coding in c until I go completely insane

(1+2)/10
trolled

The IEEE flag for compilers is a lifesaver. Coming for someone who writes scientific applications on supercomputers, it's a pain in the a$$

Edit: Forgot about 17 digits. You'd need 16 bytes then. So long double should work.

Meanwhile Excel =SUM() function done this peacefully 😆😆

Song name at the end?

bro just use a double :sob:

0.1 + 0.2 = 0.3

1 attemp only: I guess that the final 4 is something about the bits (is a float number, then it has 4 bits of lenght)

Damn, that’s annoying.

No wonder my conversion from farhenheit to celsius was wrong

❤❤❤❤❤

The reason is because of the floating point representation in computers. Since computers only store binary numbers some small errors are introduced when doing the conversions. If you want to learn more about it you can search IEEE 754 representation of floating points for starters.

"Python help me out here" is such a mood 😂

I actually suffered a similar bug in a production system that was supposed to work on really small percentages (which again blow up to be big after being multiplied by number of clients which ran in 100s of millions). This was in a big 3 software firms.
And I was a new hire.

"Ohhh, whaatt the fuck" sounds too relatable

Acho que consegui...

What about to use long double. I know it's very implementation defined but why not to try?
For example, clang supports 128-bit long double on software level when using some compilation flag...

What IDE do you use???

IEEE-754, baby :). Try having fun with some floating point DSP code. Shout if you get a denormalized number ;)

guy went ask help to Java LMAO

Integers are your friend.
Just assume that integers are tenths.
1 + 2 = 3

limited attempts is literally hell

For my great suprise C# actually gets this one right.

Now I'll be honest, you did one thing wrong, you used pure float in C, you should have gone with long double (128 bits).

well, you could have used a more mathematical proof
1+2=3
so by dividing everything by 10
(1+2)/10=3/10
so, by printing the result of (1+2)/10, I thing you would have been able to get the level of accuracy you want, since it will be adding ints rather than floats

I just started learning how to code today, so I have no idea of what I just watched...I do know he fucked up tho😂

Aren't Floats designed to be warped so it's more accurate closer to zero, and less accurate at super high numbers? But because of the warping, .1 + .2 would never actually equal .3, would it?

This challenge was impossible to begin with.

why are you using argc and argv if you're not going to put any arguments in your program?

Bro it's just a bonus, don't force javascript to do what it can't for just a bonus.

float problem tho.

Why « int argc, char **argv » in your « main » ? …

Maybe mess around with compiler settings, study the floating point standard?

You make coding fun again. Re ignited the fire in me. Thanks bro from christchurch new zealand. Chur

IEEE 754 shenanigans be like

insane? i was insane once. then they trapped me in a room. a rubber room. a rubber room with rats. the rats made me insane

Well, you see...
0.30000001192092896, if rounded, will still be 0.3. So arguably, C answer is still correct.

Use Ruby: "0.1".to_r + "0.2".to_r == "0.3".to_r
=> true
... Subscribed to save the sanity ¯\_(ツ)_/¯

IEEE754 standard for finite precision floating point. In C, "float" is usually 32-bit single precision. For your 2nd and 3rd attempts, those were double precision.

Easy. Just convert to a string, append more zero's and convert back to int.

i find it funny that i only know why this happens because of an sm64 video

Still now almost all hardware level programming done by either C or C++..
So my favourite language for bit Byte and register level Access...

Should have used doubles or long doubles.

Java Masterrace would have saved you

Tbh it's not really a C problem, inaccuracy is an inherent property of all floating points numbers... but it's fast so who cares?

Computer can't do math because the memory is limited to 32 bit or 64 bit 😂😂

Does anyone have the technical explanation of why this is the case?

this thing won't work because in C floating-point numbers are represented using binary approximations
so you gotta use gmp or whatever, pick your poison buddy

Printf your original answer and tack on (concantonate?) a string of 15 zeros.