MT4 doesn't have long to live - page 76

 
OnGoing:
Personally, I don't like everything about trading operations. Syntax + execution. The first is overcomplicated, the second is clumsy.

You're right about the syntax of the trades. It's more complicated. You will have to spend an evening or two to figure it out.

Then it's easier - in the end, all the settings are hidden in the functions you write for yourself. By the way, I use standard (library) classes for trading.

I just added my own handy features to them and use them.

I don't agree about sluggishness. Everything flies very fast at runtime.

Most likely, this is just a subjective impression you have, because of the amount of information and features. It all seems like a big, big thing.

And your gut says, "If it's big, then it's retarded". It seems to be logical. But it is not. Everything works very fast. Very fast. Much faster than on quad.

The tester is not a criterion - there were conceptual miscalculations by the developers. Fix it already.

 
MetaDriver:
I've used ready-made classes too. But the performance was meant 'live', i.e. a demo, not a tester.
 
OnGoing:
I've used ready-made classes too. But the execution was meant "live", i.e. a demo, not a tester.
It's hard to imagine what you have done there, so that the speed of execution was noticeable. There is a mechanism implemented there that allows you to make trades faster. In general, ask questions on Five, they will tell you. Maybe even I will. )))
 
OnGoing:
I used ready-made classes too. But the execution meant "live", i.e. demo, not tester.

I don't know. Of course, the execution depends on

1. the broker.

2. the ping.

Shake the broker - he's wrong. Most likely. And check the ping - just in case...

 
tol64: In MT5 I still for 9 months I can't get to any limits in implementing ideas.

In general yes, but the particular problem is the inability to work in mql5 with multidimensional dynamic arrays - it's very hard to port off-the-shelf C code

OnGoing:Personally I don't like everything about trading operations. Syntax + execution. The former is overcomplicated, the latter is clumsy.

hmm, have you actually tried to understand the trading operations? or is it just... just what people around you are saying?

Is it hard to write this kind of code to open a position?

CTrade order;
order.Buy(lot);
maybe i'm being too rude, but you are just a chatterbox , next time write for example that in mql5 it's much harder to create custom indicators - that would be like the truth.
 
Hello, I read that mt5 is also certified for GBOT. What about the history of traded instruments? I think I have to wait for the platform to be bought by a broker who is a member of the exchange and uploads the history there itself?
 
Mathemat:

...

And if it's all about visualisation, it's worth something too. Instead of storing several hundred transaction results, it's much easier to do it like this:

There are 11.5 thousand objects. This is the curve of paper profit in the zero window. It's clear and illustrative. We take screenshots at certain moments and then analyse them.

Do you think that by using the same CQG it will all come out quickly and beautifully, without straining your brain?

You don't need 11,500 objects here. All that is needed here is data per number of bars * 2. If 11,500 objects are needed, it's a method problem, not a computational problem.
 

MetaDriver:
...
Так вот. В тех платформах, которые типа "для трейдеров", там тоже много чего не хватает. Причём, там это - невосполнимо.
Здесь же (MT4-MT5) я сам могу восполнить то, чего мне не хватает
...

The point is that customisation options have been cut to a minimum, and there is no guarantee that this cutting will not continue. For example, you'll wake up tomorrow and suddenly find out in a new MQ happy report that from now on, in order to maintain the integrity of the solution, you can't plug in third-party dlls to the project. This uncertainty and the clear trend towards trimming integration with third party products makes MQ's solution highly risky when it comes to integrating MT5 into a large trading project/infrastructure. And "out there" this is just make up for it, as there are usually no such tight constraints. There is not and never will be a single trading platform that will solve all professional trading tasks, at least at an acceptable level, because these are too different and serious tasks, the solution to each of which lies in completely different areas of the IT industry.

 

Lured them in, though. Parasites. :))

But seeing the positive attitude of almost all participants towards MQ. But combined with no understanding of their concept of the trading system, and therefore no understanding of the reasons for doing this or that, I want to say a couple of words. :)

1) MQ made the system a lower level language, giving third party developers the opportunity to modify ( customize ) the system to the various needs of users.

2) MQ made the system more secure so that the user and the "brokers" would feel more secure and make more use of third party developments.


So the philosophy is - we made a base, we made a system in which you can make commercial products for traders. We separated the part that must not (cannot) be changed by third party developers and implemented it in the form of our system. Thus we simplified the way for third-party developers to implement their systems, and thus significantly widened the potential of our system, compared to others. The challenge for us was to secure the environment, and we did.


:) All the words above are mine, just spoken for MQ. If I'm wrong they will correct them. But they don't say them IMHO only because they are already tired of repeating them. And as this discussion shows not everyone understands them.


******************

As for the "problem" of trading several Expert Advisors and some other EAs on a single symbol, of course, this is not an improvement. It is in fact a problem. We have to make a grouping of orders with the possibility to determine a section or a filter, say, using some kind of a group identifier.

In other words, this part also belongs to the system part and should be implemented not as a custom setting but as a part of the system.

That is, some kind of clearing on the user side.

******************

As for the net position and FIFO, it is obvious that even the most advanced people do not understand its essence and how it works. It seems to me that this part should be clarified.

*********************

And one more thing - it seems to me that MQ should stop treating its users ( not "brokers" ) as some blondes, because blondes always follow only eggheads.

***

I have not watched MT5 for a long time, but the trading system is not convenient. I have to enter too much stuff and fill it with a file. If you want to place pending orders on an chart, use a mouse to create them, copy pending orders with a mouse, delete orders on a chart and use group operations with orders. This is not what we have.

We have to redesign the interface and do it in a professional manner. The current interface is from the 90s.

**

As for me personally ( personally (!) once again ) .... I don't think I will use MT, even if I have to trade in another system, which uses calculation on paper, and in which 10% of indicators are not present in MT. And I don't want to use another system even if I have to use an MT system which doesn't have 10% of the indicators which MT has. Why? Probably because all this stuff is not the main thing in trading. All these, as they say here, balls are really not necessary. What is needed is simple, clear, ergonomic, quick, intuitive and visible. In short, MQ guys - understand that the most important thing in programmes is the INTERFACE, it's the case, it's the capping, it's the style, look at eql.

 
Mathemat:
I know what the main complaint was: netting, which ruined the previous accounting of warrants.

This? Any other complaints?


Once in some thread it was promised by the developers to write an example of two EAs on the same account and instrument. I guess this was never done.
Reason: