Has anyone wondered why and how much the trend building "walks"? - page 7

New comment
 

Experimented too:

 
The screenshots above are from five, there's something similar in four too, but the range is much lower. It's probably normal +- a pixel (and by normal I mean quadruple).
 
No on the fours, the range is definitely bigger than the pixel, the problem is in both terminals.
 
220Volt:
I also looked at the quadruple as well, there is something similar there, but the range is much lower. Maybe it's normal +- pixel (by normal I mean quadruple).

Holy shit, a pixel.

The start from one point on the structure at two identical coordinates, however, the result is different.


 
I've been doing some research and this is what came up: the same thing happens not only with trend lines, but also with Fibo levels (I'm 100% sure). It was on MT4. Apparently this is the case with all chart patterns.
Документация по MQL5: Стандартные константы, перечисления и структуры / Константы индикаторов / Стили рисования
Документация по MQL5: Стандартные константы, перечисления и структуры / Константы индикаторов / Стили рисования
  • www.mql5.com
Стандартные константы, перечисления и структуры / Константы индикаторов / Стили рисования - Документация по MQL5
 
220Volt:
I've been doing some research and this is what came up: the same thing happens not only with trend lines, but also with Fibo levels (I'm 100% sure). It was on MT4. Apparently this is the case with all chart patterns.
Fibo fan behaves like trend lines (I don't think I'm much mistaken). As for the Gann fan, it's quite a problem.
 

Question for experimenters.

Is it a principled position to refuse scaling? If so, why?

 

Re. From another forum.

romanqz 

Есть такая шутка, что в MT даже паралельные прямые пересекаются,
связано это с тем, что очень большой размер пикселя используется
при построении графики, отсюда большая погрешность

Who can comment on this? Could this be the source of the problem?

Something I don't quite understand, how pixels can be different...

PS yes, the example given was thinkorswim, which has a higher resolution...

 
Silent:

Question for experimenters.

Is it a principled position to refuse scaling? If so, why?

It doesn't matter how you select the scale (manual or auto), the effect will show when it starts to change.

P.S. You might think it's a trifle, what a pathetic item, but it can be very important. Please do get it up to speed.

 
220Volt:

It doesn't matter how you select the scale (manual or auto), the effect will show when it starts to change.

Set it hard, in points per bar. Does it keep changing? I don't.

I'm not saying there is no problem. I'm suggesting a solution. It suits me just fine.

123456789
New comment