Closed ErikDeBruijn closed 13 years ago
I've implemented RR 5D checksum's before, if it turns out to be a problem on my mendel I can implement it np. I'm going to be trying out your branch tonight with the Arduino Mega firmware (it looks to be using an identical protocol to 5D) so any changes I make should be pushed to github by tomorrow.
GREAT! Would love to merge it in here!
perfect :)
Hey D1po1d! Your build queue seems like an awesome feature... I was thinking of building something like that, though I'm not much of a programmer yet. I'd love to collaborate on that!
Thanks Erik! Feel free to post issues to the build queue fork and fork it as you want (post a pull request when you get something awesome!). It's fairly beta in that its feature set is very small atm although it seems to work flawlessly on my makerbot which is very concerning (key word: seems). I'll post my feature requests to it so you can see what I'm thinking in terms of future development. My current project on that is setting up a way to serialize build queues so that we can have a single file contain an entire machine of printed parts and meta info on how strong to print each object, etc.
Didn't get the RepRap firmware working last night unfortunately so I'm looking at switching to rick pollack's stepper extruder mods for the makerbot firmware since that may be allot easier (+ makerbot firmware seems to be much more reliable).
Finally got around to this.. Haven't tested my latest commit yet (added retrys for bad checksums) but checksumming was working alright when i tested my mendel.
see: http://github.com/D1plo1d/ReplicatorG/commits/feature-5d-checksums
AWESOME :) Will check it out as soon as I have slightly more time (starting next week, after my graduation)
Closed, this is now implemented because of the awesome work done by D1pl01d!!
Currently, no checksums are used (the protocol just feeds the filtered GCode, leaving out M10[123] codes and such). I've printed quite a lot without checksumming having no problems (500+ parts so far), but in EMI noisy environments this could cause problems.
The 5D firmware allows for a one byte checksum (IIRC). We should send this and find a way to do re-transmits reliably.