Meme detail

Text detected in the image

STOP USING FLOATS BINARY DATA WAS NOT SUPPOSED TO HAVE DECIMAL PARTS YEARS OF COMPILER DEVELOPMENT yet NO REAL-WORLD USE FOUND for anything other than char and int Wanted to use decimal numbers anyway for a laugh? We had a tool for that: It was called FIXED-POINT ARITHMETIC 'x==x can be FALSE',' is a number', 'the sum of 1 and 2 is 1 0.30oouoooo04'—statements dreamt up by the utterly Deranged LOOK at what Floating-Point Numbers have been demanding your Respect for all this time, with all the circuits and data types we built for them (This is REAL Floating-Point Arithmetic, done by REAL computers): fld TBYTE PTR [rdi] fyl2xp1 x2 = number * 0.5F; fdecstp = number; fld st(@) i * ( Long * ) &y; fstp TBYTE PTR [rdi] fld TBYTE PTR [rsi] i = 0×5f3759df -( i >> 1); fstp QWORD PTR [rSp+16] V = * ( float * ) &i; movsd XmmO, QWORD PTR [r50+16] V = y * ( threehalfs - ( x2 * y * y ) ); ucomisd xmm1, xmme sign exponent (8 bits) significand (23 bits) 0 0 1 1 1 1 1 0 0 0 1 0 = 0.15625 3i 30 2322 (bit index) 0 'Hello, I want to know if f is a real number and less than -1 please?' Sure,that'll be f!=f?0:-1/0.f==fllf==-1.f/0?0:-1.f>f' They have played us for absolute fools

Popularity