What would you do with your fork of the NETMF?

[godefroi]

@ Reinhard Ostermeier - Microsoft isn’t trying to promote anything with NETMF. It’s just a matter of supporting the current version whenever a new version of NETMF is released…

[Reinhard_Ostermeier]

@ godefroi - Promoting now VS Versions with NETMF was meant ironically (should have placed this one there

What I don’t understand is why they drop support for VS2010 in NETMF 4.3.
So far this only resulted in that it’s not used by one of the biggest board manufacturer.

[godefroi]

The guys at Microsoft had no reason to leave it in, so if there was any reason to take it out, they would have. 2012 vs 2013 might have been a simple change, but was 2010 that easy?

Remember, Microsoft made no money from it, sold no products related to it, so they had no reason to keep it compatible.

[ianlee74]

@KiwiSaner

I think if all energy was concentrated on making one single mainboard with very good documentation and examples and on which every single feature worked flawlessly, then it would be a huge boost to people’s confidence in using NETMF. I do not think it is so important that if works with the latest Visual Studio or other platforms… only that it works perfectly and the documentation is clear and concise.

Therefore I would vote for less diversification and more emphasis on perfection.

I agree as long as that one board is in the Cerberus price range. I can’t in good conscience go tell microcontroller noobs that they need to go buy a board that costs 3x more than an Arduino just because the software is friendlier.

[Cuno]

@ianlee74

@KiwiSaner

I think if all energy was concentrated on making one single mainboard with very good documentation and examples and on which every single feature worked flawlessly, then it would be a huge boost to people’s confidence in using NETMF. I do not think it is so important that if works with the latest Visual Studio or other platforms… only that it works perfectly and the documentation is clear and concise.

Therefore I would vote for less diversification and more emphasis on perfection.

I agree as long as that one board is in the Cerberus price range. I can’t in good conscience go tell microcontroller noobs that they need to go buy a board that costs 3x more than an Arduino just because the software is friendlier.

Here´s the rub, isn´t it? Quality requires a lot of focus (hmm, and how exactly would we agree on a truly focused, i.e. minimal, feature set for such a board, which just might appear quite boring?) and yet a large amount of engineering time (for work on a lot of non-sexy stuff).

Unless GHI, or SecretLabs, or whoever else can increase their sales to hundreds of thousands of boards per year, the economy of scale will work against them, compared to Arduino (with cheaper components) and Pi (with even higher volumes).

Subsidizing a technology is not very attractive for a small company like GHI or SecretLabs, so the best bet might be the ST Discovery boards. ST Microelectronics has an incentive to subsidize these boards. But they have little incentive to use Gadgeteer sockets, so this addresses only plain-vanilla NETMF.

Of course this all applies to the maker market, where every penny counts. For businesses, even small ones, an evaluation board for a few hundred bucks should be ok - everything else would be penny-wise and pound-foolish.

[taylorza]

@Cuno

so the best bet might be the ST Discovery boards. ST Microelectronics has an incentive to subsidize these boards. But they have little incentive to use Gadgeteer sockets, so this addresses only plain-vanilla NETMF

I have often thought of this, what would the cost be of creating an extension board that provides Gadgeteer sockets? I like my STM boards very much, and would love Gadgeteer sockets even for my native work.

[godefroi]

I’m the exact opposite. I think that the Gadgeteer stuff has only consumed resources that could have been used instead to improve the NETMF, and for little gain. The world already had lots of modules, widely used and cheaply available, and what the world didn’t need (this is all my opinion) was a whole new module standard.

[Brett]

The Gadgeteer ecosystem was created for a different target audience to you and I. But I love the simplicity of a standard connection framework that it gives.

[Gus_Issa]

@ godefroi - you are right for people who understand hardware. It is made for software developers or hardware developers who need to get things working quickly nor cheaply, that is in the hobbyists side.

Of course, you or anyone can just use plain nermf if that is what they prefer. It is about giving options. Thousands of modules sold says there are developers who find it beneficial many of the orders are from customers who come back to get more modules for their toolbox.

One of the users is us. We use gadgeteer extensively to prototype custom projects before wev design the final board. Proof of concept using gadgeteer is very cheap, professionally speaking.

[taylorza]

As a software guy just wanting to play, the Gadgeteer standard removes a level of complexity that I appreciate, but I speak from very limited experience in that my introduction to the MCU world was Gadgeteer so I am admittedly biased.

[KiwiSaner]

@godefroi

The world already had lots of modules, widely used and cheaply available, and what the world didn’t need (this is all my opinion) was a whole new module standard.

Modules (hardware) is one thing. But proper drivers and documentation to accompany them is something entirely different. What Gadgeteer did very well was to provide really easy and standardized way to interface to the modules.

I’ve been using PIC microcontrollers for a long time before I fell in love with Gadgeteer. One of the main attractions was all the modules available and how easy it was for me to quickly make a prototype of almost anything conceivable.

[Simon_from_Vilnius]

@taylorza

@Cuno

so the best bet might be the ST Discovery boards. ST Microelectronics has an incentive to subsidize these boards. But they have little incentive to use Gadgeteer sockets, so this addresses only plain-vanilla NETMF

I have often thought of this, what would the cost be of creating an extension board that provides Gadgeteer sockets? I like my STM boards very much, and would love Gadgeteer sockets even for my native work.

I have also suggested this a few weeks ago, given the availability of Discovery boards, it could be the easiest way to promote Gadgeteer. Shield is easy. What may not be easy is changing the quartz resonator…

[Reinhard_Ostermeier]

I’m currently designing a room temperature and humidity device. The device must fit into a Standard German sub surface box (diameter < 55 mm). The N18 + temp sensor + 3 buttons must fit in the upper part (70x55x13mm).
Because of size and price reasons I use a MedusaMini as CPU, but as Modules I use Standard Gadgeteer modules (see picture). Initially I wanted to make a single custom board (even without display), but I always feared the board design (I’m a SW guy).
But now the electrical part is quite easy.
For the SW part: Iwould love to see a NETMF board like the MedusaMini (mainly size) for max 20$, then I would have used it here.

About prices: The components costs me about 72$ + 10$ for buttons, … per device plus 1x USB client SP (10$) for programming.
The work here is the fun part for me, so I don’t count it in.

An original Temperatur module from Bush Jäger with display (without humidity sensor, not custom programmable) costs nearly 300$, and they sell quite a lot of these.

What I want to say. The Gadgeteer concept really has it’s place, even for final ‘products’. At work I have professional people at hand to which I just say, I need this or that, and they build a prototype within a week (design / order borads, testing). At home I don’t have this service, so Gadgateer is really welcome.

[Brett]

@Simon from Vilnius

I have also suggested this a few weeks ago, given the availability of Discovery boards, it could be the easiest way to promote Gadgeteer. Shield is easy. What may not be easy is changing the quartz resonator…

We could just change the firmware instead… (but this is totally off thread If anyone has the STM32F429 Discovery yet, we should work on that as a target platform. Mine is held up somewhere in Element14’s ordering system and probably won’t arrive until March

[Justin]

@Brett

If anyone has the STM32F429 Discovery yet…

[Brett]

Module Maniac, what are you waiting for !!!

[Justin]

For Oberon to release the firmware ;D

[Cuno]

@Gus

It is about giving options.

See how difficult it is to sell “focus” Options are much sexier

[njbuch]

Whatever GHI, Oberon and Microsoft thinks about this: You are messing things up, by not collaborating on it!

By joining forces we would have an even better firmware for our cool projects - and the technology is ahead of netmf in the current situation. It should be the other way around.

Maybe you are afraid of each other, or are not mature enough to see the benefits.

[mhectorgato]

@godefroi

@ taylorza - they call that AOT compilation - Ahead Of Time. The reason Microsoft gave for not implementing it was that native code is bigger than IL. I bet that’s not really an issue nowadays with our 2mb-onboard-flash chips.

That single feature would single-handedly remove the biggest disadvantage that NETMF has- the interpreted code performance penalty. Add the ability to handle hardware interrupts with managed code, and you’re set. It also eliminates the need for RLP.

I also think WPF could go- if you want complex graphics, there are better suited boards for that (and much cheaper as well!). One can build the firmware without WPF, though… Cerberus doesn’t have it, for example.

What impact would AOT have on breakpoints in VS? I’m guessing it could still be accomplished, but I don’t know much about what happens under the CLR covers.

Granted I’m don’t develop products with a commercial mindset — but I don’t want to go back to debug.print based debugging.