How do you programmatically distinguish between a PROFESSIONAL fetus and a DILIETAN fetus? - page 17

 
DmitriyN:

Not funny.

I'm serious. There are hardly any fish here.

Have you roughly set yourself a task? Have you prepared a sample of nerds and pros to check?

Which one of you is theXpert or FAQ the best programmer? How do you rate it? How do you compare? Do you think it's impossible?

I don't think about it. And I'm not a programming guru.
 
DmitriyN:
I can't say, I don't know. I know I won't even look at a 1kb EA or a 300kb EA.


Dimitry, it's not about the size. Although, I'm not looking at such EAs either. 5-25, my opinion. But there are exceptions.
 
DmitriyN:
I can't say, I don't know. I know I won't even look at a 1kb EA or a 300kb EA.

Do you count inludes with libraries in the file size?
[Deleted]  
TheXpert:
Are you counting the inludes with libraries in the file size?

No, otherwise we get confused. Just pure mq4 code as a single file.

grell:
Dimitri, it's not about size. Although I don't look at such EAs either. 5-25, my opinion. But there are exceptions.

So the blue curve, you yourself have admitted that there is a certain golden range.

===

Another question. The number of externs. If there are none - bad. If there are dozens of them - bad, too. What is the optimal range and how to determine it?

 
DmitriyN:
So the blue curve, you yourself have admitted that there is a certain golden range.


You can't judge it that way. I dismiss the overly contrived and overly primitive, the implementation has little effect.
 
DmitriyN:
Another question. The number of externs. If there are none, bad. If there are dozens of them - bad, too. What is the optimal range and how to determine it?

6:)
[Deleted]  
grell:
6:)
You see, you have a middle ground here too.
 
DmitriyN:
You see, you have a middle ground here too.


You know as well as I do why there are six.
[Deleted]  
grell:
You know as well as I do why six.
I don't. Why not?
 
DmitriyN:
No, otherwise we get confused. Only clean mq4 code as a single file.

Hmm, then the correlation must be reversed. The bigger the file, the worse the encoder -- on average.

Then it still makes sense to measure the ex4/mq4 size ratio.