Errors, bugs, questions - page 678

 
Yedelkin:
No, as I understood it, they offered you a way "how to trade at opening prices in the tester, but taking into account ticks". You need such an option in the tester, don't you? You can reproduce the results of its work by a little completing your code and running it in M1 OHLC mode (e.g., introduce an option/switch in expert for testing)... You can even run it in "all ticks" mode.

Why not just keep the old mode and add a new one?

You won't have to make a mess of things and think what and how. I have enough problems with history synchronization, I don't need to think about what prices my orders will be triggered at...

 
Interesting:

Why not just keep the old mode while adding a new one?

Nah, don't ask me a new question :) Did you understand the way you were suggested (in my interpretation)?
 
papaklass:

For quality, would the "all ticks" mode not suit you?

All ticks is accurate and long, the one I propose to keep is a middle ground between very good quality and speed which will appear in the next build.

In particular, my opening price pass is almost the same as in all ticks mode, but the speed is higher. If you run an all ticks optimisation in the cloud, no balance is enough there.

Interesting:

And why not just leave the old mode, while adding a new one?

You don't have to make a mess and think about what and how. I have enough problems with history synchronization, I don't need to think about what prices my orders will be triggered at...

I support you!
Документация по MQL5: Стандартные константы, перечисления и структуры / Константы индикаторов / Ценовые константы
Документация по MQL5: Стандартные константы, перечисления и структуры / Константы индикаторов / Ценовые константы
  • www.mql5.com
Стандартные константы, перечисления и структуры / Константы индикаторов / Ценовые константы - Документация по MQL5
 

There is a two-dimensional array, it needs to shift the last column one whole column backwards

Prev--- original two-dimensional

Del--- service two-dimensional

ArrayCopy(Del,Prev,0,0,-1);
ArrayCopy(Prev,Del,0,0,Index-2);
ArrayCopy(Prev,Del,Index-1,Index-1);


Did I get it right?

 

Happy New Build! )))

Why didn't you update the Help? I want to know more details. )) Especially it is not clear, what are custom frames.

 
See section Working with Optimisation Results. Examples will be added.
 
Rosh:

See section Working with Optimisation Results. Examples will be added.

Thank you.

//---

In the Opera browser, the forum indicators on the right and left of the list are not fully displayed:

 

Build 619. Test, which used to go all the way to the end, steadily interrupts at 11%:

2012.03.24 00:55:15 Core 1 connection closed
2012.03.24 00:54:26 Core 1 log file "C:\Program Files\MetaTrader 5\Tester\Agent-127.0.0.1-3003\logs\20120324.log" written
2012.03.24 00:54:26 Core 1 stopped on 11% of testing interval
2012.03.24 00:54:26 Core 1 2011.01.24 23:00:00 OnTimer: LastTickReceived = 2011.01.24 23:00:00
2012.03.24 00:54:26 Core 1 456654848 bytes not available
2012.03.24 00:54:25 Core 1 2011.01.24 22:03:01 CTrade::PositionOpen: instant sell 0.10 EURJPY at 112.516 sl: 113.516 tp: 104.416 [done at 112.516]
2012.03.24 00:54:25 Core 1 2011.01.24 22:03:01 order to sell 0.10 EURJPY at 112.516 [#297 sell 0.10 EURJPY at 112.516]

 

The situation is rather comical.

Springtime aggravation, chasing a martin. I have been working on my own, but I have not had the chance to do it myself.

I have placed an initial deposit of 3000, naively believing that there will be a miracle and I will be rewarded with a grail. I optimized the profit and minimum drawdowns, ran the optimization. As a result at one point the position did not roll over due to insufficient funds, which was recorded in the log. Then the position was closed by TP. There are a lot of entries in the journal, so the entry with the error went down to the history and I did not see it. After forward testing, almost ran to the boss, face kicking.

Dear Metakvot, a big request to add a tab in the tester, which will be written only bugs, sometimes, as in my case, they are single and well masked in the general journal. Of course, I know you won't smell like that before you open a pose, you should check everything, but agree when you write a skeleton strategy missile and drive, it's not so good for small checks, besides error checking bloats code, which is already debugged, reworked and rewritten several times.

 

How can the script, knowing the indicator handle, get the name of the indicator buffer from the buffer number?

Reason: