ATMEL Mega1284P evaluation board avalible

Just stumbled onto this and it looks very interesting. I was looking to start using the 1284p so if i could get it in an arduino compatible board that would be great. I will keep an eye on this thread and hope all the assembly goes well.

Thanks Subterranean, I am looking forward to attempt ass'y on Saturday.

..nice boards.. I would assembly the very first pcb manually, and in a step by step manner. In case of errors it could be difficult to debug when fully assembled. E.g. start with power, ftdi, 1284p and so on.. and measure the voltages, currents.. just a hint.. Pito

pito: ..nice boards.. I would assembly the very first pcb manually, and in a step by step manner. In case of errors it could be difficult to debug when fully assembled. E.g. start with power, ftdi, 1284p and so on.. and measure the voltages, currents.. just a hint.. Pito

I usually do this with any new board. It avoids any unnecessary hair pulling when something is not working!

Not sure quite how to do that with surface mount. Need to stencil on the solder paste for the Rs, Cs, LEDs, USB connector, SD connector, and surface mount ICs, especially the FTDI chip.

:) need a flux only.. there is nothing on the board one can not solder manually.. just put a lot of flux (i.e. amtech 223), place the ftdi or other chips/r/c and solder the pins.. it is easy.. I am soldering 0.5mm 64pin chips - no issues by today.. P. 8)

You must have much younger eyes than I do :D I stopped to get some +3.0 reading glasses on the way home so I can see the little parts more clearly tomorrow.

Still trying to get my bootloading capability back, I seem to have lost the touch, see this thread, writing things as I go.,67129.msg494716.html#msg494716

I am using 3+ as well :astonished: purchased in a supermarket for this exact purpose.. Important is not to solder your nose in as well.. :stuck_out_tongue_closed_eyes:

First pair of "Bobuino"s are assembled & minimally programmed! Spent a couple hours with my good friend Skyjumper, my son, and wife, got a little assembly line kind of thing going, me putting down solder paste & handing out parts, son placing parts on 1st and wife on 2nd , Skyjumper guiding us along in our first ever surface mount assembly project and driving the oven, group effort to find & fix the solder shorts on the FTDI chips. Skyjumper got them bootloaded for me via ISP (I somehow managed to not bring a mini-USB cable out of all the toys we took down, only had regular USB and the large square kind between us). Time to figure out how to do that with my tools, and write some sketches to show everything works now ...

Skyjumper, thank you again for all the help, and your wife for lunch!, and letting us use your tools.

Nice progress. Good luck on the alfa software/firmware tests. Suspect there will be a need for a nice single zip file that includes all the nessesary file changes and additions and a file for installation instructions.


Thanks Lefty. That's what I'm thinking also.

Beautiful job on the boards! Way to go!


Thanks :)

Very impressive looking boards you got there!

Thanks Mowcius. Finally got enough software pieces in place late last night that I can now select the board, compile, and make Blink - tho not on the pin I expected. I think I am using the Sanguino-based pinout, I recreated it as pins_duino.1284.cxx (just added some comments, no actual changes). I'm not sure how to follow the assemblage path to see that it gets used. I have the L LED on the SCK line, yet I need to toggle D15 instead of D13 to make it turn on/off. So I'm thinking I don't understand the software setup fully yet. Going to play some more today, see what the other pin assignments ended up as.

Here's what I want it to be, not sure how to make it so.

Ok, here's what I've figured out by working 0-31 thru Blink: PB0-PB7 are working as D8-D15 >> wanted D4-D13 PD0-PD7 are working as D24-D31 >> wanted D0-D3, PD4, D8,D9, PD7 PC0-PC7 are working as D16-D23 >> hadn't assigned #s PA0-PA7 are working as D0-D7 >> wanted D14-D19

What file do I change so that a shield expecting D0-D19 gets D0-D19?

.. try pins_arduino.c .. the pins are referenced by their position (the order) within

digital_pin_to_port_PGM[] digital_pin_to_bit_mask_PGM[] digital_pin_to_timer_PGM[]


PS: in case you want to use sdfat lib you have to modify sd2PinMap.h (add the 1284p defs, SPI port defs, e.g. "BahBots" like)..

Ok, that's gonna take me a bit to comprehend. Seems to be a ton of stuff listed under

#if defined(__AVR_ATmega1280__) || defined(__AVR_ATmega2560__)

and if not it is mapped to the 168

// these arrays map port names (e.g. port B) to the
// appropriate addresses for various functions (e.g. reading
// and writing)
const uint16_t PROGMEM port_to_mode_PGM[] = {

So this would need to add something like this?

#if defined(__AVR_ATmega1284__)
and add similar stuff as the 1280/2560

and change the existing #else to this?

#if defined (__AVR_ATmega168__) || defined(__AVR_ATmega328__)

or leave it as is, with the #else etc. being after the last of the 1284 code?

.. forget 1280 and 2560 - those have nothing to do with 1284p in any respect, 1284p=644p=32 as the pins are concerned..


and when required modify pins_duino644.cxx == 1284p

See my last post on sdfat chages when runnig with 1284p/ P.

Crossroady, you're very welcome, and also welcome back any time. We can get a full fledged assembly line going and make the last 10 boards!

I sent you a spreadsheet that matches port designations to arduino "digital pins." You do not need to modify the avr-developer cores, they work out of the box, including the pins files. Because the DIP package has different pin out than the SMT package we used on my board, you can't use my spreadsheet to map the physical pin number directly to the digital pin. But you can use it to map port to digital pin number.

If you need me to, I'll add another column for DIP physical pin number. but I expect you are ontop of that.

The data iin that spreadsheet is all tested. I also used that to add 1284 support to the NewSoftSerial library.