[sdiy] PIC 16F1788 CPU bug

[Pete Hartman]

I keep writing and deleting replies, seeing as how I don't have the
same standing as some of the other folks speaking here :-)

I think the value of the libre tool chain for AVRs and Arduino in
particular is really a small part the overall cost/benefit
consideration, even when we're talking about our types of
one-off/small run applications.

I've worked a fair bit with Arduino and just recently started working
on PICs (though I did do a PIXaxe project last year, not the same bag
at all).

AVR's costs come in the development platform itself.  While you *can*
run individual AVRs and some do, it's not the common pattern--most
things are geared toward Arduino.  At $10 - $65 a pop, depending on
which Arduino you prefer, that can be spendy, particularly if you're
looking to embed something.  Lots of folks have talked about Arduino
synth modules, but I only know of two that actually made them
available, for example.  I can do a bare Arduino-programmed AVR (and
have), but it still uses a bit more overhead--I haven't seen an option
to run one on an internal oscillator like I can a PIC, for example.

On the other side, I don't think the development tools for PIC are all
that expensive.  I have a recent project I'm using the free MPLAB X
and XC8 environment for, and that is plenty for my purposes.  I paid
more for my PICKit 3 than I did for the FTDI buddy I got when I was
first doing Arduino a couple years ago, but only about $15 more.  I
had a slightly steeper learning curve to get the oscillator and
interrupts working correctly, but I was still able to get a working
prototype after 2 or 3 days of part-time effort.  It *is* fair to say
that the PIC documentation is harder to wade through and find answers
though.

If I wanted to do really time-sensitive stuff and resort to assembler,
I think there are more examples and resources out there for PIC in
that case than for AVR as well, but that could just be from lack of
looking for those resources...

Thanks

Pete



On Fri, Nov 29, 2013 at 12:55 PM, Tom Wiltshire <tom at electricdruid.net> wrote:
>
> On 29 Nov 2013, at 16:10, Andre Majorel <aym-htnys at teaser.fr> wrote:
>
>> Seeing as most of us deal with small runs or prototypes, that
>> would make AVR the logical choice. And yet PIC projects seem to
>> outnumber AVR projects.
>>
>> Am I missing something ?
>
> No, I don't think so. You're maybe overestimating the amount of logic that goes into these decisions.
>
> Having spent several years playing with this stuff, I reckon I could now make a logical and informed choice about which is the better platform for a given task. However, at the point I was trying to decide which platform to opt for, I knew virtually nothing about it and it was mostly dumb luck. The careful choice of the 16F684 processor for all my early projects was entirely the result of the fact that that's what I had available with my first PICKit 1 programmer!
>
> We think that these choices are worked out, but mostly they're not. You only find out later if you made the right choice or not, and by then it's too late. Luckily, in the case of processors, I don't think it matters hugely, and certainly not at our level. You can make any of them do any of the stuff that we might want them to do, although some will do some bits more easily and others other bits more easily.
>
> T.
>
>
>
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy