MQL5 looks very complicated language ...

 
Dear MQL5 developers. I really hope that you will continue to support MetaTrader 4 at least 2-3 years after the MetaTrader 5 official start. Unfortunately MQL5 looks very complicated language for non professional programmers. I afraid that you will loss a lot of popularity of your platform because of this.  I am a professional developer of expert advisors, but I am not a professional programmer. I think that 90% of the developers are like me, so I can tell you that you made wrong decision to change totally the MQL conception. You will satisfied 10% of your auditory with this new powerful language, but you will disappoint the rest 90%. I am really sorry to tell you this and I really hope that we should not learn a totally different language when you develop MetaTrader 6.
 
Autotrader posted  :
Dear MQL5 developers. I really hope that you will continue to support MetaTrader 4 at least 2-3 years after the MetaTrader 5 official start. Unfortunately MQL5 looks very complicated language for non professional programmers. I afraid that you will loss a lot of popularity of your platform because of this.  I am a professional developer of expert advisors, but I am not a professional programmer. I think that 90% of the developers are like me, so I can tell you that you made wrong decision to change totally the MQL conception. You will satisfied 10% of your auditory with this new powerful language, but you will disappoint the rest 90%. I am really sorry to tell you this and I really hope that we should not learn a totally different language when you develop MetaTrader 6.

I agree with you Autotrader. I created a number of EAs under MT4 and have really struggled with MT5. Functions that were simple in MT4 are really complex in MT5. It seems that all the useful high level stuff in MT4 has been broken down into small pieces. Even the simple process of getting a High[] Low[] of a tick is now beyond the average to inexperienced programmer. There seems to have been a thrust towards making it as close to C+ as possible (and there are probably advantages in this somewhere), whereas I would have thought that to have reached a large audience a leaning to BASIC would have been benificial. Maybe the code makes it easier for brokers to offer MT5 and this is the main thrust.

-Jerry

 
I like the new MQL5. Way more powerful than MQL4. And I gotta tell you, I don't know C++. I can barely read the code in C++. I don't know about objects too there, but at least here classes and structures are necessary to give more power to the EAs. Just compare a C++ script with an MQL5 script and you'll see that the C++ is way harder to understand than MQL5. The only thing I dislike here are custom indicators. These have become a pain to both write and debug, and hidden conceptual errors that don't appear on chart can appear as bugs when you attempt to use the indicator in EAs. I think that also this C++ish look of the language is helping developers, because it's approaching the MQL community towards the more fancy institutional communities working with Java and C++, thus allowing easier algorithm porting from and to MT5 - especially when brokers will come with a Level II MT5. The institutional community always considered retailers to be losers. A turn to BASIC would have proved them right - think how worse would be to turn retailers to BASIC, when institutionals are packing more and more speed in their C++ish HFT toys. If you want a simplistic syntax, try Vertex!
 

The trading matter dose not need powerful and complicated language like MQL5. I can develop profitable strategy with several rules and one indicator. By my opinion MQL5 is a conceptual mistake. MQL4 is more then powerful for the trading. But the major irritatingly thing is that every new MetaTrader is with a  new programming language – very frustrating!

 

However, any new MetaTrader appears at about 4-5 years, so you don't need to adapt too often. I could say that you are right, that trading wouldn't need such a complicated language, if it would be fit for only indicator-based, one-chart retail trading. But if MT5 will be ready to jump to Level II, then it must be ready to at least come with the technical features if it can't offer the low latency of the big stations. Probably you might like eSignal - it's simpler, but documentation is murky and unclear..
 
TheEconomist:

However, any new MetaTrader appears at about 4-5 years, so you don't need to adapt too often. I could say that you are right, that trading wouldn't need such a complicated language, if it would be fit for only indicator-based, one-chart retail trading. But if MT5 will be ready to jump to Level II, then it must be ready to at least come with the technical features if it can't offer the low latency of the big stations. Probably you might like eSignal - it's simpler, but documentation is murky and unclear..

I don’t tell that MQL5 is bad language, but it is not easy understandable like MQL4 and this will make very little number of specialists happy, but very big number of people like me crazy. I am amazed, how smart people like MQL team can’t understand that. This is the same thing like to publish Hamlet in a child book, because the publishers are too smart to publish a fairy-tales.

 
Autotrader:

I don’t tell that MQL5 is bad language, but it is not easy understandable like MQL4 and this will make very little number of specialists happy, but very big number of people like me crazy. I am amazed, how smart people like MQL team can’t understand that. This is the same thing like to publish Hamlet in a child book, because the publishers are too smart to publish a fairy-tales.

An important change from MQL4 to MQL5 is to use code libraries by preference, which is the Object Oriented way of thinking.  A well written set of libraries, and MT5 is already delivered with a very good start, will make MQL5 EA coding far simpler, easier to read and more robust than anything written in MQL4.

Paul 

 
phampton posted  :

An important change from MQL4 to MQL5 is to use code libraries by preference, which is the Object Oriented way of thinking.  A well written set of libraries, and MT5 is already delivered with a very good start, will make MQL5 EA coding far simpler, easier to read and more robust than anything written in MQL4.

Paul 

 

Yes ... but for a little number of professionals.
 

I really can't believe there are people claiming to develop "working" EAs  and at the same time find  MQL5 difficult .

C++ is the universal programming language. It is a blessing for us -programmers- that MQL5 is C++. Otherwise , It would lack the power to develop serious systems. Any EA with a remote chance to be profitable , needs to be a very complicated system. Proffesional -real life- Algotraders  are more than 20.000 lines of  C++ , and even so they are *barely* profitable.




 

MQL5 new features are more open and powerfull compared to mql4.

 

 

 

I am totally on the other side.

MQ5 is far better than MQ4, and on the contrary, i hope mq4 will be foret soon, and ALL the stuff

written in years for MQ4 will be converted in MQ5, that's a really way to learn fastly the mq5 language.


Even more on the contrary, i think that mq5 will not loose popularity over of mq4, and mq4 will losse

popularity over mq5, because mq5 is far more complex, better, and the future.


The natural evolution for metatraders programmers is mq5.


I am a basic programmer, but i'm happy to move to mq5.

So i can learn from the beginning, at least.. a kind of restart.


Bye

Reason: