28 !!! currency pairs, 1 expert. Another grail, but this one I think no one has ever shown. + DEMO ACCOUNT - page 13

 

Hello, everybody!

Folks, help me out here, because I don't know what I'm doing!

Podskite where you can download the platform for the contest!

And what is it (Expert file ( *.MQ4 or *.EX4):?

I will be grateful to you for help!

 
oreshenko:

Hello, everybody!

Folks, help me out here, because I don't know what I'm doing!

Podskite where you can download the platform for the contest!

And what is it (Expert file ( *.MQ4 or *.EX4):?

I will be grateful for your help!


You don't have to download anything for the contest, you just have to register and send your Expert Advisor file (you can also send *.mq4 and *.ex4). But isn't it too early to enter the contest with such knowledge?
 
Why, you can set up a MACD Sample and see if you get lucky.
 
klerk:
Why, you can set up a MACD Sample and see if you get lucky.
Indeed, they say that newcomers are lucky :)
 
Rosh:
granit77:
As for the "pictures" as you put it, these are the actual test results on the latest build of the EA layout at the time, the source code of which I had previously offered to send to Rosh if the developers had a need for it. There was no request, hence no need.
Honestly, I don't remember (I'm browsing a lot of forums and have to quickly slip for the most important stuff) that you suggested sending the source code, but if it helps find a potential bug in the tester, I'm all for it. I guarantee non-distribution and destruction of the EA after testing.

EA has been sent and the test is now finished. The results may be useful in understanding the limitations of testing certain types of EAs.
The tested Graal uses a separate indicator for a senior TF, manually set on the chart. The Expert Advisor searches for chart objects set on the chart by the indicator and makes decisions based on its analysis.
The Expert Advisor has no ability to "look into the future" and the indicator just can!

With the author's permission, I quote Rosh:
"The point is that the indicator thrown on the visual testing chart knows nothing about the testing that is going on and can refer to historical data of the higher timeframe without any problems. This is why such a tricky peeping happens. But if you would call the indicator from the Expert Advisor, it would be impossible......
Briefly: indicators drawn manually on a visual testing chart are not subject to the same restrictions on peeking into the future, as opposed to indicators called from an Expert Advisor under test".

The second conclusion that can be drawn from this experience is that indicators using data from higher TFs are not visually
more "beautiful", than they really are, due to "peeping" of the future. The data, transmitted by such
indicator will not correspond with the indicator lines, that are visible in the tester.
 
granit77:
Rosh:
granit77:
As for the "pictures" as you put it, these are the actual test results on the latest build of the EA layout at the time, the source code of which I had previously offered to send to Rosh if the developers had a need for it. No request was made, so there is no need.
Honestly, I don't remember (I browse a lot of forums and have to quickly slide around looking for what's most important) that you offered to send the sources, but if they help find a potential bug in the tester - then I'm all for it. I guarantee non-distribution and destruction of the EA after testing.

EA has been sent and the test is now finished. The results may be useful in understanding the limitations of testing certain types of EAs.
The tested Graal uses a separate indicator for the older TF, manually placed on the chart. The Expert Advisor looks for chart objects set on the chart by the indicator and makes decisions based on its analysis. This is the whole thing.
The Expert Advisor has no ability to "look into the future" and the indicator just can!

With the author's permission, I quote Rosh:
"The point is that the indicator thrown on the visual testing chart knows nothing about the testing that is going on and can access the historical data of the higher timeframe without any problems. This is why such a tricky peeping happens. But if you would call the indicator from the Expert Advisor, it would be impossible......
Briefly: indicators drawn manually on a visual testing chart are not subject to the same restrictions on peeking into the future, as opposed to indicators called from an Expert Advisor under test".

The second conclusion that can be drawn from this experience is that indicators using data from higher TFs are not visually
more "beautiful", than they really are, due to "peeping" of the future. The data, transmitted by such
indicator will not correspond with the indicator lines, that are visible in the tester.

If you do not mind, please answer the following question: (I cannot make it clear)

I have a symbol chart open on H1. If I write in an Expert Advisor
Tenkansen_current  =iIchimoku(Symbol(),Н4,9,26,52,MODE_TENKANSEN,0);
And the Ishimoku chart is not open, do I get a correct result?

Regards - S.D.
 
Sart:
If you could please answer the following question: (I cannot get any clarity)
I have a symbol chart open on H1. If I write in the Expert Advisor
Tenkansen_current  =iIchimoku(Symbol(),Н4,9,26,52,MODE_TENKANSEN,0);
and there is no Ichimoku chart open, do I get a correct result?
Regards - S.D.
I think it is correct, at least when testing, there is no peeking into the future here.
You do not need to have an open chart to obtain the indicator data.
For the test, you can temporarily perform a charting of objects with this value after obtaining the value of Tenkansen_current, it will be very clear.