Re: [vhdl-200x] Regarding Fixed Point Algorithmic User's Guide

From: David Bishop <dbishop@eda.org>
Date: Wed Oct 09 2013 - 08:42:55 PDT
On 10/09/2013 08:06 AM, Sharad Sinha wrote:
> Hi group,
>
>   I had a look at the Fixed Point Algorithmic User's Guide. I am not 
> sure about the utility of including functions to calculate exp(), 
> log() and some trigonometric functions like sin(). The reason I am not 
> sure is that vendors, at least the FPGA ones, provide IP cores to do 
> these operations. Of course they have their own limitations etc. but 
> that is another issue.
>
>  Is there really a need to include some of those functions in the 
> package? If yes, where would they find most utility? Synthesis, 
> testbench or both?
My thought here was that synthesis vendors could create algorithms that 
could perform these operations.

Also, if you create your own "exp" or "log" function, you now have 
something to verify it against.   Which is basically what I did for some 
of these.

If a function can be done in REAL, then there should be a way to do it 
in fixed and float as well...


-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.
Received on Wed Oct 9 08:43:21 2013

This archive was generated by hypermail 2.1.8 : Wed Oct 09 2013 - 08:43:31 PDT