mt5/mt4? - page 11
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
We have already created and implemented the fifth platform. Each time we launched it was accompanied by hooting and mass criticism in the forums.
,,judging by the pressure and achieved versions of MT, you are not going to stop, so it follows that MT6 is not far off (just a matter of time)...
If you want to be sure of the correct answer to your question, let's start with the release of MQL5, MQL6, MQL7 ... So the programming language will also change with a new version (MQL5, MQL7, etc.).
,,judging by the pressure and achieved versions of MT, you are not going to stop, so it follows that MT6 is not far off (just a matter of time)...
MQL5, MQL6, MQL7 and etc.).
We will have 3 more years until MT6 before we start thinking about it.
We are not going to change the language anymore, since we have found the right architecture through trial and error, having gone through all the pitfalls. We will only expand the functionality.
Customers are not seeing it that way. OpenCL seems to have impressed only a couple dozen people at the most (myself included). But there is no OpenCL in MT4 and it will not be supported by MT4. So I definitely need a "five". Or I will have to use crutches like dll in "four".
A really profitable system cannot be primitive and requires a lot of computation. I do not believe in long term profitability based on simple and well known indicators. Just do not give me an example of ATS winners. In most cases it was a fluke, apart from a few likely exceptions (if you count not only the nominal winners, but all those in the top 20).
Then I have a puzzle for you. You must realize that the selection of an optimization parameter in MT4 is available only for students. That's why MT5 has an opportunity to write your own optimization criterion. How can I set my own criterion in MT4? I can't even imagine that yet. And therefore I don't consider MKL4 as a serious product for development of a serious system.
The fact that you can't imagine something speaks only of your imagination and cleverness.
I once made my own optimization criterion for MT4.
A really profitable system cannot be primitive and requires a lot of calculation.
Everyone has different opinions. For me, the simpler the more profitable. The main thing is to find regularities and separate them with filters. Filters are simple, all based on mash-ups. I remember someone in the military department telling me how in Vietnam American soldiers abandoned their jammed M16 rifles and picked up Soviet Kalashnikovs. Then they told how the Ak-47 worked reliably even after being driven over by a Humvee, or submerged in water or buried in the sand. And the M16 broke down in all those cases. That is how a trading system should be as reliable as an Ak-47, so that deviations in quotes when moving from broker to broker or seconds delays in price arrivals do not break the system. Lots of calculations only make the system more sensitive to deviations.
The highlighted in blue - yes, I agree. But the last sentence of the quote is not necessarily correct at all: the result of extensive calculations can be some statistics, and at the same time a test of its stability.
The main thing is to find patterns and separate them with filters from the huskies. And the filters are simple, all based on mash-ups.
I'm almost certain that these filters on the mashups are applied depending on the context, whose evaluation is based on non-automated reasoning ("gut feeling").
P.S. I've looked at a couple of your works in kodobase. From your comments on the Hodrick-Prescott filter, it appears that it doesn't provide any significant advantage over conventional mashups either. But conventional mashups don't provide any benefit when applied out of context.
The highlighted in blue - yes, I agree. But the last sentence of the quote is not necessarily correct at all: the result of extensive calculations may be some kind of statistic, and at the same time a test of its robustness.
It's almost certain that these filters on the mashups are applied depending on the context, whose evaluation is based on non-automated reasoning ("gut feeling").
P.S. I've looked at a couple of your works in kodobase. From your comments on the Hodrick-Prescott filter, it appears that it doesn't provide any significant advantage over conventional mashups either. But conventional m ashupsdon't do any good if applied out of context.
The regular dash rules! No flair. You just have to know how to use it. You don't need to be an astrophysicist to predict sunrise.
The clients don't see it. OpenCL appears to have impressed only a couple of dozen people at most (myself included). But there is no OpenCL in MT4 and it won't be supported by quadruple. So I definitely need a "five". Or I will have to use crutches like dll in "four".
...
OpenCL, multithreading, 64-bit and other technical "stuff" are a thing in themselves (even MQL5). On their own, they won't be commercially successful. Ordinary users will have the same question "Cool! - But why?".
MT5 has to offer something special, something that MT4 doesn't have. And in my opinion there is such a possibility. The independent developers can create a lot of unique and complete products, like graphic panels, indicators and automation tools, which are fundamentally impossible to create in MT4. And then there will be the first swallows like: "I use MT5 because there is a program, which is not available in MT4".
It has been explained many times.
The complexity of the system and the demands for simplicity are such that all the dirty work has to be removed and everything has to be done automatically and transparently. Traders in MetaTrader 5 don't even need to think about any history load - just open whatever you want of the available ones and everything will work, including testers, cludes, etc.
And once again the explanation fails miserably when you connect to one broker with a history of well over a year, to another with instruments on the fingers count, when calendar history is needed, when synthetics are needed, when...
And why do I need such cleanliness?
I still want to believe that this is a temporary solution dictated by marketing.