Jump to content

[Piccolo] Driver for Mac OS X?

Recommended Posts

I agree that we could be more hobbyist friendly by being supported in tools like Energia, but I respectfully disagree on a few of your points.

  1. We are working on cross platform support for our compilers.  We already have Windows and Linux support, so all we are missing is OSX support.
  2. C code that is compiled with a GCC compiler can easily be recompiled with a TI compiler.  C is C.  The only thing that changes between compilers are thinks like compiler pragmas.  These are typically very easy to swap over between compilers.  The one point I will concede here is that our memory width is 16 bits wide instead of 8 (i.e. one address location holds 16 bits of data and a "unsigned char" = 16 bits).  This can cause some issue when code is ported from a byte wide platform to C2000.  That said these issues can typically be worked around.
  3. I think Energia and the other platforms are successful because of how easy to use they are and their cross platform support.  Certainly GCC enables this in most of the solutions out there, but GCC is not a requirement.  A closed source compiler with redistribution rights and cross platform support could enable the same functionality in an Energia type environment.

Trust me, I of all people want this board to be successful.  I want to see the C2000 LaunchPad supported in Energia, and I want to see lots of people coming up with cool projects using the board.  I'm fighting to get us the tools we need to make this successful, but its going to take time.  C2000 has traditionally been a closed architecture, but "the times they are a changing" in TI.  I just ask that as we work to improve the ecosystem surrounding the tools that everyone be patient.


Happy Hacking!


Link to post
Share on other sites



Points are well-taken, but the fact of the matter is that TI is asking the embedded hobbyist community to throw out their existing (and working) tools for a closed compiler that runs on two platforms for a relatively obscure chip (even if it is cool hardware).  Even if you target three platforms (Windows / Mac / Linux), it's not truly cross-platform.  It only works on three targets in which TI has dictated.


And there really is nothing wrong with this.  You have to use the right tool for the job.  Many people who program embedded professionally on a daily basis have no problems using and paying for CCS / IAR / etc.  There are great benefits for many people to do this, but it's not for everyone.  I am not trying to drag along the old debate of free (speech) vs free (beer).  Hobbyists ultimately become professional engineers / programmers and they often opt for the architectures and tools that they know best.  I recently opened up my induction cooktop to repair it and wasn't surprised to find AVR.  Even forgetting about the fact that it is in Arduino, you can program it with anything.  I could program and flash it with a Raspberry Pi if I wanted to torture myself.  I don't even use my AVR boards, but there are good reasons why it is popular.  We're drowning in a sea of cheap or free EVBs.  It's good for the hobbyist and good for the manufacturers (if they get an audience), but you need to make your chips and tools stand out as being the most accessible.


I only suggest GCC as a method to improve the ability to adopt C2000 among hobbyists.  It certainly isn't a requirement, but would help.  TI seems to interpret "We want better cross-platform support" as "We want Code Composer Studio for Linux and Mac".  These are not one and the same.  My observations have been that many Linux users try CCS v5 for Linux and then never use it again.  I tried it several times over the past year or so and determined that I could do eveything that I wanted for MSP430 and Stellaris with open tools instead.  When I reimaged my OS this weekend, I didn't even bother installing CCS along with my usual toolchains.  You can't cater to everyone, but you are already catering to the professional engineer with CCS for Windows.  The hobbyist is entirely different.  I can't imagine how much easier it might be for TI to just spend some time putting effort into open tools that already exist rather than reinventing the wheel and trying to port the entire CCS suite of tools to three platforms.  It took a year just to get MSP430 compiler support in Linux after the CCS 5.1 release.  I'm pretty sure that it still doesn't have Launchpad / ez430 support debugging after a year and a half.  Oddly, C2000 is better than even the MSP430 in this case.


I could actually sum this up into a single sentence:  If C2000 was supported by GCC and had an open debugger, it would already be part of Energia and we wouldn't even be having this discussion.

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Create New...