Features of the mql5 language, subtleties and tricks - page 35

[Artyom Trishkin]

Alexey Kozitsyn:
20-30 seconds!? .... Omg... I think you did something wrong... My tick indicators are much faster. Several ticks on a chart.

...

[Alexander Puzanov]

Artyom Trishkin:

Run it on a symbol you've never used, and for which you simply have no data at all.

And, yes, it's good that you got a result - that's what I want to achieve. Gathering opinions on how best to do it.

I tried MOEX FORTS, all I have is 8 symbols for 6 TFs = 48 handles, I've never used it: even faster.

I mean, the number of handles is not an indicator, it doesn't affect the speed of execution. The copied into the CopyBuffer volume for sure.

[[Deleted]]

Artyom Trishkin:
...

As you wish, but the problem is with you, not with me...

[Artyom Trishkin]

Alexey Kozitsyn:
As you wish, but you have a problem, not me...

I asked for help, not a statement of fact. You would have just passed by then.

[fxsaber]

Artyom Trishkin:
I asked for help, not a statement of fact.

Youtz.

[[Deleted]]

Artyom Trishkin:
I asked for help, not a statement of fact. You would have just passed by then.

Are you offended with someone? Or you didn't notice the help? Or do you think your smileys should have been interpreted differently? If you want help, do as it was written above, and do not yorne. And then we'll see if further help is needed.

[Vitaly Muzichenko]

Artyom Trishkin:

I set 5000 bars of history in MT4 and MT5. I have a visual instant start in MT4, and a waiting period of 20 to 30 seconds in MT5. And that's assuming that the EURUSD symbol is used repeatedly for tests - there is a history in both terminals.

It is not necessary to tell the "impatient" user in every way to wait. It is necessary that there are no unreasonable delays when starting and switching the timeframe.

That's why I am interested in possible methods to solve this problem. For now. I'll experiment with them later - what really happens.

Similarly, the solution has not yet found, I read the branch, and hopefully someone will suggest something.

[Artyom Trishkin]

fxsaber:
Youz.

Thank you. What's in it for you now?

[fxsaber]

Artyom Trishkin:
Thank you. What will it give now?

Hope, Faith, Love! I would freeze the project for a few hours and wait for a decision from the developers.

[kaus_bonus]

With the release of build 1583 we will return to the issue of mismatching the number of ticks in the history with the number of ticks in the tester on low-liquidity stock tickers

https://www.mql5.com/ru/forum/170952/page29#comment_4845672

https://www.mql5.com/ru/forum/170952/page29#comment_4845795

let's run

KS      0       23:45:05.124    Core 1  GOLD-9.17: history ticks synchronized from 2017.04.03 to 2017.04.14
KF      0       23:45:05.124    Core 1  GOLD-9.17,M1: history cache allocated for 484457 bars and contains 347 bars from 2017.01.17 23:25 to 2017.03.31 19:46
MF      0       23:45:05.124    Core 1  GOLD-9.17,M1: history begins from 2017.01.17 23:25
ID      0       23:45:05.124    Core 1  GOLD-9.17,M1 (Open-Broker): generating based on real ticks
NM      0       23:45:05.124    Core 1  GOLD-9.17,M1: testing of Experts\Advisors\ExpertMACD.ex5 from 2017.04.02 00:00 to 2017.04.16 00:00 started with inputs:
ИТОГ
2017.04.16 23:45:05.124 Core 1  GOLD-9.17,M1: 242507 ticks, 227 bars generated. Environment synchronized in 0:00:00.040.
 Test passed in 0:00:00.181 (including ticks preprocessing 0:00:00.020).

result - good! there are changes.

Then the tricky part starts:

we run the multicurrency Expert Advisor on "cold" condition.

2017.04.16 23:54:12.000 Core 1  GOLD-9.17: history ticks synchronized from 2017.04.03 to 2017.04.14
2017.04.16 23:54:12.000 Core 1  GOLD-9.17,M1: history cache allocated for 484457 bars and contains 347 bars from 2017.01.17 23:25 to 2017.03.31 19:46
2017.04.16 23:54:12.000 Core 1  GOLD-9.17,M1: history begins from 2017.01.17 23:25
2017.04.16 23:54:12.000 Core 1  GOLD-9.17,M1 (Open-Broker): generating based on real ticks
2017.04.16 23:54:12.000 Core 1  GOLD-9.17,M1: testing of Experts\открыть-закрыть 2 пары.ex5 from 2017.04.02 00:00 to 2017.04.16 00:00 started with inputs:
ИТОГ
2017.04.16 23:57:02.105 Core 1  GOLD-6.17: generate 284232 ticks in 0:00:00.050, passed to tester 287584 ticks
2017.04.16 23:57:02.105 Core 1  GOLD-9.17: generate 15208 ticks in 0:00:00.020, passed to tester 242507 ticks

We should run it for the second time, without changing anything.

2017.04.16 23:58:51.687 Core 1  GOLD-6.17: passed to tester 287584 ticks
2017.04.16 23:58:51.687 Core 1  GOLD-9.17: passed to tester 242507 ticks

as far as i understand, now all ticks get to the tester. is it true, in this case not all from the first run?

s.s. as far as they now coincide with CopyTicks did not check.

https://www.mql5.com/ru/forum/188047