In fxpmath does the signed syntax represent the negative numbers as 1's complement only? If yes, how do I change it to 2's complement?

Question

from fxpmath import Fxp

c = Fxp(complex(3.2500+2.0625j), True, 8, 4) # (val, signed, n_word, n_frac)

print(c)

print((c).bin()) # binary representation

d = Fxp(complex(2.0625+5.2500j), True, 8, 4) # (val, signed, n_word, n_frac)

print(d)

print((d).bin()) # binary representation

y= Fxp(complex(c*d), True, 16,8)

print(y)

print((y).bin()) # binary representation**

This is the output I see-

(3.25+2.0625j)

00110100+00100001j

(2.0625+5.25j)

00100001+01010100j


(-4.125+21.31640625j)

1111101111100000+0001010101010001j

when I multiply these two numbers, the real part is -4.125 and the conversion to binary shows just the 1's complement of the answer, I need this to be shown in 2's complement, is this possible using the fxpmath function? The real part and the imaginary part again consists of integer and fraction bits each 8bit wide.

Please show the code for this multiplication & conversion to binary. — Scott Hunter, Sep 20 '22 at 17:32
@ScottHunter I have added the code in the main thread for reference. thank you for looking into it. — mohammed zeeshan, Sep 20 '22 at 17:51
The 'signed' in the fxpmath syntax is converting the negative number into 1's complement whereas I am looking to get the answer in 2's complement. — mohammed zeeshan, Sep 20 '22 at 18:00
Update your original question to include your code rather than posting it in an answer to your original question. — j_b, Sep 20 '22 at 18:00
@mohammedzeeshan: That's not how Stack Overflow is structured. It's not a thread, with a chain of replies; you've posted in the answer section. You should edit that information into the question instead. — user2357112, Sep 20 '22 at 18:01
Looking at the examples on the github page, it looks like the `bin` method already gives two's complement. Are you *sure* you're getting one's complement? — user2357112, Sep 20 '22 at 18:04
@user2357112 yes I am sure, ill go ahead and paste the output as well in the main question. — mohammed zeeshan, Sep 20 '22 at 18:06
@user2357112 my mistake, I was performing the complements for the integer bit separately, understood it now. thanks for your time. — mohammed zeeshan, Sep 20 '22 at 18:20

In fxpmath does the signed syntax represent the negative numbers as 1's complement only? If yes, how do I change it to 2's complement?

0 Answers0