Floating-point 7.1 input precision

[crybe86]

Hello,

I want to convert a 10 bit signed Integer into Float format with the Floating-point IP 7.1

When I enter the following values in the "Precision of Inputs" tab, I still get a 16 bit input. Can someone explain this behaviour? I couldn't find a reason for this in the documentation, but maybe I have overlooked something? Or is it a bug?

Many thanks in advance!

Christian

[balkris · Accepted solution]

you need to paid dummy bits to make sixteen. You can use LSB 10 bit for your input and pad rest to 6 bits . this is requirement of AXI interface

check core product guide for detail

http://www.xilinx.com/support/documentation/ip_documentation/floating_point/v7_1/pg060-floating-point.pdf

[crybe86]

Thanks for your explanation! That cleared it up.