Interesting topic for many: what's new in MetaTrader 4 and MQL4 - big changes on the way - page 52
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
I need a mql5 tester to be called from an EA.
What's the use of such self-restriction?
Now (with the "Two MT - one language" scheme) it can simplify many "test by trade" schemes. Moreover, such a tester can be easily ported to C++. I'm not sure about reverse porting at all.
In short, it's like this for now, and then we'll see.
It's a strange approach. In my opinion, it is logical to first find the best place to trade (the biggest potential profit).
You write your own SINGLE trading API for yourself. If you want to change platform and there, for example, MultiCharts is offered instead of MT5. Then you do not need to change the robot or the tester at all. Write only pad - a mutual converter (Gateway) between your trading API and the API of your broker. Don't forget that MT5 in particular is also a trading API.
Don't tie yourself to platforms and any particular third-party trading API. Write your own once. And from there, it's easy to go anywhere. Robots and tester will not change.
The same Stockshop is written on this principle. In FOREX, for example, you may trade on LMAX through it. I.e. it's just a question of writing a bridge between the two APIs.
Actually MT5 works on the same principle. You write in MQL5, and someone writes for MT5 gateways. The main task of developers is to get hooked on their trading API (MQL5) so that it would be hard to get off. Then the client becomes yours. That's why the hatred and jealousy of talking about other APIs and platform-independence is strong.
It's a strange approach. In my opinion, it is logical to first find the best place to trade (the biggest potential profit).
You write your own SINGLE trading API for yourself. If you want to change platform and there, for example, MultiCharts is offered instead of MT5. Then you do not need to change the robot or the tester at all. Write only pad - a mutual converter (Gateway) between your trading API and the API of your broker. Don't forget that MT5 in particular is also a trading API.
Don't tie yourself to platforms and any particular third-party trading API. Write your own once. And from there, it's easy to go anywhere. Robots and tester will not change.
The same Stockshop is written on this principle. In FOREX, for example, you may trade on LMAX through it. I.e. it's just a question of writing a bridge between the two APIs.
Actually MT5 works on the same principle. You write in MQL5, and someone writes for MT5 gateways. The main task of developers is to get hooked on their trading API (MQL5) so that it would be hard to get off. Then the client becomes yours. That's why the hatred and jealousy of talking about other APIs and platform-independence is strong.
MetaDriver:
It's a strange approach. In my opinion, it is logical to first find the best place to trade (the biggest potential profit).
You write your own SINGLE trading API for yourself. If you want to change platform and there, for example, MultiCharts is offered instead of MT5. Then you do not need to change the robot or the tester at all. Write only pad - a mutual converter (Gateway) between your trading API and the API of your broker. Don't forget that MT5 in particular is also a trading API.
Don't tie yourself to platforms and any particular third-party trading API. Write your own once. And from there, it's easy to go anywhere. Robots and tester will not change.
The same Stockshop is written on this principle. In FOREX, for example, you may trade on LMAX through it. I.e. it's just a question of writing a bridge between the two API's.
In fact MT5 operates on the same principle. You write in MQL5, and someone writes for MT5 gateways. The main task of developers is to get hooked on their trading API (MQL5) so that it would be hard to get off. Then the client becomes yours. Hence the strong hatred and jealousy in talking about other APIs and platform-independence.
I am already addicted to it. I`m thinking about treatment, but I`m afraid of withdrawal. I don`t need to be motivated, I understand everything myself. Again, you're basically right.
Logically I do not agree, MT is a constantly evolving infrastructure that over time will conquer all the sites standing, to change the site, even in MT infrastructure is not a problem.
But there is no need to write any Gateway at all, you can concentrate on the main thing, on trade ideas and on the actual chopping of $.
SZY
This is a strange approach. In my opinion, it is logical to first find the best place to trade (the biggest potential profit).
Again, how, may I ask, is this best platform sought? Only by developing algorithms and testing them within a site, so you have to write a Gateway for each site just to make sure that the site is a dud.
Again, how do we search for the best site? Only through development of algorithms and testing them within a marketplace, those for each marketplace we have to write a Gateway, just to make sure that the marketplace is no good.
I've written an entire fact sheet so that, in particular, such questions don't arise again. I myself only trade where it is most profitable. Even Renat likes this place.
There is no way to calculate the highest potential profit? Try to compare two similar crosses at different brokers. Where the potential profit is bigger - it's more profitable to trade there. Or at least compare the spreads.
Of course, the issues of quality of execution, withdrawal, etc. - Of course, the issues of execution quality, withdrawal rates, etc. are important. But it's all so simple now.
Logically I do not agree, MT is a constantly developing infrastructure which in time will conquer all worthwhile sites, changing the site even in MT infrastructure is not a problem.
But there is no need to write any Gateway at all, you can concentrate on the main thing, on trade ideas, their verification and the actual chopping of $.
SZY
Again, how, may I ask, is this best site found? Only through the development of algorithms and testing them within the site, those for each site need to write a Gateway just to make sure that the site foul.
And you're right. Once again.
I'm off to bed. Good night, everyone.
Morning,..... ....
(Futures)
Ideally we need a multi-threaded (+ working in the cloud) tick tester with visualization as in MetaTrader 4 + https://www.mql5.com/ru/forum/146025/page8#820498
+ possibility to import one's own tick history and any other history into the tester.
i.e. time, (Last) Ask, Bid, Vol on each tick.
I completely agree with Urain
You will get twice as much information, but it will be only half percent more useful.
Even if MQ will make your request (which I highly doubt), you will immediately begin to grumble that it's all bullshit, let's use a tick history.
And that's because the tick history really contains a lot of additional information against bars, and for the tick history we need a full storage of Ask and Bid for each tick,
And the bar history does not change much in informativity, if it is stored as AskBid or simply Bid.
https://www.mql5.com/ru/forum/146025/page47#827030
And those who need to filter, thin out ticks ( hrenfx ) - can do it from tick history.
If MetaTrader 5 is really a terminal designed for exchanges and for accurately testing your strategies in it.
In my opinion, all the "fights" over Ask history make no sense, because without a tick history it cannot be done normally.
And there is no tick history in the next 10 years, that's a fact.
I do not understand something else - why should the spread on a minute bar or on some other bar be counted by one value (maximum or at opening it does not matter). Why not take the average spread over a period, isn't that more logical?
With this approach at least it will be possible to explain why the results in the tester are the way they are.