The future of automated trading - page 23
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 don't know about easier. I've been using the standard way since about 95...
The standard way requires a recursive pass through files/folders, and here all "one-touch".
P.S. Quietly the moderator ran and pressed the "ban" button)))
Time to go to bed)))
The standard way is to recursively go through the files/folders, but here it's all "one-touch".
P.S. Quietly the moderator ran and pressed the "ban" button))
Time to go to bed)))
Apparently, we all have forgotten that we can write our own functionality for the terminal in the terminal's native language, including our strategy tester, which allows us to test it on any data, including "purchased" ticks. And in general, everything that the terminal lacks, everything can be implemented by means of MQL5. In some time such solutions will surely appear in abundance, to fill the niche of custom testers that is still empty.
But I believe the future of automated trading consists not in terminal-type program solutions, but in such special programming languages. If only there were a language and an executable environment for it, then everything would be possible.
That's why I admit the possibility of such specialized programming languages as for engineers, medics, etc.
I repeat - everything. No one writes their own software products consisting of 100% of their own components these days. All programs use some form of operating system library.
This is especially noticeable in Linux programs, which is why Linux programs weigh very little.
We are talking about a specialised programming language for trading operations. Everything that is necessary for trading (and even much more than necessary) can be implemented in MQL. The specific functions that go beyond trading and data analysis, service functions, etc., can be solved with the help of operating system libraries.
After all, no one would call the Matlab language cheesy, would they? This is a special programming language for mathematical functions. So it is here too.
I'm talking about specific languages, not general-purpose. And the foundation of the programming language for traders has already been laid. There's no need to mix up the flies with the cutlets.
ZZY I also wrote my own tester in MQL4. It implemented everything I needed without using dlls and API. That said, I've never said anywhere, nor have I ever thought MQL language was cheesy.
I wrote my tester in Matkad. Does it mean that the functionality and the environment are good? If a trader-programmer chooses a different language for research (the time to debug and test an idea is of great importance). You can write everything in assembly language, and in machine code absolutely everything. The question is when you are going to trade, in this life or in the next century.
The foundation has been there for a long time and the future we are talking about here has already come http://www.kroufr.ru/forum/index.php/topic,11085.0.html.
It's just coming (maybe it is coming) for MT, but with a business scheme like the one I see now, there is no chance. Just take your mind off it and imagine two robots fighting, both have the same energy capability. Only one has a reaction time of milliseconds, and the other robot has at least 3 seconds, even if it is located on the championship server + re-quotes are shoved to him.... betting on who will win? ( I bet on the first robot)
Here's a quote. It's 3 years old.
In 2007, the American company IBM announced that by 2015 the number of traders on the London Stock Exchange will shrink by 90 percent, since most of the trading will be done by robots. IBM predicts that the company that develops a robot capable of reacting as quickly as possible to changes in the market will win the trading algorithm race.
In 2007, IBM announced that by 2015, the number of traders on the London Stock Exchange would decrease by 90 percent, as most trading would be done by robots. According to IBM's prediction, the company that develops a robot capable of reacting as quickly as possible to changes in the market will be the winner in the race of trading algorithms.
The foundations have long been in place, and the future we are talking about here has already arrivedhttp://www.kroufr.ru/forum/index.php/topic,11085.0.html
It's just coming (maybe it is coming) for MT, but with a business scheme like the one I see now, there's no chance. Just take your mind off it and imagine two robots fighting, both have the same energy capability. Only one has a reaction time of milliseconds, and the other robot has at least 3 seconds, even if it is located on the championship server + re-quotes are shoved to him.... betting on who will win? ( I bet on the first robot)
Are you changing the concepts and think everyone here is a fool? Where did you see that the terminal reacts in 3 seconds? Run any Expert Advisor and measure the execution time of a trade order on MetaTrader 4/5 servers. Don't confuse artificial execution delays and technological delays. Read on the forums what the howling is for traders who trade on our platforms (especially FORTS) using domestic terminals.
I wrote my tester in matcd. Does it mean that the functionality and the environment are good? If traders-programmers choose another language for research (debugging time, idea checking is the most important thing). You can write everything in assembly language, and in machine code absolutely everything. The question is when you are going to trade, in this life or in the next century.
The foundation has been there for a long time and the future we are talking about here has already comehttp://www.kroufr.ru/forum/index.php/topic,11085.0.html.
It is for MT it is just coming (maybe it is coming), but with a business scheme like the one I see now, there is no chance. Just take your mind off it and imagine two robots fighting, both with the same energy. Only one has a reaction time of milliseconds, and the other robot has at least 3 seconds, even if it is located on the championship server + re-quotes are shoved to him.... betting on who will win? ( I bet on the first robot)
Here's a quote. It's 3 years old.
In 2007, the American company IBM announced that by 2015 the number of traders on the London Stock Exchange will shrink by 90 percent, since most of the trading will be done by robots. IBM predicts that the company that develops a robot capable of reacting as quickly as possible to changes in the market will win the race of trading algorithms.
If you are trying to argue, would you be so kind as to think of what you are arguing instead of clinging to phrases taken out of context?
Is Matcad a specialised language for trading? You may write it in XML, if you like. I find it easier to write trading robots and testers in a language specialised for this purpose. I see the future of auto-trading in it.
That's nonsense. What serious investment fund would leave their work to any, even a super duper cool robot. I suppose there are always people behind the robots, and tearing the hairs on the top of their head (if there are any left:))
A properly designed and written robot will be more reliable than a human. :)
P.S. 666 is my reputation))
You are changing concepts and think everyone here is a fool? Where did you see that the terminal reacts in 3 seconds? Run any Expert Advisor and measure the execution time of a trade order on MetaTrader 4/5 servers. Don't confuse artificial execution delays and technological delays. Read on the forums what a howl is on traders trading on our platforms (especially FORTS) using domestic terminals.
1.
No, I'm not stupid. And I am not substituting concepts. There is a delay in the execution of the order, everyone has it. Its magnitude is important, if you add to technological delays and artificial (as in the championship, read the rules paragraph 4.8 processing time of trading orders from 2 to 7 seconds) .
I also advise you to open any rules, any company using MT and find there stitch on the execution time ....-normal market 1...4 seconds... can take up to 3 minutes...
Read the claims on KROUFR - 99% of the claims are quality and execution time of trade orders....
2.
What you propose to do, measure the time on the servers, for me is not feasible, because I do not have a server, I have a terminal, and I can only measure the ping to your server. And I can assure you that it (ping) is much less than 1 second, and if the speed of execution of my orders depended only on it (would be critical for me), I know how to reduce it.
The formula: 3 min - my ping= .... who makes a fool of who?
3.
If you do not know what pings are and you do not know how to reduce them, you should take them as close as possible to the pings. 3. I have never seen such information for MT anywhere. Only saw the figures I gave in the first paragraph + there is real experience (milliseconds there and not a whiff).