Which STM32H7 have CORDIC & double precision support ?
Hello ST Community,
I’m currently trying to select the right MCU/board from the STM32H7 series, but I’m a bit confused between the following options:
STM32H723
STM32H725
STM32H735
STM32H7A3VGT6
My requirements:
Need CORDIC support for fast math operations
Require high-precision calculations (double precision floating point)
Planning to implement a high-performance data processing pipeline (DPU-like workload)
Would prefer a board with good STM32CubeH7 example support
My confusion:
Do all the above MCUs support CORDIC, or only specific variants?
Is double precision (64-bit floating point) actually supported in hardware on these devices, or only single precision (FPU)?
Which of these would be the best fit for heavy math/DSP workloads?
Are there recommended Nucleo/Discovery/Eval boards for these MCUs that already have:
Working examples (DSP / math / AI)
Good documentation and community support
What I’m trying to avoid:
I don’t want to choose a device and later realize:
It doesn’t support CORDIC
Or double precision is too slow (software only)
Or there is limited example code available
If anyone has experience with these MCUs (especially for DSP / math-heavy applications), your suggestions would really help me choose the right board.
Thanks in advance!
