[ARCHIVE!] Any rookie question, so as not to clutter up the forum. Professionals, don't pass by. Can't go anywhere without you - 4. - page 167

New comment
 
Zhunko:
It's the same thing but through the ass. Battered operations are primary. They're faster.
You could argue. The result of a "fast" bitwise operation will have to be processed afterwards somehow.
 
For example, try "packing" three numbers in the range 0...1023 into a single int followed by unpacking exclusively by bitwise operations and you'll see what "through the ass" means
)))
 
MikeM:
For example, try to "pack" three numbers in range 0...1023 into one int followed by unpacking exclusively by bitwise operations and you will understand what "through the ass" means
)))
Packing is calculations. Only at once.
 
Then it will indeed already be easier, simpler and faster to pass variables by reference, and to return success, or failure (error) of the operation.
 

Yeah... can't you just use global variables? And change them in a function... even if it's 100500 parameters....

global at the level of the EA - not the terminal....

 

anything you can't do, anything you can't do :)

it's just a question of code semantics, nothing more.

 

The question arose as follows. Rough testing by reference points and testing by all ticks.

I know that testing by control points produces a rougher variant of a price chart. But I want to understand if a candlestick is drawn within one tick in the tester, the hell with it. But its HLOC corresponds to the real one? I am interested, first of all, in peaks (high/low). The real tick chart spikes are reflected in the HLOC of the candlestick in the rough test? What is inside of it (candlestick) is not important for me. But the correspondence of candlesticks and the main thing - peaks - to real ones is critically important. The testing on all ticks is very different. Why?

 

as someone said earlier - it's possible that the spread can be changed by the terminal - try disconnecting from the internet when testing - or rather, not logging in with the terminal...

and then the "tick" result will repeat from time to time...

And look at the algorithm - opening and closing trades... If you use any indices... they usually redraw on zero bar...

... so it turns out that the MAA crossed the price once... but on ticks it will turn out that it happened 1000 times during your period... (conditionally)

 
About the candle formation, yes. The most important question is whether HLOC is appropriate. I.e. is the RMS of ticks used to calculate rough candlesticks or are peaks taken into account?
 
are taken into account... It's kind of like testing takes data from lower TFs... and use it to model price behaviour in a candlestick... including HLOC
1...160161162163164165166167168169170171172173174...632
New comment
Reason: