The function of decomposing color into shades. - page 18
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
You're right about "R = Source_R", but I left it that way because it makes more sense how the algorithm works.
The algorithm certainly needs some more work. It's not perfect. Maybe at the end I'll come up with an algorithm similar to yours. But by going all the way, I'll get incomparably more. After all, I will pass a path of knowledge and understanding, and therefore the creative area in which I implement my algorithm will be much wider than if I simply copied someone else's solution. Therefore, I always choose my own path.the line
is the source of the mad brakes.
Try profiling it.
This way will be much faster:
the line
the source of the rabid brakes
Try profiling it.
It'll be much faster that way:
Shit, my algorithm has the same speed as yours))
Thank you, Nikolai!
I'm just not familiar with this kind of operation.Shit, my algorithm has the same speed as yours)).
Thank you, Nikolai!
Not yet. There are still a couple of things to fix.
Removed redundant features. Now only the main function:
You got me, Andrew. :)
Well, yes, 18 pages are devoted exactly to discussing the problem. And not a single word about OOP, Russian variable names and future of autotrading.)
Well, yes, 18 pages are devoted to discussion of the problem. And not a single word about OOP, Russian variable names, and the future of autotrading )
Why, I told about Cyrillic several times only in this thread. But Peter is unbreakable in the matter of OOP and Cyrillic, that I gave up and think that my "cheating" over him is over until I see his code with his own classes without Cyrillic and using debugging. I've already repeatedly told him that without OOP and debugable style (i.e. without Cyrillic) GUI will never see the light of day, and if it does, it will be immediately booed off and pelted with rotten eggs and tomatoes. What can you do, if Peter with his fantastic persistence and efficiency (that he manages to write code without the debugger) has the same fantastic degree of conservatism and stubbornness.
But without Peter this forum would not be so charismatic. ))It's probably going to be a little faster that way:
I agree.
I agree.
Nikolai, let's be objective.
By removing the drawing that is irrelevant in this test, I made the script as simple as possible, and revealed the following discrepancies:
As you can see, the centre of the colour range in your algorithm is shifted upwards. The brightest band should be in the centre. This is confirmed by the Windows palette:
Below, I attach a script for testing:
Box 127 is the centre of the shade array. In this cell, it should be the same colour as in the palette. I have a margin of error of 1, you have 63.
The blurring and over-painting prevented me from concentrating on verifying that the algorithm was working correctly. That's why I suggested checking against the printout in Alert.
SZY. Actually, the centre of the array is cell 128, but it doesn't change the essence.New interesting facts:
No offence Nikolai, but your algorithm is faster and shorter (fact), but it works incorrectly (also fact). It lacks a well-thought-out conceptual framework.
My algorithm originally lacked speed (but you helped) and debugging. At the same time, it had a well-thought-out and correct concept from the beginning.
That is why there is such a difference.