SPRADC4 june   2023 AM62A3 , AM62A3-Q1 , AM62A7 , AM62A7-Q1

 

  1.   1
  2.   Abstract
  3.   Trademarks
  4. 1Introduction
  5. 2AM62A Processor
  6. 3Vision Pre-processing Accelerator (VPAC)
    1. 3.1 Vision Imaging Sub-System (VISS)
    2. 3.2 Lens Distortion Correction (LDC) Block
    3. 3.3 Multi-Scalar (MSC) Block
  7. 4Deep Learning Acceleration
  8. 5Camera Mirror System Data Flow and Latency
  9. 6End-to-End Functional Safety
  10. 7Example Demonstration
    1. 7.1 Hardware Equipment
    2. 7.2 Software Components
    3. 7.3 Latency Measurement
    4. 7.4 Future Improvement on Latency
  11. 8Summary
  12. 9References

Latency Measurement

The glass-to-glass latency of this demonstration was measured by streaming a digital stopwatch and calculating the time difference between the actual stopwatch and the displayed stopwatch. Figure 7-3 shows an example of the latency measurement:

  • The actual real-time stopwatch was at 09:52:32 when the picture was taken.
  • The displayed stopwatch was at 09:52:26 when the picture was taken.
  • The displayed stopwatch lagged the actual stopwatch by 60 ms, which was considered the glass-to-glass latency.
GUID-20230607-SS0I-PHB2-14SD-DDKRBSX6ZMSR-low.jpg Figure 7-3 Glass-to-Glass Latency Measurement

This demonstration did not perform deep learning analytics and thus the latency did not include what is introduced by the deep learning processing. According to Table 5-1, the total latency is 57 ms without the deep learning block. Considering measurement errors, the measured latency of 60 ms is very close to the estimated value.