[Archive!] Any rookie question, so as not to clutter up the forum. Professionals, don't pass it by. Couldn't go anywhere without you - 2. - page 195

 
alsu, thank you!!!!!
 
DhP: Your train of thought is beyond the comprehension of ordinary people...
What's not to understand? If you don't ask questions about what you don't understand, it's your own fault for not understanding
 

Thank you all!!!

And here's another one, there's some kind of mistake, I'm not such an expert on the єquel 4... If you can help me. It's not my indicator, I'm writing it down.

 
DhP:

There is probably no clear answer to this question.

It does not matter how much an EA can earn, what is important is that it does not lose the deposit.

Aim to create an Expert Advisor that you could entrust with any amount of money.


At least some figure, maybe from your own experience or the experience of friends and acquaintances...

There is a starting point in ANY comparison, in relation to which it is easier to make further comparisons. In fact, I already have a figure (I'm only talking about data from the tester from 1.1.2010 up to now). I've managed to get close to 5000 points from 1500-2000 points during this period, analyzing the code's operation and adding functions to improve the operation. Is this good, bad or satisfactory? There is no possibility of running the code 7/24 for now. And leaving the computer unsupervised is not a good idea, so there is no real-time data from at least the Demo...

 
kulon:


At least some figure, maybe from your own experience or the experience of friends and acquaintances...

In ANY comparison, there is a starting point against which it is easier to make further comparisons. In fact, I already have a figure (I'm only talking about data from the tester from 1.1.2010 to the present). I've managed to get close to 5000 points from 1500-2000 points during this period, by analyzing the code and adding features to make it work better. Is this good, bad or satisfactory? There is no possibility of running the code 7/24 for now. And leaving the computer unsupervised is creepy, which is why there is no real time data from at least the Demo...

Better get the data from the tester before 1.1.2010 and skip the data from 1.1.2010 to the present without changing the settings of the Expert Advisor.

This will show you what your EA is capable of.

 
eddy:

artmedia70, I need to wait for the first tick in init and remember its time in milliseconds, understand?)

I don't need to "work something out at the first start" but to work something out after the first tick after that start


You're a persistent writer :)

Stand on the word init, press F1 and reread it several times. And read not only the paragraph about init, but about start as well. And try to analyze them (a set of words in paragraphs).

 
SergNF: Stand on the word init, press F1 and reread it several times. And not only the paragraph about init, but also about start.
i have read it. what do you want me to pay attention to?)
 
eddy:
I read it. what do you want me to pay attention to?)


No mention of ticks

init() - функция, вызывающаяся в процессе инициализации модуля. 
В случае ее отсутствия при инициализации не вызывается никакой функции.

The only mention of ticks

start() - основная функция. 
У экспертов вызывается после прихода очередного тика. 
У пользовательских индикаторов вызывается при пересчете после прикрепления индикатора к графику, при открытии 
клиентского терминала (если индикатор прикреплен к графику), 
а также после прихода очередного тика. 

This means that init() is only executed once, regardless of whether there are ticks or not. Moreover, when executing init (if the function is heavy), the ticks may be skipped. But the point of the start function is precisely the execution of the function when a new tick arrives. Again, if the function is heavy and slow, the ticks may be skipped, but this function is called at the arrival of a new tick of the symbol in the chart, to which the indicator is attached

 
DhP:

Better take the data from the tester before 1.1.2010 and skip from 1.1.2010 to the present without changing the EA settings.

This will show you more clearly what your EA can do.

Before 1.1.2010 it's not so good...
But an EA cannot work well at all times, on any TF, with any pair...

This is what optimisation is for!

 
kulon:

Until 1.1.2010 it's not so good.
But an EA cannot work well at all times, on any TF, with any pair...

That's what optimisation is for!

In that case, how are you going to determine that times have ended and a "new" time has begun?

And that optimisation carried out from 1.1.2010 to the present will start laying golden eggs for you on Monday?