Hello!

In this simply example of code I got different results, depended on the placement of db and the length of the defined variable.
The program prints “Hello World!”, the only difference is where the definition of the string is located.



(1) The result is "llo World!" - missing two first characters.

When you put db on the end of the program(2) the result is "Hello World!" - correct one.

I can't see the difference, and why this should be wrong.

When you write "string1 db 'Hello World!', 13, 10, 0" or “string1 db 'Hello World!', 0” works in both cases, but I don't understand why.

From some trials I know that 10(line feed) is not the case, it depends of the string1 length, but why it should make a difference where is placed.

The program is executed in the first line !
You have to put data separate (into another segment) or at the end of the program. You could also use jmp print in the first line to skip the data section.

Your code is executed from the beginning regardless if you think this is text or code. The CPU doesn't know and try to execute your data "Hello World" which gives unpredicted results.

Exactly, you are making a .com executable, right? This format is just a flat binary that starts executing from the first byte, it doesn't have an "entry point". So your data gets executed as if it was code, who knows what kind of crazy instructions result from the "Hello World!" string... You could check with a DOS debugger...