Why "21030018" convert to float normal , but "21030019" shown as scientific notation

[BruceQiu1996]

i can't get answer from anywhere.

platfrom : .net 7

[yaakov-h]

This seems like an IDE issue and not a language issue, however you haven't even mentioned what IDE you're using.

[BruceQiu1996]

visual studio 2022

[TahirAhmadov]

float is a floating point number, which means it's stored in an in-precise manner from the get-go. In your case, the way the number is converted to the underlying bits changes when you go from 21030018 to 21030019 - for the former, it's able to preserve all the digits, and for latter, it's not. The precision is specified to be 6-9 digits, and your number is straddling that line, which explains this behavior.
https://learn.microsoft.com/en-us/dotnet/csharp/language-reference/builtin-types/floating-point-numeric-types
If you are worried about preserving all the digits, you should use decimal (or even int/long), not float/double.
PS. This probably belongs more to Stack Overflow and not C# language design board.

[huoyaoyuan]

20130019 is exceeding the precision of float.

        float a1 = 21030018f;
        Console.WriteLine(a1);

        float a2 = 21030019f;
        Console.WriteLine(a2);

outputs on .NET Core:

21030018
21030020

However, I have no idea why the two values show differently in VS debugger. ToString uses both normal form on .NET Core, and both scientific notation on .NET Framework.

[BruceQiu1996]

Hello, can you tell me why the 21030019 exceeded the accuracy, but 21030018 did not. I can't understand it

[jnm2]

https://www.h-schmidt.net/FloatConverter/IEEE754.html may help you visualize what the 32 bits are doing to hold the two numbers.