Shop OBEX P1 Docs P2 Docs Learn Events
Programming the Altera MAX 10M50 device with a P1V image — Parallax Forums

Programming the Altera MAX 10M50 device with a P1V image

ozpropdevozpropdev Posts: 2,792
edited 2015-08-31 12:05 in Propeller 1
Hi All
I managed to get P1V running on Altera's Max10 Dev kit over the weekend.
This board is fitted with the 10M50 device and fits 8 cogs nicely with 160K of hub ram.
Her's the relevant results of the 8 cog 160k hub test build.

Device	10M50DAF484C6GES
Total logic elements	17,491 / 49,760 ( 35 % )
Total combinational functions	15,784 / 49,760 ( 32 % )
Dedicated logic registers	5,542 / 49,760 ( 11 % )
Total memory bits	1,441,792 / 1,677,312 ( 86 % )

A lot of time was wasted trying to program the board via Quartus programmer.
Success was finally achieved by ignoring the Altera documentation completely.
Here are the steps to program the device from Quartus Programmer Version 15.0.2

Before compiling the verilog code make sure to set the configuration as follows.
In the "Devices" dialog press the "Device and Pin Options" button.

1_10m51.jpg

Compile your code. Quartus will generate a .sof and a .pof file.
Now start the programmer.

1. Press "Auto detect"
Quartus will give you 4 choices, select te first option (10M50DA) Ok

2_10m50.jpg

2. Highlight the first device and press "Change file"

3_10m50.jpg

3. You have two choices at this point.
A. Select the .sof file for ram based fpga image.
B. Select the .pof file for Flash programming of your image.

4. Check the program/configure checkbox.
5. Press "Start" to program the device.
6. Have fun! :)



771 x 234 - 56K
436 x 213 - 21K
1042 x 544 - 103K

Comments

  • Cluso99Cluso99 Posts: 18,069
    Congrats Brian. Now to check the pricing on the 10M50DA.
    BTW What speed are you clocking at?
  • Ray,
    P1V became erratic at 150 MHz so stepped back to 143.75 MHz (50MHz x 23 / 4)
    So far seems Ok, but further stability testing needed though. :)

  • Cluso99Cluso99 Posts: 18,069
    Pretty impressive.
    Shame the 10M50DA chips are all well over $100 ea from Mouser/DigiKey and nil stock :(
  • After a few more tests clock has been slowed to 125 MHz.
    I suspect that the bit shift, rotate instructions are the area causing the erratic behaviour.
    I base this "guess" around the fact that my tests seem to have trouble in the multiply/divide opeartions in spin.
    When time permits a test suite needs to be created to isolate the troublesome instruction.
    Perhaps a tweak here and there to the Verilog might allow the clock speeds to increase again.
    Still happy with 125 MHz though. :)
Sign In or Register to comment.