So adding 2's complement of -8 to binary numbers will give wrong result

2's complement - 8 and binary -8 are different i suppose

A, right

Binary -8 => 10001000 2's complement -8 =>11111000

-8 + 12 should give u 4 whose binary is 00000100

Oh yes

Wait

So the 1 is coming additional

I feel like my NN isnt learning that *certain* words are german, but that LONG words are german

;_;

Count the bits

And the input?

If input is 8 bit long, output is 8 bit long

I feel like my NN isnt learning that *certain* words are german, but that LONG words are german

What's NN?

The 9th is cutted

That's overflow

No problem ;)

Do they make you solve it at school?

Do they make you solve it at school?

Define you

Wozniak

Hey

New member? Read the pinned message

:D

What? Show me the operation using monospaced format

Mmm, no?

https://en.wikipedia.org/wiki/Two%27s_complement

You've written a different number up there

Here, play with it :D

#include <iostream> #include <bitset> int sign(const int a) { return a > 0 ? ~a + 1: a; } int main() { const int a = 11; const int b = sign(a); std::cout << a << " -> " << b << "\n"; std::cout << std::bitset<8>(a) << " -> " << std::bitset<8>(b) << "\n"; std::cout << "a + b = " << a + b << "\n"; return 0; }

Result: 11 -> -11 00001011 -> 11110101 a + b = 0

You need to use the same number of bits

Inputs and output have the same number of bits

No

0100100+ 1101100 ------- 1|0010000

You CUT the overflow

You don't change nothing

No

You're starting from a bad point

You can't sum numbers with a different number of bits in notation

36 is either 0100100, 00000000000000100100 or 00100100, but NOT 100100

36 is either 0100100, 00000000000000100100 or 00100100, but NOT 100100

Last one is, as an unsigned, 36. The others are +36 if ints, or 36 as unsigned

So it can't be rapresented in Two's complement

You need more bits

Like 011111111

-255 is at least 100000000

55 is something that you'll do

-255 is at least 100000000

-255 is not encodable in 8 bit

-255 is not encodable in 8 bit

They're 9 :)

yea

no, -255 would be 111111111

where the first bit represents the sign

practically you usually put the sign bit way upfront

100000000 00000000 00000000 11111111

oh yea

no, -127 would be 11111111

7 bits represent the number => [0, 127] 1 bit for the sign

oh yea

He means that 11111111 means 255 if it's an unsigned char

again you cannot represent -255 with 8 bits

sorry?

not at all

you need atleast a bitfield of 9 bytes

using 9 bits (one's complement) (range [-255, 255]) -255 would be 1 1111 1111 255 would be 0 1111 1111 using two's complement (range [-256, 255]) -255 would be 1 000 0001 255 would be 0 1111 1111 by duplication of the first bit yo can keep the same value with larger bitfields

100000000 + 000110111

you can play with it here http://www.binaryconvert.com/result_signed_int.html?decimal=045050053053

Lol

#include <iostream> #include <bitset> typedef std::bitset<16> bits; int change_sign(const int a) { return ~a + 1; } int main() { const int a = 255; const int b = change_sign(a); std::cout << bits(a) << " -> " << a << "\n" << bits(b) << " -> " << b << "\n\n"; std::cout << " " << bits(a) << "\n" << " " << bits(b) << "\n" << " -------- + \n" << "1|" << bits(a + b) << std::endl; return 0; }

Result: 0000000011111111 -> 255 1111111100000001 -> -255 0000000011111111 1111111100000001 -------- + 1|0000000000000000

Never mind, I'm just messing around

1 0000 0001 ( -255) + 0 0011 0111 (55) ----------- 1 0011 1000 (-200)

so I have my SFINAE container that returns true, but how can I make it return false if it fails?

-255 -> 11111111 11111111 11111111 00000001 55 -> 00000000 00000000 00000000 00110111 -200 -> 11111111 11111111 11111111 00111000 What's wtong about that? :D

dudes stop discussing binary, I have two hours of music theory later on and I don't want to have an headhache when doing that xD

dudes stop discussing binary, I have two hours of music theory later on and I don't want to have an headhache when doing that xD

Oh, right. Now let's talk about shifting...

STAWP

xD

STAWP

Now it's a binary channel. I want to rename it to 0/1

Now it's a binary channel. I want to rename it to 0/1

don't

i was just gonna ask about bitstring representation of sets

dudes stop discussing binary, I have two hours of music theory later on and I don't want to have an headhache when doing that xD

Okay, have a nice music theory. I'm stwapin

Okay, have a nice music theory. I'm stwapin

thanks ❤️

music theory is only in about 2 hours