MT4 doesn't have long to live - page 51
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
Don't get all flustered - we are talking about debugging MQL5 code.
And abstract strategies can be tested in Matlab...
You MUST debug your code on any ticks. :) Even on silly ones, because silly ones can occur too - it's a random process.
So don't talk about debugging, okay. :)
I don't think it's only MQs that are preventing you from trading successfully.
I personally do not see the logic that MQs are doing something to their detriment on purpose. And it is even clear why they do it. On the contrary, if everyone at MQ was making cool money they would sell more servers.
Debugging code MUST be done on all ticks. :) Even on stupid ones, because stupid ones may occur - it's a random process.
So don't talk about debugging, okay. :)
What do you mean?
;)
Well look - MT is the terminal and the server. OK? OK, you have run on your ticks and got super - but the server does not produce such ticks, so why build something into the terminal that will never happen?
It is very unhealthy to think others are idiots
I don't think anyone is an idiot. And you in the first place. But agree - the root of not successful trading is not MT for sure. :)
You MUST debug your code on any ticks. :) Even on stupid ones, because stupid ones may occur - it's a random process.
So don't speak about debugging, okay. :)
Debugging should be done on ticks with known properties, including stupid ones. But it is useless to do debugging on ticks specially dissected by someone else. And don't pretend that it's some kind of unseen thing. In all system development packages you can input quite a specific set of data.
As I understood here in general they are asking (begging) for a perfectly simple thing: give an opportunity to slip your quote to the tester input. No, you can't, there will be gigabytes.
Can we have some decorum?
You're a guest, you're being discussed.
Or have the Metakwots done you any harm? This is not the place to vent your grievances and complexes.
IMHO
Debugging should be done on ticks with known properties, including stupid ones. But it is useless to do debugging on ticks specially dissected by someone else. And don't pretend that it's some kind of unseen thing. In all system development packages you can feed quite a specific set of data into the input.
As far as I understand they ask (beg) for a very simple thing: give an opportunity to input their quotes into the tester. No, you can't, there will be gigabytes.
That's right.
That would be a wonderful feature of the product. Let it be used by only 0.5% (it's hard to say how many) of users.
But the mere fact that it is possible is already a huge plus.
Of course we do not know how this violates the integrity of the system, but Rinat did not talk about it.
And I haven't heard any argument that the system developers don't need it.
No problem with that argument. I personally do not need a tick history. Those TS which are more or less workable with me trade at opening prices on timeframes no smaller than M15.
And the argument that they allegedly need ticks for arbitrage and spidering is also unfounded.
Arbitrage trades are essentially a rebound strategy and positions are not protected by stops. The profits are penny wise but the risk is high. That's why any decent backfires against the wool and the deposit will be ruined. The system should be able to calculate the probability of large moves in order not to get in their way.
Spreading, i.e. arbitrage on several highly correlated symbols, is also possible only by calculating the cyclic behavior, and not the tick cycle but the annual one. Futures with different delivery terms have cycles and at certain times of the year tend (but not necessarily) to diverge in price or converge. In addition, for spreaders, there are special types of contracts, which the broker must close if the total profit on open positions for different instruments has reached a certain point. There are no such types of contracts in MT5 at the moment.
No problem about the arguments. I personally don't need a tick history. Those TS that I have more or less workable, trade at opening prices on timeframes at least M15.
And the argument that we need ticks for arbitrage and speculation is also unfounded.
I have not got a good enough profit to make a profit, but I have not got a good enough loss. The profit is tiny and the risk is big. That's why any decent reverse trade against the wool and the deposit will be ruined. The system should be able to calculate the probability of large movements so that it does not try to get in their way.
Spreading, i.e. arbitrage on several highly correlated instruments is also possible only by calculating cyclicity and not by ticks, but yearly. Futures with different delivery terms have cycles and at certain times of the year tend (but not necessarily) to diverge in price or converge. In addition, for spreaders, there are special types of contracts, which the broker must close if the total profit on open positions for different instruments has reached a certain point. There are no such types of contracts in MT5 at the moment.
why discuss what you don't trade and don't know in principle?