The function of decomposing color into shades. - page 3
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
HSL
I had a look. Horrified. Looked again. :)
The question is, how long does it take to get into solutions like this? I came to the conclusion that I'd get more value if I found the solution myself.
...
* * *
1. As correctly pointed out above - there are HSL and HLS systems, which have long been standardised, and transitions to them from RGB are done with standard formulas. There's also the Lab system. What's the point of reinventing the wheel ?
...
2. I'd like to see concrete examples of applications, even without revealing the source code - just the idea of where it all could be usefully applied ?
1. About the bike. Look, George, Nikolai provided a solution from CCanvas class. As a developer, what do you think - is it convenient to read/use/develop ? Personally, I find it uncomfortable. Here's why:
Now about my solution.
So, this solution is not a bicycle at all like you think.
A lot of people write small useful attachments. But so far I see that only fxsaber has presented a library which, in my opinion, is very useful for literally everyone. I don't know of such a voluminous, yet generally useful work. In this case, with Canvas, it's even sadder - I can't imagine where it all could be applied in a way that would be useful to any discernible part of the forum audience.
Yeah, I'm sad too George. For almost three years on this forum I've been trying to explain why trading software needs a GUI. I have solved a huge number of tasks to create it. Among these tasks were both simple and incredibly complex. But I have never solved a more reinforced task than to explain to people here why they need it.
What is absolutely clear to me is incomprehensible to them. It's a phenomenal difference in understanding.
You write hundreds of trading robots. Don't those robots need adjustments? They do. So - they need adjustments. Controls.
You trade in the market. Don't unexpected events happen in the market, and you want to be informed right away? Yes, they do. So you need message windows.
You do research on market patterns. Isn't it better to observe patterns through charts and tables? Better. So you need windows with charts and tables.
You're optimizing robots. Isn't it better to display the results of optimization in report windows? Better. So you need report windows.
You collect trading statistics for each trading robot. Isn't it better to show those statistics in details in the report windows? Better. So, you need windows to display statistics.
The same can be said about risk management settings, about AI, MO and other things. The more complex the program, the more settings, tables and graphs it must have. The more it has to interact with the user.
After all this, do you still think that trading programs don't need GUI? If so, I have nothing more to say to you.
George, I debug the code in my head. :) I've always done that. I look for errors by reasoning logically. The Alert() function helps a lot;
That's cool. I can do that too, although I use output to a log file, but it's a lot of stress for me. It's much more convenient to look at variables directly.
That's cool. I can do that too, although I use output to a log file, but it's a big strain for me. It's much more convenient to view variables directly.
The thing is, you can do without a debugger, but I wasn't ready to sacrifice the convenience of native language development for it.
I always choose the best solution. Trust me. If I don't know it, I know where to find it. Developing in a native language gives you an advantage that no debugger can cover.
Yeah, I'm sad too George. For almost three years on this forum I have been trying to explain why trading software needs a GUI. I have solved a huge number of tasks to create one. Among these tasks were both simple and incredibly complex. But I have never solved a more reinforced task than to explain to people here why they need it.
What is absolutely clear to me is incomprehensible to them. It's a phenomenal difference in understanding.
You write hundreds of trading robots. Don't those robots need adjustments? They do. So - they need adjustments. Controls.
You trade in the market. Don't unexpected events happen in the market, and you want to be informed right away? Yes, they do. So you need message windows.
You do research on market patterns. Aren't patterns better observed through charts and tables? Better. So you need windows with charts and tables.
You're optimizing robots. Isn't it better to display the results of optimization in report windows? Better. So you need report windows.
You collect trading statistics for each trading robot. Isn't it better to show those statistics in details in the report windows? Better. So, you need windows to display statistics.
The same can be said about risk management settings, about AI, MO and other things. The more complex the program, the more settings, tables and graphs it must have. The more it has to interact with the user.
After all this, do you still think that trading programs don't need GUI? If so, I have nothing more to say to you.
The thing is, Peter, that all of this I believe should be "once and for all, permanently" crammed into the bot. My trading robots don't have ONE (!!!) setting. If I were to offer them to anyone (for free or for money) - they would have only one setting - the percentage of risk. That's it!
Look.
Of course, the robots are constantly being adjusted. However, these corrections, in my opinion, should not come from gui interface. Ideally, the robot should automatically search the history, select the best parameters, save them for itself and keep working. It does not have to make any adjustments "from a human".
Information on unexpected events is indeed required. But, in my opinion, the output is enough. Message window - how is it better than the same Print? The window is useful when we need to display information in a structured way, in the form of tables, links and the like. However, this is not at all a case of "important events" !
Outputting optimization results - again, for human perception, it is true, it is kind of convenient to have a separate window. But how should these results be used? In my case, they are used to create a block of code that is then "loaded" into the bot. Don't you think the window is unnecessary in this case? The optimization results are generated in my file as a block of code that is then written to the bot through copy-paste. A separate window on the chart, I'm afraid, will not allow for that.
The same goes for statistics - I don't need them in windows, I need them in report files!
Message to the user, I've often said that I am a supporter of the simplest and "dumbest" TS. Such systems don't need any "user interaction".
Summary:
I would love to have a GUI but I'm afraid that the GUI I have doesn't cover my needs and that its complexity increases dramatically.
However, I admit that others may find the GUI useful. First of all - for semi-automatic trading.
That's why - I say, Peter, you need to "dramatise the idea". Then the audience will get interested, and then the users will get interested.
Take the example of the TC league. Now few people remember the thread in which I introduced the idea of the League a year and a half ago. Mostly there was skepticism and only a couple of interested participants. In fact - same thing as you, I can see direct analogies. Anyway, on another forum - one member told me so - "dramatise the idea". Make your work memorable, or better yet, vivid. He gave me some examples, of which I only remember Lavoisier (Lavoisier was proving the phlogiston theory wrong. Experiments on burning various substances in a sealed vessel did not impress his colleagues. So he burned a diamond. The experiment was very expensive, and it attracted the attention of many, the experiment was talked about, and at the same time - they started talking about the phlogiston theory, against which more and more facts began to appear). So this participant pointed out to me that I should open a branch in which to regularly publish the results of the TC set (at that time the TC League was not yet so called), and arrange something like a "competition report" to make it interesting. I opened up the branch (past), and made sure it did. Even found two people who helped me with testing. Now - a new branch has been opened. And any regular on the forum, I think, knows what the TC League is all about. Many chuckle ironically, others watch with interest. But I'm sure everyone has long since "linked" in their mind my name, the Bloody Fork (my avatar) and the TC League. By the way, the "sound", spectacular name "TC League" seems to play a part too.
You have to have something ready that would be of interest to the people. Say, here, with the same, semi-automated trading, in which windows and GUI are very appropriate. Especially since the GUI for "brightness" has a lot of possibilities.
User interaction, I have said many times that I am in favour of the simplest and "dumbest" TCs. Such systems should not have any "user interaction".
George, there seem to be a lot of supporters ofthe simplest and "dumbest" TCs here. Therefore, I can't explain anything to them...))
Of course, robots are constantly being corrected. However, this correction - in my opinion - should not be from the gui interface. Ideally, the robot should automatically run through the history, select the best parameters, save them and go on working. It does not have any human settings.
Dumb robots do not really need a lot of settings. A simple strategy, optimization and go!
Information on unexpected events - indeed, output is required. But in my opinion, the output is good enough. Message window - how is it better than Print? The window is useful when we need to display information in a structured way, in the form of tables, links and the like. However, this is not at all a case of "important events" !
Why do we need a foreign car, if the cart serves us well?) What is there, structured output of information? What for?! ))
Summary:
I would love to have a GUI but I'm afraid the GUI I have doesn't cover my needs and it increases the complexity of the software seriously.
However, I admit that others may find the GUI useful. First of all - for semi-automatic trading.
That's why - I say, Peter, you need to "dramatise the idea". Then the audience will get interested, and then the users will get interested.
Take the example of the TC League. Now few people remember the thread in which I introduced the idea of the League a year and a half ago. Mostly there was skepticism and only a couple of interested participants. In fact - same thing as you, I can see direct analogies. Anyway, on another forum - one member told me so - "dramatise the idea". Make your work memorable, or better yet, vivid. He gave me some examples, of which I only remember Lavoisier (Lavoisier was proving the phlogiston theory wrong. Experiments on burning various substances in a sealed vessel did not impress his colleagues. So he burned a diamond. The experiment was very expensive, and it attracted the attention of many, the experiment was talked about, and at the same time - they started talking about the phlogiston theory, against which more and more facts began to appear). So this participant pointed out to me that I should open a branch in which to regularly publish the results of the TC set (at that time the TC League was not yet so called), and arrange something like a "competition report" to make it interesting. I opened up the branch (past), and made sure it did. Even found two people who helped me with testing. Now - a new branch has been opened. And any regular on the forum, I think, knows what the TC League is all about. Many chuckle ironically, others watch with interest. But I'm sure everyone has long since "linked" in their mind my name, the Bloody Fork (my avatar) and the TC League. By the way, the "sonorous", spectacular name "TC League" seems to play a part too.
You have to have something ready to go, something that would be of interest to the people. Say, here, with the same, semi-automated trading, in which windows and GUI are very handy. Especially since the GUI for "brightness" has a lot of possibilities.
It's more enjoyable for me to read...))
George, there seem to be a lot of supporters ofthe simplest and most "oaky" TCs here. So, I can't explain anything to them...))
Dumb robots really don't need a lot of tuning. A simple strategy, optimization and go!
Why do we need a foreign car if the cart serves us well?) What is there, a structured output of information? What for?! ))
So far, there is only a painted foreign car. Why do people need a painted foreign car when the cart is real and serves?