Hello Gurus:
Is there a 1/2 precision (16bit) floating point library for the 280x series of micro controllers?
I have 32 bit (float and double) routines, and also 64 bit (long double.)
Thanks, Mark.
This thread has been locked.
If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
Hello Gurus:
Is there a 1/2 precision (16bit) floating point library for the 280x series of micro controllers?
I have 32 bit (float and double) routines, and also 64 bit (long double.)
Thanks, Mark.
I believe we only have support for single precision (32-bit) and double-precision (64-bit) in our floating point library.
Thanks,
Sira
I am curious about the reason for asking - do you mind telling me what kind of application this is for?
Thanks,
Sira
Hi Sira:
Actually, for the current project I am working on, this is unrelated.
We are making an audio processor using an ADAU1701 DSP, and I was researching using single precision (32bit) floats vs double precision (62bit) long doubles and ran across something during my general searching on the internet about half precision (16bit) math.
I like to try to think outside the box as much as I can and it is always good to have as much related info rattling around in the back of my mind (and, yes, as I get older, my mind is becoming more of a rattle trap.)
Mark.
Mark,
Thanks for letting me know. It is definitely something I can keep an eye on going forward.But the trend continues towards more computing power and more performance, so this may be a niche area that we may not end up focusing on.
Thanks,
Sira