Traumflug

Nice chunk of work! > Did you mean literally to put calls to dda_join_moves() into a loop equal to the number of moves in the queue? Yes. Reasons to leave...

> Could it be skipping/dropping steps? That's always worth a consideration, of course. There's a z_disable() in dda.c approx. line 725. I think you can safely remove this. Other than...

Not to turn down your great work, more to give you a bit an idea in which direction to change things when searching for new algorithms: ideally delta movements would...

Today on IRC, #reprap-admin: [00:06] <thejollygrimreap> #ifdef lcd [00:06] <thejollygrimreap> in mendel.c [00:08] <thejollygrimreap> LCD isn't defined anywhere as far as i can tell [00:08] <thejollygrimreap> yay!! [00:08] <thejollygrimreap> it...

Put months into getting an algorithm step-accurate and people will barely notice. Put an evening into whipping up some display code and people will get crazy. :-) Back to topic,...

> I am interested in the approach that dc42 and ambrop7 have taken, but I do not think it will ever be feasible in Teacup without 32 bit hardware. There's...

Not that I exactly agree with triffid, but let's decide that later.

> I think that triffid is correct if one is actually stepping through the lookahead queue. Assuming an acceleration of just 1000 mm/s2 and 64 segments of 1 mm in...

> I have added the ability to more easily switch between time based and distance based segmentation to facilitate testing. Excellent idea! > dc42 is segmenting to the 1us level,...

P.S.: it might even be possible to do the segmenting after running dda_create(). Then you already have all the ramping numbers and can pass them to all the segments as...