STMicroelectronics has joined the ARM mbed project

@ Valkyrie-MT - I’ll respectfully disagree with you. My favorite NETMF board of all time had 512KB of flash and would still be plenty powerful for most things I build today.

https://www.ghielectronics.com/catalog/product/256

Then head over to Saelig and buy up a couple of Fez Ultimate kits while they’re cheep :slight_smile:

I think more flash is always better, but only as far as it is on-chip. At 512KB, you have to make some sacrifices…

Of course but my point was that a 512KB board isn’t DOA for NETMF. It is actually quite cable and we got by with it for several years. Since the Panda-II was retired, GHI has never really filled the gap of a $30 premium board that we enjoyed with the P-II.

A very valid point Ian, there is a gap on a good replacement for the USBizi chipset. There are NXP Cortex M3 chips that are compatible with the LPC2387 but limited to 512K.

The bigger premium offering are great but it will be good to have a simple and cheaper option for simple projects.

Absolutely agree with both of you. In my opinion, this is the market where NETMF and microcontrollers in general shine. Leave the big stuff to things that do it better.

I am curious on your thoughts here, speaking of panda.

So why cerbuino is not an option in your opinion? Is the the premium firmware or the missing large header that was replaced with gadgeteer sockets? Or $40 is to much and cerb40 won’t do the job?

My thoughts?

$40 is too much, in my opinion. Cerb40 is pretty much the ideal form factor, for me, but $30 is still steep for what is there. I prefer the original, without the power supply components, but would have liked it to include the RTC crystal.

On top of that, the premium firmware would be nice. I see no value whatsoever in the OSHW offerings that can’t be compiled with GCC, since the list of people with access to commercial compilers is going to be very short, given the cost involved. The downside is a lack of essential features, the upside is… well, there isn’t one, since it’s not like I can modify the firmware and build my own version. Maybe switching over to the 427/429 version would solve this problem, given that space won’t be nearly so tight.

Also allow me to add that Gadgeteer sockets, for me, are a big downside. I won’t pay the kind of money I would need to to get into the Gadgeteer game, and besides, I ENJOY the soldering and breadboarding. I really like to solder. Maybe I’m crazy.

@ godefroi - so to you personally, panda is not needed as cerb40 got you what you need for $25? But you wish it had the rtc Crystal?

No, I love my Panda II. I actually got involved RIGHT as the Panda II came out. I was probably one of the first to buy one.

If I could put the Panda II firmware on the Cerb40, I’d do it in a heartbeat (but let’s keep the ENC28J60 for networking, it’s cheaper and much more capable than the WizNet).

Panda II got the firmware right, Cerb40 got the form factor right. I would really like to see them for $20 instead of $30, too, but I’m a cheapskate :slight_smile:

Well, you’ve got plenty of options then! Discovery boards are cheap and plenty of space to prove your soldering skills there :slight_smile:

The idea of standartized sockets is frickin awesome, price is not. And price of the Gadgeteer cable is terrible!

However, guys, I think you are getting totally off-topic.

Speaking of topic. Nordic is in the mbed game as well with a chip that runs your code and BLE.

https://mbed.org/blog/entry/Nordic-Bluetooth-Smart-SoC-running-mbed/

Yeah saw that. Those chips are really cool, I actually own a nrf51422 devboard (which can be reprogrammed and become nrf51822). Hardware is nice, prices are nice. But development atmosphere is [em]terrible[/em]. Community is small and helpless, the only thing that works out-of-the-box is Keil IDE, and that is a total piece of sh*t…