Derive algebra

[Фролов]

Tell me how you can implement + 0.5 for example double

Code: Pascal  [Select][+][-]

1.     d := 10;
2.     for i := 0 to 50000000 - 1 do
3.     begin
4.       d := d + 150.5;
5.     end;
6.  

Should be - 7525000041,32991, but in fact 7525000010

0.1 = 0.00011001100
0.5 = 0.1

[wp]

What is the problem? What is d?

You are adding 50 million times the value 150.5 to 10, i.e. you are calculating 10 + 50E6*150.5 =7975000010. And this is exactly what the following program is printing:

Code: Pascal  [Select][+][-]

1. program Project1;
2.  
3. {$mode objfpc}{$H+}
4.  
5. var
6.   d: Double;
7.   i: Integer;
8. begin
9.   d := 10;
10.   for i := 0 to 50000000 - 1 do
11.   begin
12.     d := d + 150.5;
13.   end;
14.   WriteLn(d:20:1);
15.  
16.   d := 50e6*150.5 + 10;
17.   WriteLn(d:20:1);
18.  
19.   ReadLn;
20. end.

[Фролов]

What is the problem? What is d?

You are adding 50 million times the value 150.5 to 10, i.e. you are calculating 10 + 50E6*150.5 =7975000010. And this is exactly what the following program is printing:

Code: Pascal  [Select][+][-]

1. program Project1;
2.  
3. {$mode objfpc}{$H+}
4.  
5. var
6.   d: Double;
7.   i: Integer;
8. begin
9.   d := 10;
10.   for i := 0 to 50000000 - 1 do
11.   begin
12.     d := d + 150.5;
13.   end;
14.   WriteLn(d:20:1);
15.  
16.   d := 50e6*150.5 + 10;
17.   WriteLn(d:20:1);
18.  
19.   ReadLn;
20. end.

Yes d = Double

I would like to repeat this. For one measure(One line). There is a possibility?

Code: Pascal  [Select][+][-]

1.     d := 10;
2.     for i := 0 to 50000000 - 1 do
3.     begin
4.       d := d + 150;
5.       d := d + 0.1;
6.       d := d + 0.1;
7.       d := d + 0.1;
8.       d := d + 0.1;
9.       d := d + 0.1;
10.     end;

[guest60499]

I think you are running into the issues with floating point numbers as outlined here.

[wp]

I still don't understand. I still don't know what you want to achieve and what is wrong. Maybe you should post the question in your native language.

[Фролов]

I still don't understand. I still don't know what you want to achieve and what is wrong. Maybe you should post the question in your native language.

Выполнить за один такт, + 0.5 для типа double  (Run for one clock cycle, + 0.5 for double type)

7.5250000413299112E+009

Code: Pascal  [Select][+][-]

1.     d := 10;
2.     for i := 0 to 50000000 - 1 do
3.     begin
4.       d := d + 150;
5.       d := d + 0.1;
6.       d := d + 0.1;
7.       d := d + 0.1;
8.       d := d + 0.1;
9.       d := d + 0.1;
10.     end;
11.     WriteLn(d);  
12.  

Это отнимает 80% 1.088237 sec Я думаю, это бы занимало порядком 0.203127 sec

(It takes away 80% 1.088237 sec I think it would take about 0.203127 sec)

[guest60499]

Are you trying to create a delay by performing floating point operations?

[Фролов]

I found what I need. But it generates classes and shits into memory ... But tearing out the right thing is possible

- Я нашёл то, что надо. Но оно порождает классы и гадит в память... Но выдрать нужное возможно

https://github.com/rvelthuis/BigNumbers

Code: Pascal  [Select][+][-]

1.     d := BigDecimal.Create('10.000');
2.     d := d + 150.5;
3.     d := d + 150.5;
4.     d := d + 150.5;
5.     d := d + 150.5;
6.     d := d + 150.5;