Hi All:
I know many of you who have G100's are waiting with baited breath as to status, so I thought
I better start doing progress updates. Just so you have an idea as to whats going on..
Almost everything is done for the preliminary release. Im just working on homing. Once thats ready,
I will put out a "G100-Testers Pack", hopefully, this week if I dont get swamped with everything else.
The Testers pack will enable you to upgrade your firmware, and run programs, homeing, limits, arc's, lines,
all should be functional. Runtime accuracy reports well by numbers, table testing is unknown.. Ill get Brian to rehook
his mill for further mill testing..
The new firmware gives lots of messages to indicate a failure, and why, and so far, I have not seen it fail to
reconnect after a failure, and all failures have been my fault and right now, only in homing. (Which is a bit complex,
so it may take a couple of tries to get that quite right for all.).
I do have extended error message capability if someone gets into huge trouble. Other than Steve and myself, only Illya
has tried it, and he connected fine, so Im very hopefull this will be much less painfull than any other G100 update.
Fair warning though, I havent yet repartitioned for the spindle axis, though I dont expect trouble there. It will be automatic, if the
C axis is disabled in MAch3, it will automatically become a spindle axis for the G100. Analogue output is already working
at all times as before. The spindle axis will give you a frequency output ( 0 - 4Mhz), not a pwm. Homing may be single axis at
a time at first, then switched to multiple axis at once, as the new firmware can seamlessly join all homing axis together
by specification. Spindle partition may, or may not, be in the preliminary release, but shoudl be pretty quick afterwards if it
isnt done this week.
Movement seems good, but Im limited to my small tester machine at the moment for movement tests, and since the G100 now uses
full Gcode commanded moves, on a move by move basis, it means that CV may have to be reinvented a bit in the plugin communications
to Mach3. We'll see how that goes. But in general, it seems very fast and efficient, no connection or message problems seen as yet,
and I suspect many of the past gremlins will not reappear, but again, expect new ones while this is all worked out.
I want to fast track this one, youve been waiting far too long for me to hold it for reams of testing, so the G100Pack will be a
separate download, till we get reports from enough that they prefer it, and that its working OK for you. We'll go from there.
Its capabilities are very high, it'll take awhile to find the right mix of them to use. This is very unlike the last one, much cleaner,
and its library is very complete in terms of what it takes care of, and what Mach3 takes care off.
So far, Im very very impressed. I hope in the end, you will be as well. But as I said, dont expect perfection immediatley, you know
how this goes, space shuttles dont fly without lots of prelaunch glitches.. some easy, some hard. I just dont want to raise expectations,
youve all been there, so I think you understand. Steves done an enormous job, you have to see his code to understand what I mean,
but its a major library rewrite, nothign simple and not just a "bug fix", and now that he's done is job, I have to complete doing mine.
I may do a video before release just so we're all on the same page, but it may not be required, there really isnt much to do on your
end, if a G100 exists, it is found, and attached to, doesnt matter what the IP is in it, it will warn if things are wrong, and allow you to
correct them for the most part. It wont just hang, it'll tell you WHY its hung, and allow a total reconnect. I havent seen this happen
except when I mess up in the code for a special op like feedhold, or home.
Im aiming for Thursday or Friday release of the preliminary G100Pack update. Firmware and Software.
It WILL be for the brave and the stupid. I know theres lots of those out there.. ... ... brave I mean .. :-)
Thanks,
Art