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.

GPMC Differences between Pandaboard and TMS320DM8147

Other Parts Discussed in Thread: TMS320DM8147

Ultimately, we will use a TMS320DM8147 chip in our design. However, we will use a Pandaboard initially to perform some studies, including connecting the Pandaboard to an FPGA circuit using expansion connectors J3 and J6. I have a few questions:

1. is the Pandaboard faster or slower than a TMS320DM8147? I believe that the Pandaboard'sCPU has two ARM cores, whereas the TMS320DM8147 has one ARM core plus a DSP component that we will not use. Should we disable one of the Pandaboard cores in order to make an apples-to-apples comparison of the two systems?

2. Just to avoid other "gotchas", are there other differences between the two systems that we should take into account? The Pandaboard uses a multiplexed address/data bus on connectors J3 and J6, but I believe a non-multiplexed bus is available with the TMS320DM8147; is the non-multiplexed bus is faster?

3. With "long" wires connecting the J3/J6 expansion connectors to our FPGA, I'm concerned that the CPU bus might be ruined by reflections, etc. Of course we can add wait states for accesses to the FPGA, but perhaps the other peripherals on the Pandaboard that use the GPMC bus will no longer work. Any comments?

Thanks,

- David.

  • Hello David,

    I looked at the link. I am not able to relate those questions to design.

    For (2), yes in DM8148 device we support non-mux mode of operation in GPMC interface. Definitely non-mux mode throughput will be higher than mux mode.

    If you are looking for some other info, kindly provide more detail on forum questions.