What would it take to get everyone to finally switch to MT5? (collecting opinions) - page 15
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
The main question in the topic is. A clear, logical, thoughtful presentation is welcome.
I see. Make it easy to quickly test ideas.
Yes, I think that was the key to success in MT4: clear and simple product, language doesn't require from user to have deep knowledge of architecture and OOP, we just write a set of instructions - I want to do this and that - and language allows to do it guaranteed.
In MT5 the situation became more complicated, the event model became more complex, the data model became more complicated, you need to think a lot, write additional checks, for example you can't just get data for the indicator calculations, you need to make sure that the data is really loaded, and for multicurrency indicators it becomes a real challenge.
The solution could be to create a set of functions that would pack all of the trader's tasks, as it was in MT4. In part, the developers have already made such functions like iClose, iTime, etc. What we need is more functions like that. That is, we need A layer between the system layer and the application layer.
P.S. Thank you for collecting opinions.It is possible to run ex4 in MT5. Maintain this for a couple of years.
Close the local Market to adding new products under MT4 (leave only version updates).
These are the ticks provided by the broker. I looked at them visually: at a depth of a couple of years at the most popular brokers in Russia, in several ways, they are auto-generated (e.g. have a fixed spread of 50 pips).
Don't mislead people!
On the MT5 tester the real ticks have the same spreads as the real ones (fixed or floating), which are downloaded from the server where this account is located.
Yes, I think that was the key to success in MT4: clear and simple product, language doesn't require from user to have deep knowledge of architecture and OOP, we just write a set of instructions - I want to do this and that - and language allows to do it with assurance.
In MT5 the situation is more complicated, the event model becomes more complex, the data model becomes more complicated, you need to think about many things, write additional checks, for example you can't just get data for calculating an indicator, you need to make sure that the data is really loaded, and for multicurrency indicators it becomes a very difficult task.
Who-what prevents to write in MT5 in C style without any OOP?
The data model hasn't become more complex, just a bit different, and in general there are simple functions for everything without any OOP. Yes, and for most systems this model is needed in a much reduced form. In short, everything is about the same.
Who-what's stopping MT5 from writing in C style without any OOP?
The data model has not become more complex, just a bit different, and, in general, there are simple functions for everything without any OOP. Yes, and for most systems this model is needed in a much reduced form. In short, everything is more or less the same.
Stop explaining the obvious ))))
SZS: I remembered my childhood, it was the time... this topic reminds me of the children's debates in the early 90s about what's better: Coca Cola or Pepsi Cola, somehow those who thought Coca Cola was cooler won, evidently it was to show their taste for the beautiful even as kids, because Pepsi was in free access, while lucky beggars were not .... and then sprite came along and shattered all the stereotypes - it turned out this children's drink was created to wash down vodka with lemon ( Stopka ?? ) by adults ))))
Who-what's stopping MT5 from writing in C style without any OOP?
The data model has not become more complex, just a bit different, and, in general, there are simple functions for everything without any OOP. Yes, and for most systems this model is needed in a much reduced form. In short, it's about the same.
For example, it concerns multicurrency indicators - what can be easily and freely done in MT4 is not so easy in MT5. There was a topic in which developers considered this issue, but their opinion was mainly focused on speed and optimality, and not everyone needs it, as well as usability and simplicity. Developers were called to show the canonical correct code how to get time-synchronized data array in a guaranteed way, but they avoided it, as far as I remember. There are also times when you want to test some ideas and want to do it quickly. For many users, c.m.c. this is now out of reach as the barrier to understanding the internals in MT5 is higher than MT4. I'm not calling to abandon OOP and new cool things that are made in MT5 at all, but make an additional layer that would make mql5 similar to mql4 and some steps are already made (iClose functions, etc.)
On the MT5 tester, real ticks have the same spreads as real ticks (fixed or floating), which are downloaded from the server where this account is located.
If a broker launched an MT5 account 2 years ago, what real ticks from three years ago will they offer?
Why is this even necessary? Another fix idea? One platform has some advantages, another has other advantages. Some like one, others like another. Strength in diversity (which allows you to survive a changing situation).
I'll answer your question at the same time, as well as the ridiculous outcry that I want to destroy MT4.
The inevitable trend is that MT4 is slowly becoming obsolete. It cannot be changed. As Renat said: "Every architecture/skeleton has its own development limits". MT4 reached them a long time ago, and since it is no longer evolving (even if it were, it would quickly reach its ceiling) it could easily lose its position in the future.
Competitors are waiting and breathing down the back of MT4. They are surely doing their own development and time is playing into their hands. To remedy the worsening situation and keep up with development, the company is developing and implementing MT5. Technologically, MT5 is head and should not be beaten by competitors for years to come, but only if it is popular. Failures with MT5 implementation threaten the monopoly on the market as MT5 becomes less scary for competitors. MQ is left to rely solely on MT4, which is much easier for competitors to cope with technologically.
Attempting to introduce MT5, is a struggle for future niche dominance and striving to secure it for years to come. Failure to do so could easily fall prey to competitors and be forced out of the market in the near future. Therefore, once again, it is a question of survival. No one questions "why rearm the army? Don't World War I weapons shoot? Let them fight..."?
It would seem, what business is it of mine or other users? That's the company's problem. However, as I'm used to thinking big picture, I see the prospects for my project as tied to MT5. It is a powerful technological underpinning, convenient for realization of my most ambitious ideas. And I also need MT5 to be mainstreamed, so I don't end up being redundant along with the obsolete platform in a few years.
You can do it without OOP, but it's not only about OOP, and as I wrote above - you have to take into account more nuances, for example, concerning multicurrency indicators - what was easy and free in MT4 is not so easy in MT5. There was a topic in which developers considered this issue, but their opinion was mainly focused on speed and optimality, and not everyone needs it, as well as usability and simplicity. Developers were called to show the canonical correct code how to get time-synchronized data array in a guaranteed way, but they avoided it, as far as I remember. There are also times when you want to test some ideas and want to do it quickly. For many users, c.m.c. this is now out of reach as the barrier to understanding the internals in MT5 is higher than MT4. I'm not calling to abandon OOP and new cool things that are made in MT5, but to make an additional layer that would make mql5 similar to mql4 and some steps have already been made (iClose functions, etc.).
In general, imho, MK's mistake is the lack of backward compatibility of mql code, as it is common in all languages. I.e., MT4 prog must go to MT5, but vice versa, it is not necessary. Abandoning all the groundwork and starting from scratch to change the platform - who the hell needs it. MT4 will live forever! Till mammoths die out)).