Problems with numerical accuracy?

Vegard Aaker
Thu May 31 05:56:00 GMT 2012

I only managed to print 16 bits of accuracy on the embedded board. But
I am not certain that printing the FP value will give me the answer to
how many bits are used. Do anyone else has a better suggestion to how
I could check the accuracy. I cannot find any information about it on
the spesification of the real time OS.

On the computer I use doubles.

On Wed, May 30, 2012 at 11:32 AM, marco atzeri <> wrote:
> On 5/30/2012 11:08 AM, Vegard Aaker wrote:
>>  Hi,
>> I am writing a optimization algorithm that solves a quite large linear
>> system of floating point numbers. The algorithm works well on a PC,
>> and should be quite robust to floating point errors. But when I am
>> using a microcontroller the algorithm crash, most likely due to
>> numerical issues.
>> I am using the Ethernut 5 board and Nut/Os that use newlib as the
>> floating point library. Now I am wondering what kind of accuray I can
>> expect from newlib compared to for example a PC (running windows)?
>> Should I expect much more accuracy if the microcontroller contains a
>> FPU?
>> Kind regards
>> Vegard Aaker
> number of digits for your calculation on board and PC ?
> On PC are you using floats or doubles ?
> A windows FPU could make calculation with 64 or 80 bits.
> Regards
> Marco


- Vegard

More information about the Newlib mailing list