Skip to main content
TBles.1
Associate
October 17, 2022
Solved

Floating-Numbers Problem with SPC560P

  • October 17, 2022
  • 3 replies
  • 1619 views

Hello,

I am working with a SPC560P and have a problem with float numbers. As soon as I write a calculation with float numbers in my loop, the loop becomes extremely slow (microsecond range for a simple calculation in the loop).

What can be the reason for this?

Do I have to activate or set anything else and if so where?

Please help me

Thanks

Timo Blessing

    This topic has been closed for replies.
    Best answer by ODOUV.1

    Hello,

    SPC560P is based on a powerpc core e200z0 which have no Floating-Point Unit.

    When you use floating point operations, it is calling software libraries that simulates an FPU on an integer CPU, so it is much longer than hardware calculations.

    Best regards.

    3 replies

    ODOUV.1
    ODOUV.1Best answer
    ST Employee
    October 17, 2022

    Hello,

    SPC560P is based on a powerpc core e200z0 which have no Floating-Point Unit.

    When you use floating point operations, it is calling software libraries that simulates an FPU on an integer CPU, so it is much longer than hardware calculations.

    Best regards.

    TBles.1
    TBles.1Author
    Associate
    October 17, 2022

    Hi,

    thank you for your answer.

    Is there any controller with an Floating-Point Unit, that is familiar with the SPC560P?

    Best regards

    ODOUV.1
    ST Employee
    October 17, 2022

    If you want to stay in SPC56 familly,

    there are the SPC564L or SPC56EL based on core e200z4 (with FPU)

    You can use the Pdroduct selector to compare them:

    SPC56 L Line MCUs - STMicroelectronics

    If it is not enough, you also have SPC57 and SPC58 families.

    If you want an FPU, do not use those based on powerpc e200z0 but use more (z2 z3 z4)

    Best regards