KILOCORE needs more investigating

propn00b

When I took a look at the project of the week, kilocore, I thought that this would deserve more investigating. I thought that it was a great idea. I am new to propeller and I dont have ANY budget so I can't do this myself.
:idea::idea::idea::idea:

obrienm

PropnOOb,
Hi, welcome to the propeller, thank you for the review. you are in for a new level of HSCMOS design.
Do you have a functional propeller system? like a
protoboard USB,
protoboard + propplug,
DIP propeller + propplug + eeprom,
propstick USB
spinstamp + propclip
schmartboard
backpack
professional development board
or
demoboard

To start any project involving multiple props you ideally need at least 2 propellers with their own programming interfaces with one as host and the other(s) as the grid - something like what is attached below. I find the best scenario is multiple DIP propellers with a propplug for the grid and a protobard usb for the host (this is cheaper than inadvertently frying a propstick usb or spinstamp by accident). Both usb connections can be run from the same pc using 2 IDE's. However, on a budget you can use a single programmer as long as you have 2 eeproms (1 for the grid and the other for the controller). In my own latest system, I have standardized on 2 ways of running the grid cogs. 1) propellent.exe the code directly from the pc host and 2) run a control system interpreter (pretty slow spin at this point) that executes a fixed # of commands directed from the host propeller 1 command at a time in SIMD fashion - so once the interpreter is stable I don't intend to update it very often. I currently was not able to run more than 12 propellers off a single eeprom so I distribute 1 eeprom per 8 propellers. In my so far 32 propeller breadboard prototype running at around 1A I have 4 eeproms that i update manually by F11ing the grid interpretter 1 eeprom at a time (avoiding having 5 usb programmers in total - not exactly fast development - this is why I hash out the code on a single 5 breadboard system of 9 propellers first).
Tell us more about your ideas for multichip parallelism as there are several designers here that are very interested in this work. If you trully need a starter 3 propeller system, let me know what you are missing from the set of (single breadboard, wires, usb cable, (host propeller/eeprom/usb-programmer combination - ideally protoboard usb + soldered headers), grid propellers (1 would do for now - but 2 or more would verify interchip communication), at least 1 595 and 165 for a base off chip i/o for the grid, a 374/574 or 595 for eventual fanout to a larger grid(try not to drive loads directly from the propeller), some LED's and resistors, or an LED grid with 595 row/column drivers-(no resistors required - see my "non-EE approved" 8x8 led display matrix driven by a cog on the controller propeller with 2 595''s/64 LED grid that depends on the inherent resistance of the HC595 as row and column driver - I have been lucky so far - it stays under 20mA) and power plug) and I will send you a portion of what is missing (are you in North America?).
For myself, I have made a some circuits involving arrays of LED's and the temperature/humidity sensor parallax sells over the past 3 years - as the strength of the propeller is in running multiple peripherals via shared HUB RAM between cogs. My focus these days is solely on SIMD arrays of multi-chip parallax propeller chips. The website will be updated prior to the upcoming contest this winter when the project is complete. Don't worry about HW or SW knowledge - I for example come from a pure SW background with some LS TTL experience. The forum will advise appropriately.

thank you
/michael