Anyone Seen The Maple?

Well, consider using native USB capability to simultaneously continue to emulate a serial port, while also providing for other USB endpoints. Or DMA capture of anything, or timer support beyond PWM, or sleep modes, or debug features...

Starbug ... Have you used your hammer on the Maple yet?
Looks like the MIT boys are solving your problems? :wink:

Concurring in part and dissenting in part... :wink:

I think the current object implementation for the serial port might actually make it easier to implement multiple USB endpoints. At least for those who stick to the simplest stream-like methods.

DMA would certainly be difficult/impossible to do in a way that preserved downward compatibility with code written to do the kind of port I/O, and even bit-banging, that's common now. But I think it could be done in a way that fits within the Arduino paradigm, so it would be workable and less "alien" than switching to a different "OS" model.

Debug features are pretty much doomed to being very CPU-dependent, with little chance of encapsulating them in a way that makes them consistent from target to target.

Otoh, features like pulseIn() and the Mstimer2 library should be portable to almost any CPU, and it seems feasible to add some more-sophisticated capabilities that would also be portable. When you add targets with unique timer features, you can use ifdefs to enable them much the way that the existing hardware serial library handles the multiple ports on a Mega.

Overall, I think the Arduino community would be better served by extending to platform to more-powerful CPUs, even at the expense of sacrificing easy access to some of their features, than by telling people "Once you exceed the limits of the ATMega family, you must switch to a very different environment". Certainly some people will want/need to, but most, especially artists and others who just want a flexible component to use in their "non-computer" projects, will be happier to have a range of choices that work "the Arduino Way".

I have yet to reach the endpoint of atmega capabilities.. by a long shot.
However, a lot stronger chips do sound interesting, due to the 2kb memory restraint on my atmega328 I won't even bother making a neural net driven system, but it could be interesting to implement such a feature in a stronger chip.

So please, do go on with the ARM stuff, it is interesting stuff..
I also agree that the arduino makes the hardware, software, and where those two collide, accessible for a lot of people.. it would be a good thing if it could be extended to more powerful chips.

it would be a good thing if it could be extended to more powerful chips.

Again I think there is a clearly more logical growth path ahead for the Arduino platform via the Atmel Xmega series of processors. The Arm path is not a good next step IMHO for the Arduino developers to take.

Lefty

Not to get off the main subject of the Leaf Labs ARM Maple but
I have yet to see any useful development of the Xmega
line of microcontrollers. Just maybe, the developers are going to
fight the same problems as the MIT boys did. Looks like
the MIT boys have won their battle.

Apple with their ARM Ipods, Iphones and IPads knows where the
future is ... The ARM processor.

Retrolefty you bought a Arduino Mega for $65, see what you
get for $15 less with the Maple. ::slight_smile: ::slight_smile: ::slight_smile:

Retrolefty you bought a Arduino Mega for $65, see what you
get for $15 less with the Maple.

Actually I bought a Seeeduino Mega for $50. :wink: :wink: :wink: It's a better board with more features and more I/O pins.

Don't get me wrong, I will be keeping a close eye on the Maple and when I read enough positive feedback from more users I will most likely give it a try.

Lefty

Those of you looking at Primer2... its better to ditch it... Support seems to be kind of non-existent. though it packs in a lot of things at the price point, trying to do things beyond RIDE 7 and Circle OS seem to have limited support. :frowning:

Support seems to be kind of non-existent

Primer and Primer2 are "traditional" vendor demonstration platforms; some chips in a neat setup that is supposed to attract the interest of "real engineers" who might actually use those chips, thousands at a time, in some consumer product. The idea that a demo platform might be useful on its own, as opposed to just being a sort of digital pheromone, has yet to really catch on with chip vendors. As you say, "support" as seen in the arduino community, is pretty non-existent. (oh, you can probably get support for a product design, and such support can range from incredibly valuable to useless, which is how chips live or die, but that's not really the same thing as we have here.)

This is mostly because such support is really difficult to do. It's a sort of chicken-and-egg problem: if you have enough people with varied levels of expertise and helpfulness, they end up supporting each other. But in order to GET enough people tor each that level, you need to have enough people providing "support" to start with. I don't know how many people that is, but it seems to be more than the average semiconductor company assigns to a particular chip. Look at how many near-equivalents there are for Arduino that haven't quite made it:
Atmel Butterfly, PIC USB Bitwhacker, Coldfire TowerSystem, STM32 Primer and Primer2, mBed...

Alert ... The Maple is now available ... again.
For the ARM trailblazers, on this board, here
is your opportunity. I, myself, will wait until all
the libraries are done and completed. :sunglasses: :sunglasses: :sunglasses:

Primer and Primer2 are "traditional" vendor demonstration platforms;

West,

you got it bang on the point... i think i got desperate to try out arm and made a momentary decision...should keep away from such products in the future...

should keep away from such products in the future...

I tiptoed into this hobby by finding active user forums that could give lots of opinions and their stories of success and failures with their projects and learnings. If I don't find an active and growing forum to support a given hardware/software platform, I would just mark it as not ready for prime time and move on. That's how I found the Arduino and happy I did it that way.

Lefty

i think i got desperate to try out arm and made a momentary decision...should keep away from such products in the future...

Don't feel bad pracas - I got burnt on the Xduino. From now on, I will let others trailblaze the ARM boards for me.

:frowning:

I also ended up buying the STM Discover. Yet to try it as it has no support for linux.

it as it has no support for linux.

I have to admit that multiple host operating systems was one of the first things that attracted me to Arduino. I mean, it's POSSIBLE to do PIC and AVR development on Mac/etc, given virtual box this and Wine that and Fink the third thing and fiddle with the source to the fourth thing till it compiles, and maybe MacPort the other thing. But it was annoyingly ... annoying.
Arduino was trivial; even pleasant.

I haven't quite figured out why the Arduino environment hasn't become the sort-of-linux for small microcontrollers. (linux is great for vendors of larger CPU. Do a little bit of work and you get to demo a whole operating system, without having to license or agree to much of anything.) I guess gcc is a pretty high bar, and arduino doesn't really show off an individual products capabilities that well. But still; the landscape includes many architectures that already have gcc support (ARM being the big example, but also PIC32 (MIPS), MSP430, Coldfire, and some of those Renesas H8 chips.) Maybe in another couple of years.

Do a little bit of work and you get to demo a whole operating system, without having to license or agree to much of anything.

True. The MS licenses are annoying. I think linux is the way for the future with people realizing that they can get for free almost all of what they are currently paying for. As for why arduino has not become the linux for small mcus, i think its more to do with the ego of developers who believe that something thats easy for the beginner is not good for real-time development.. i feel more that way after reading the arduino haters on this post..

People just dont realise that they are falling victim to corporate monopoly at times.

If anybody else fancies a look at the STM discover board, the compiler is now 32k limit not 16k, and the board is available for £5.25 in the UK from Farnell. A knee jerk reaction from ST prompted by the success of the Arduino maybe?

A knee jerk reaction from ST prompted by the success of the Arduino maybe?

Well ST seems to be eying the students / institutional market in a futuristic manner. But it would take a lot to repeat anything like the arduino... lets see how it comes out.

The Batman has capitulated and put an order in for the Leaflab's Maple.

I see that the German ARM trailblazer "Starbug" is giving the Leaflab Maple leaf blowers
some hard hitting questions on their forum. From what I can see, his hardware problem on the Maple
(faulty / intermittent pushbutton switches) is fixed.

Tell, you all, how it goes...
:wink: :wink: :wink:

Well Batman, I'm putting an order in a few days...I hope you have done some path breaking. keep us posted.