A little surprised :) Thought I'd share and ask a NOT rhetorical question. - page 5
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
Question to the developers:
Will the Expert Advisor with indicators and the same Expert Advisor, but with transferred indicators in its code ("all in one"), be different in terms of speed of execution in the tester? In which direction?
Most likely, there will be no definite answer to this question. But all the same I ask you to somehow more or less clarify this question.
I remember a popular comment on 4th forum that in absence of IndicatorCounted call in EAs, indicators that depend on their previous values will work with a big braking effect. Because the expo will have to calculate the entire buffer on each tick.
So if we are talking about primitives like Buf[i]=(High[i]+Low[i])/2, we can of course implement them in the Expert Advisor and it will certainly work faster.
Why are you making a sad face with incomprehension? Does knowing what will be faster solve anything for you?
Have fun
The developers, if they wish, will confirm, not to lie, that they can always transfer indicators to the Expert Advisor's code with absolutely identical results in the Tester. But at the same time we will not lose any speed and will get a serious opportunity for further algorithmic optimization. Having conducted such optimization, "all in one" will be in the TESTER always faster than its indicator "twin brother" by definition.
The variant "all in one" is needed only for those who appreciate the speed of execution in Tester and Optimizer. Moreover, the variant "all in one" can be easily transferred to your self-written "rough" optimizer to perform the optimization much faster.
No IndicatorCounted() is needed for the tester and optimizer.
Here we go... With this approach, you get a wild brake on calculations.
You need to learn the theory and look at the standard indicators. They are almost all economical with IndicatorCounted() and recalculate only last bars.
Just the correction of one of the errors of buffers reinitialization in the last build of MT4 has revealed the problems of indicators that do not consider IndicatorCounted().
Here we go... With this approach, you get a wild brake on calculations.
You need to learn the theory and look at the standard indicators. They are almost all economical with IndicatorCounted() and recalculate only last bars.
Just the correction of one of the errors of buffer reinitialization in the last build of MT4 has revealed the problems of indicators that do not consider IndicatorCounted().
What is the habit of taking things out of context? I was speaking about the absence of IndicatorCounted() in Expert Advisors. Allegedly, because of this absence, all-in-one EAs have to recalculate the same thing a hundred times. But the IndicatorCounted() is not needed in explicit form in Expert Advisors, it can be simply implemented through static variables.
Initially, we are talking about competently written indicators and competently written Expert Advisors "all in one" for TESTER. We are not discussing the analysis of bad handedness. We are comparing the speed of two approaches in writing EAs for Tester. Exactly where the issue of speed is important.
Here we go... With this approach, you get a wild brake on calculations.
You need to learn the theory and look at the standard indicators. They are almost all economical with IndicatorCounted() and recalculate only last bars.
Just fixing one of the errors of buffers re-initialization in the last build of MT4 revealed the problems of indicators, that do not consider IndicatorCounted().
In the tester, the bars are received sequentially and there are no communication interruptions. The time of the last bar is memorised, due to which only one bar is recalculated. But this will only be a toy for the tester.
....
It is originally about competently written indicators and competently written all-in-one EAs. There is no debriefing of the crooked hands. What is being compared is exactly the speed of two expertly written EAs.
Do not be silent about the context. You are talking about "all-in-one" EAs solely for work in the tester.
Don't be silent about the context. You are talking about an all-in-one advisor solely for the tester.
Especially for you, I'm writing again, as in all the posts above in capital letters, speed is important for Tester and optimizer. Therefore, speed is only considered for the Tester and optimizer.
SPEED and only speed.
Suppose you have two identical results in TESTER, one on indicators, the other one "all in one". The second one works much faster. It is obvious that in order to optimize it, you will run the second variant (because it is faster) and find the necessary parameters and insert them into the Expert Advisor with the indicator, which you will run on the real account, because there mechanisms are more reliable for the real trade.
Once again, we are discussing the SPEED in the Tester.
Especially for you, I write again, as in all the posts above it was written in capital letters, speed is important for the Tester and optimiser. That's why speed is only an issue for the Tester and the optimizer.
I do not have to personally write anything, I can read, my attention is steady, my memory is good. Moreover, this message, to which you personally responded to me in capital letters, demonstrated my awareness of the konokst.