Which unit is used in computer for arithmetic calculation?

Answer 1

The bit. One bit contains a binary memory cell that will return a 1 or a zero.

In abstract terms, the bit tells the system the answer to ONE UNAMBIGUOUS question with a yes or a no. In Boolean math it is called true or false.

In hardware terms, when current flows through the cell, it will either boost it (High=1=Yes) or lower it (Low=0=No) returning the answer to the question.

For making accurate arithmetic calculation, you must always use integers (-infinity...-4, -3, -2, -1, 0, 1, 2, 3, 4....infinity) using floating point numbers or floats (3.14, 1.7, 0.99999999999....etc) is banned by the IEEE and ISO 9000 and other standards organizations for good reason, if your ERP system uses floats and calculates enough transaction using floats, you will either lose tons of $ or have some 'splainin to do to clients and the IRS.

if the calculation involves money use integers.... so $1.50 must be read in as 150 cents and returned as another integer (let's say we calculated the tax and the answer is $1.61, the integer 161 must be returned and converted to $1.61 OUTSIDE of the program that did the financial calculation).

Point = use ints whenever poss. and always use ints for $.