synthesisable floating point

[Bruce Warkentin]

Hi

I'm looking for VHDL floating point (IEEE-754) code to do 32-bit adds and
multiplies. I've found a couple out there but have no idea as to their
reliability/performance. Does anyone have any experience with
this? Any sources to recommend?

Thanks much

Bruce

--

 

[Mike Treseler]

Bruce Warkentin wrote:

I'm looking for VHDL floating point (IEEE-754) code to do 32-bit adds and
multiplies. I've found a couple out there but have no idea as to their
reliability/performance. Does anyone have any experience with
this? Any sources to recommend?

Consider doing this in software.
A pentium uP can do a IEEE-754
multiply in 3 or 4 clock ticks.

-- Mike Treseler

 

[Bryce]

I'm looking for VHDL floating point (IEEE-754) code to do 32-bit adds and
multiplies. I've found a couple out there but have no idea as to their
reliability/performance. Does anyone have any experience with
this? Any sources to recommend?

Check out http://www.annapmicro.com/ . They definitely have a
floating point VHDL library, but you may have to ask for it--I don't
see the API on their web site.

 

[Jim Lewis]

There is an IEEE effort going on under 1076.3
to create floating point packages.

For details see: http://www.eda.org/fphdl/

Cheers,
Jim
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Jim Lewis
Director of Training mailto:Jim@SynthWorks.com
SynthWorks Design Inc. http://www.SynthWorks.com
1-503-590-4787

Expert VHDL Training for Hardware Design and Verification
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~


Bruce Warkentin wrote:

Hi

I'm looking for VHDL floating point (IEEE-754) code to do 32-bit adds and
multiplies. I've found a couple out there but have no idea as to their
reliability/performance. Does anyone have any experience with
this? Any sources to recommend?

Thanks much

Bruce