calculate some of the Expert Advisor's logic in the indicator" has the same meaning as in the current topic title "In which cases does it make sense to keep the part of the robot's code in the indicator?

In this case, why change the subject to the following one - "When it makes sense to do indicator calculations in an Expert Advisor"?

It makes much more sense, and I doubt that someone will deliberately calculate indicators in an Expert Advisor and then apply the result back to the indicator.

 
Then don't pay any attention to me.
 
Andrei01:

Then why change the subject to something like "when it makes sense to calculate the indicators in the Expert Advisor"?

In fact, it makes absolutely no sense, and it's unlikely that someone will deliberately calculate indicators in an Expert Advisor and then return the result back to the indicator.



Use google search on this site, as the moderator recommended...if you are interested in this topic. Don't forget to write back here with the results. :-)))
 

It's strictly my opinion:

All logic should be contained within the EA. No calling MQL indicators in the EA. Calling something from DLL is acceptable, but only when really needed.

As for my experience, I have the following arguments:

  1. All-in-one" EA is faster than "with indicator" EA.
  2. "With indicator" has no advantage over "all in one".
  3. "All-in-one" is not tied to the nuances of interrelationships with indicators and peculiarities of their work.
  4. Dependence on MT+MQL bugfixes (4 and 5) in "all in one" is much less.
  5. Changes regarding the operation of indicators do not affect the results of the "all in one". For example, changes in MT4 since builds > 380.
  6. "All in one" does not imply a rigid interconnection of all internal units. "All in one" is also assembled from pieces, but bypassing the architectural rake of the platform.
  7. A few more points could be written.

P.S. Indicators are only needed for visualisation. That is, only for manual and semi-automatic trading. For full-fledged automated systems, indicators (aka visualization) are not needed by definition. I dare say this as the author of not trivial indicators.

 
hrenfx:

Purely my opinion:

All the logic must be contained in the Expert Advisor. No calls of MQL indicators in the Expert Advisor. Calling something from DLL is acceptable, but only when really needed
.

From my experience, I have the following arguments:

  1. All-in-one" EAs are faster than "with an indicator".
  2. "With indicator" has no advantage over "all in one".
  3. "All-in-one" is not tied to the nuances of the relationship with the indicators and peculiarities of their work.
  4. Dependence on MT+MQL bugfixes (4 and 5) in "all in one" is much less.
  5. Changes regarding the operation of indicators do not affect the results of the "all in one". For example, changes in MT4 since builds > 380.
  6. "All in one" does not imply a rigid interconnection of all internal units. "All in one" is also assembled from pieces, but bypassing the architectural rake of the platform.
  7. A few more points could be written.

P.S. Indicators are only needed for visualisation. That is, only for manual and semi-automatic trading. For full-fledged automated systems, indicators (aka visualization) are not needed by definition. I dare say this as the author of not trivial indicators.

You are deeply mistaken.

123456789...38
New comment
Reason: