On Wednesday 15 October 2014 12:00:38 Sebastian Kuzminsky did opine And Gene did reply: > On 10/15/14 4:07 AM, Frank Tkalcevic wrote: > > I was so annoyed that the estimates for my 3d printing jobs were so > > far off (estimated 2 hours, took 8) that I modified the axis code. > > It looks at the Velocity and acceleration of each axis and tries to > > calculate a more accurate time. It only looks at moves - no G64 or > > probing. It isn't perfect, but much closer. > > > > The attached file is the changes - a new gcode_properties function > > and some additional routines. > > > > My version of linuxcnc is pretty old, so a diff or patch wouldn't be > > useful. > > That change looks good to me, thanks. > > I pasted all the parts of your file into appropriate-seeming places in > axis.py, and it runs, so that's good. But i think something's off with > the estimate still. > > I did some testing with the sim/axis/axis config. > > Your code's estimate for the splash screen gcode is 2:09, which is > within a second of the measured runtime. The current version of Axis > in the master branch estimates "2.1 minutes", so they agree on this > one. > > But the estimate for 3D_Chips.ngc is way off. The current master > version of Axis estimates 0:46, your code estimates 3:44, and measured > runtime is 1:11. So I think there's some work to do still.
I haven't played with that one for a while, but that estimated time of 3.44 is still way faster that it will run cutting air on my machine. Estimated time is 13.3 minutes on my machine (without these patches, most recent 2.6.3 install), cutting air runtime was 14:30 at the speed limits of my toy just now. Fairly close IOW. I actually made that in a 2x6x6 block of pine, years ago, and would have done it in maple if it looked good, but because of the limited length of the flutes of my mill, had to do 3 passes to get to full depth. I had bookends in mind, but in pine it was at best "so ugly its pretty". :) My point is that the simulator, a 2.7.pre build, does run faster here, noticeably faster. > I pushed your updated version to git.linuxcnc.org as a branch called > 'seb/master/ftkalcevic-axis-estimate', for wider testing and in case > you want to refine the code. Cheers, Gene Heskett -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) Genes Web page <http://geneslinuxbox.net:6309/gene> US V Castleman, SCOTUS, Mar 2014 is grounds for Impeaching SCOTUS ------------------------------------------------------------------------------ Comprehensive Server Monitoring with Site24x7. Monitor 10 servers for $9/Month. Get alerted through email, SMS, voice calls or mobile push notifications. Take corrective actions from your mobile device. http://p.sf.net/sfu/Zoho _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
