• State
  • Locked Locked
  • Replies 2 replies
  • Subscribers 79 subscribers
  • Views 949 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

CCS/MSP430FR5994: MSP430FR LEA vs Cortex M4 DSP performance comparison

Aleksandar Atanasovski
Prodigy 140 points
Part Number: MSP430FR5994

Tool/software: Code Composer Studio

Hello everyone,

Is there any information how does the MSP430 FR LEA engine compares to the Cortex M4 DSP in terms of performance. I found on some Ti demo video that the LEA is approximately 4X times faster than Cortex M4 DSP. Is this measurement is performed with same CPU clock frequency on both devices (16MHz).  Cortex M4 is normaly able to run above 100 MHz so it should be faster than 16MHZ running LEA ?!

  • 0
    Guru 284380 points

    Increasing the CPU frequency does not necessarily increase other clocks for I/O or memory. However, many DSP algorithms are not primarily dependent on memory latency.

    Anyway, the point of that video is that both chips try to stay in a sleep mode most of the time, and that the more efficient LEA results in a lower power usage than the Cortex M4. If you care about latency more than about power, then choose a high-end CPU running at a high frequency.

  • 0
    Genius 5495 points

    Hi,

    The comparison is about clock cycle utilization. The LEA is limited with 16MHz however the efficiency of specialized engine in some cases is better even compared with DSPs. This is because LEA do preprogrammed operations on arrays/matrices with direct access to them rather than each element. But power consumption can be even hundreds times smaller that on high speed DSPs. The cycle counts for LEA is described in this document: www.ti.com/.../slaa698b.pdf

    Alexey

**Attention** This is a public forum