The Maple ARM board (http://leaflabs.com/devices/maple/) looks like it really packs some punch. I’m considering buying it, still trying to think of a specific application where it beats the arduino. But for the price, why not?!
Anyone tried it out yet?
It does look very nice and quite the step up in performance over the 8 bit AVR mega chips. However it’s forum is still pretty new and there does seem to be a lot of additions/corrections ongoing. Maybe not quite ready for inexperianced users? I think I will keep a look out to see if it grows to critical mass of users. Nice price for the performance increase. Lots of toys built in and 12 bit A/D is also nice.
I’m sure the jump to 32 bit processors for hobbyist is not long in the future, I hope the Arduino team keeps pace.
It seems to be just as fit for new users as the Arduino…given that its programming environment is the same, form factor is the same, and setup is similar. The major limit really seems to be their constraint to the arduino form factor…their upcoming “Maple Native” looks very interesting. I do look forward to the 32 bit arduino future…maybe with an atmel ARM?
I have been following this Maple ARM / Arduino form factor printed circuit board for one year.
They started with revison 1 on the Maple ($40) and now their latest release is revison 3 ($50)
It took them a long time to realize that increasing the speed to 72 MHZ causes a lot of noise on the board.
They finally switched from a 2 layer board to a 4 layer board to reduce the noise and crosstalk between I/O.
The lesson learned is that anyone can put a new CPU chip on a board but by increasing the CPU speed, then special design considerations will have to be made. You won’t find the noise problems until thoroughly testing the board with software. I have to hand it to the MIT present and past students, which created the Maple, on how thorough they were on testing and debugging their board. It took them over one year and three revisions!
Maybe a person needs a degree from MIT to build an ARM microcontroller board! ;D ;D ;D
Even though the Maple is Arduino compatible, the porting of the Arduino libraries will have to be completed before I make my purchase. (See below) Otherwise, it is a excellent high speed Arduino/ARM platform.
Arduino Library Porting Status
Wiring Not yet In progress <------------ >:(
Ethernet Not yet Planned
EEPROM No The Maple doesn’t have EEPROM; use flash instead. This library could be emulated?
Ethernet Not yet Planned
Firmata Not yet Planned
Matrix Not yet
SoftwareSerial Not yet Planned
Sprite Not yet
LiquidCrystal Not yet Planned
Stepper Not yet Planned
BTW … I spoke too soon - the Maple has been put on hold due to software concerns :’(
Thanks for the update. I sent them an email a few days back and yet to receive a reply Perhaps they are overloaded.
Any thoughts on the Primer2 from STM? I ended up ordering one.
This is definitely the way of the future. It seems like the gap between hobbyists and professionals is slowly closing. Soon every man in the street will be able to do what tradionally only an Electronic Engineer can do, but only the EE will know how it works under the hood. :-/
But somehow I think the Arduino type 8bit solutions will always have a place, because ARM Cortex is a waste of power for many of the simpler applications that the Arduino is aimed at.
However boards like the Maple will definitely enable the hobbyist to explore advanced projects. Which could be a good and a bad thing.
Yep i see one. It is just lying in front of me waiting for a hammer which will crash this fine little board.
But before starting my report i must really say that it was not a big bang with this board. After ordering, the board arrived here in Germany after 2 days. Wow what support i thought. I installed all the IDE and all the drivers. You have to install 2 drivers. One for the Dfu protocol and one for the serial communications. Worked good. I compiled the BLINK sketch. Okay. I tried to upload the prog. Bang. First trouble the uploader could not find the Maple. I reinstalled the serial driver and tried again. It worked. The highly complicated Blink programm was working. What a joy. I changed the parameter of the delay value and uploaded it several times to see if it works. Each time i tried to upload something i had to repeat it atleast 3 to 4 times before success. Okay i thought i had to live with that. So i tried the next sketch out of the sample directory “communication”. It was the ASCIITable sketch. Compiling was okay and in a good time. But then the upload. Horrible. I tried for at least 8 times to avoid the failure "MAPLE not found. Altough everything was there. Then during the next uploading the board stopped all work an since that time it cannot be detected as a USB Device fron Win XP. It is just lightning the bootloader LED and then you have a complete break. Reinstalling of the bootloader with orignal STM32 serial Flash Loader doesn’t show any reaction. Either the implemented software for the Dfu and serial part are that worse or the whole system is in a undeveloped state and really not in a distribution state. Maybe it is for Hackers who love hacking the system but not for me.
My wife has “verboten” me to crash the board. Se told me to be patient and give it a last try. And if that doesn’t work she told me to nail it to the same graveyard board where all the other ARM board are are buried and doing a wonderful job as Artwork. (Armite, USBizi Dev Sys, Fez Domino are already on the Plate and i suppose the Cortino and XDuino will follow as soon as i get hold of them. I is really a wonderful Artwork in my office.
A blowtorch works better than a nail. ;D ;D ;D
BTW … To save you some German Marks or Euros on the ARM junk out there here are some pointers:
Armite - Basic programming without floating point.
Fez Domino - Excellent board for SD, USB hosting but not for simple tasks. Be prepared to write drivers even
for simple tasks like turning on LEDS. Netframework was created by Microsoft. <— Look out.
Cortino - Nice looking board, for your wall, but be prepared to write low level C programs for it - that
is if you can purchase this board.
Xduino - Support is AWOL and you cannot build libraries for it.
Mbed - Good ARM stamp but you can only use their online compiler.
Maple - Work in progress - one year and counting. If the elite engineering college in this country
(MIT - Massachusetts Institute of Technology) cannot produce graduates who can make an ARM
board work after one year then who can? I would bet they will prevail over the other ARM junk in the
Could it be that a ARM based system will never be as good a begineers platform as the Arduino AVR platform? It seems that system software is the only thing holding it back, is the ARM tool chain just too complex for a simple IDE for typical Arduino users?
I think the Arduinos success has proved their is a pretty good market targeting begineers out here. Or is it still too small a market for a ARM board designer to target?
Olimex - The European board producer summed up the ARM with this statement:
Very fast: most ARM7 cores run at 60Mhz and ARM9 cores run at 150Mhz+ providing more power than old 386 Intel processors.
Low power: ARM7 cores need only approx 0.5-1mA per Mhz
Great range of peripherials: ADC, DAC, USB, SPI, UART, I2C, CAN, Ethernet, SDRAM.
Lot of internal Flash: 32KB-1MB, Lot of internal RAM: 4-256KB.
too complex to beginners, definitely this one should be not your ‘first steps in microcontrollers’
too complex to beginners, definitely this one should be not your ‘first steps in microcontrollers’
Sounds like a good challenge and opportunity for someone wishing to become rich and famous. Surely the Arduino success showed that with a little abstraction magic (pin numbering) and behind the back magic (auto function prototyping, etc) there is a pot of gold awaiting someone.
If it was so simple - why is there all the ARM junk out there?
When you increase the microcontroller speed from 16 MHZ to 72 MHZ strange things happens. You definately need
a hardware debugger connected to the ARM to troubleshoot and to sort the problems out.
Software, which is simple to use, is the key. The problem is that the ARM processor is so complex that even
the MIT boys are having fits.
Apple, which uses ARMS in their ipods, iphones, ipads seem to have very little problems … hmmm
Their boys are from Stanford?
The MIT folks are having problems because their design doesn’t include any Lego :
Okay, now to be serious again have a look at the official home page for the AUG AMI DevKit for the .NET Micro Framework from an Austrian company.
A friend of mine is working with this system and is fully satisfied regarding the hardware, software and the support. The price hurts a bit. It is about 370 ? excl. Vat and shipping. It is a professional system which is also suited for hobby. If you want to learn something you have to pay for it. But this is what you get for your money:
the AMI DEV board (MH080407 Rev. 2)
a 3.4" AMOLED display in a plastic housing
a resistive touch screen glued on top of the display
a 1 GB Kingston USB-stick containing full documentation in PDF format and install-files for SDK, samples and more
a 5 V regulated power supply plug (110…240 V AC, 50/60 Hz)
USB-flash adapter cable for SAM-BA
serial cable (RS-232 PC pin assignment - to connect to a PC)
DBGU serial cable (RS-232 device pin assignment - directly connect to a PC)
standard USB A-B cable
standard Ethernet patch cable
“Getting Started” document in printed form
I know you’ll say oh “Microsoft” again, but to be honest the system works like a charm and never had a hanger up to now.
Have a look and come back for dissussion.
Looks good but at 369 Euro + Vat + shipping is nosebleed territory on the forum.
Some users were complaining about $65 USD for an Arduino Mega.
But you get what you pay for.
It is a professional system which is also suited for hobby.
Only if your “hobby” is “defeating alarm systems for bank robberies and jewel heists”.
Andy’s comment was too mild by half: that price is more like aneurysm territory for most hobby purposes. It would be different if there were a low-cost “runtime” module, and the US$500 or so were a one-time development system cost.
Something like the Arduino makes far more sense for hobby use: it has a really low entry cost, and excellent potential (so far) scalability. It’s been scaled down as far as ATTiny chips, and could easily be scaled up to chips like ARMs and Coldfires for applications that need more horsepower. That provides a wide variety of platforms that could affordably be permanently embedded in hobby projects. And don’t have multiple steep learning curves.
And also aren’t at risk for viruses and other malware.
I never said anything against the Arduino. I am using a lot of them for all kind of projects. We were talking about ARMs. I tried a lot of boards and was’nt satisfied. But i am still looking for new boards to test them. If people cannot afford to buy such expensive bords like the AUG AMI they should leave it and stay with there own small systems. Why not. I only wanted to present a board which is working well and can be used for evaluating small boards for hobby purpose. As already mentioned a friend of mine is working with this board to create a cheap Arduino like board which later can be programmed via IDE or MF4 just like the Arduino or XDuino. The status of this project is at 80% at the moment and only God knows when he will finish it. But it is really working a bit.
friend of mine is working with this board to create a cheap Arduino like board which later can be programmed via IDE or MF4 just like the Arduino or XDuino. The status of this project is at 80% at the moment and only God knows when he will finish it. But it is really working a bit.
More info, link? blog?
My objection to the AUG board is mostly based on the fact that it runs Windoze, which bloats the platform and has a big learning curve for people who aren’t already PC geeks. For the price, it arguably makes more sense to buy something like a netbook or an ARM-based PDA that can run Linux.
But it makes still more sense to get an Arduino(-like) environment running on a more-powerful board: it’s friendlier, more efficient, and cheaper.
For people who can’t wait for that to happen, I’d recommend looking into boards like the Netburners, which are well-supported with documentation and examples, and run a variant of uCOS that has an easier learning curve than Windoze for non-geeks.