• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 82 subscribers
  • Views 435 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

Vey slow USB connection to TMS320F28335

Paul Raine
Intellectual 580 points

Hi,

We have recently started using the TMS320F28335A, which is a great micro.

However, we seem to have a really, really, really slow USB/FTDI connection.

Every time we launch the debugger it takes around 1 minute 40 seconds to connect, delete flash, load the code into Flash, and execute. (For a simple sample application)...

Is this typical for the part? Or do we have a problem with our chip or IDE?

Is there any way to speed up the connection?

Thank you in advance.

Host: Windows 7 using CCSv4 IDE.

Target: TMS320F28335A on a USB-EMU R3 Docking station.

  • 0
    Genius 5885 points

    Hi Paul

    I'd say it is pretty typical. XDS100 class debuggers are inherently slow, so switching to XDS510 class debugger (it can be one of those targeted only for C2000 devices) are faster.

    If you are considering going this way I'd recommend you either Blackhawk C2000 or Signum JtagJet2000 (signum is also most flexible if you have other devices on JTAG chain). By my experience they have way better EMI immunity, so your power stage can generate a lot of EMI before you're gonna lose the connection. Sadly Spectrum Digital XDS510LC (C2000 targeted) is quite sensitive. Note that this is just my experience with these three debuggers. Going this way also means forking for a full license if you'd like to use CCS 5.1 or higer as TI changed licence types in CCS 5.1 (no more free license for code size limited for C2000)

    Beware though that flashing the part always takes the same amount of time regardless of the license, and can be quite long. So for development I try to keep my code in RAM, which loads way faster. You also save on erase write cycles which are quite low for this part.

    Regards, Mitja