Wrong preview

no.
So if your G54 offsets are negative the plot is wrong?
By the wat what is wrong with plot? tool offset on wrong side?

Here is detail:

This is just a minor problem, it is caused due to calculation errors in some part of the code, most probably some rounding errors or unprecise calculation. Sqr(2)*sqr(2) will not give 2 but something similar

Try to set your offsets to exact numbers, like 10 / 10 / 10
The grafical deviations should be gone. I know you can't do that in real work, but just test.

Norbert

Try to set your offsets to exact numbers, like 10 / 10 / 10

It works good, but It is not possible in real world, as you wrote.

Is possible use preview from Axis to Gmoccapy? Or use old gremlin? Or .....?

Has gremlin always had this problem?

gmoccapy and gscreen uses the same code, just look at linuxcnc-dev/lib/python/gladevcp/gremlin....

Axis does use a very similar code, but not the same, axis is tk the other gui are gtk or Qt
As already mentioned, if someone likes to look art it OK, I will not do so, because I will gon on with gtk3 / python3 as the actual used combi gtk2 / python2 will not be maintained any more and hopefully the actual gmoccapy users will stay with us for long time and if python2 is not any more used then willing to use the new gmoccapy witch will be called MINOS (Machine Interface NOrbert Schechner)

I am working at that GUI, but actually I have two othe projects to do, one is a DXF integration into LinuxCNC and the other is about new hardware.

IMHO gremlin preview is only a cosmetic error, not fun, but will not damage anyone.

Norbert

Thank you for your response and I apologize for bothering you with the design issue alone.

When do you expect MINOS to be released? Just indicative. Half a year, two years?
Will it also be applicable for LCNC 2.8, or only for LCNC 2.9?

Gmoccapy is great, but this bug doesn't make it possible to predict what the machine will do, especially when there are many tracks.

Is there anything I can do to help me? I realize that LCNC is developed by programmers in their spare time.

you are not bothering me
I just wanted to make clear, that I am not going to fix this.

By the way, I have never seen it as much as in your preview. If you like, I can llok over your config. In that case just post your config folder and the File you loaded. Please note also the offsets (difference between relative and absolut numbers from the DRO's, so I am able to reproduce that.

This way we can at least exclude, that it is related to the OS or window manager, by the way what Linux and window manager do you use?

Norbert

I hope to finish MINOS this year.
It will only be available with master, as it require python3

Norbert

For testing can be used normal simulation LCNC2.8 LIVE CD


G54 must be non integer:


bug:



Configuration files are not needed, but if necessary I attach:


*.ngc files for testing:



Same problem I have with desktop computer and with RPI4.

Can I ask someone to try it too?

Whether we know if there is a bug in my hardware or if it is the standard behavior of LCNC 2.8 gmoccapy.