What are you referring to as 411? I’ve been trying to understand but I’m kind of missing it 
I have converted my script to output segments instead of curves and now it works flawlessly even though there are way more segments than there were curves before. And you still see artifacts of the technique which I don’t like.
The filters on the images are kinda heavy so causes blocky artifacts but it’s working okayish.
[attachment file=“118016”]
[attachment file=“118017”]
411 refers to my firmware version. Currently it is at 412, you can see it in the config file, when you connect a program to your board, or if you have an LCD plugged in it shows on the marlin screen.
I have your text file as a tester for the arcs but maybe we need a better test. Or not use arcs again for a while. There has been a lot of updates to Marlin (laser timing!) so Maybe I will make a 413 in the next few days. We need a stable version in a bad way that works with arcs.
Ah yes, like I thought, I couldn’t upload the latest version of Marlin due to an error in compiling so I wasn’t able to check the LCD or check it through the serial connection sadly. I’m using the raw version of Marlin by the way, it seems you are talking about the firmware specifically for the v1engineering machines I am completely DIY but still find this forum to be very good for the information that is supplied. Hope that’s ok.
No problem for other builds!
What date did you compile your firmware that the arcs work again?
Sadly no working arcs yet, only straight line segments. I tried different combinations of MM_per_arc_segment or something like that and jerk, both the old vs the new jerk method with calculated settings. But it seems my controller just gets choked up on G02 and G03. I was advised to try out Klipper but for Klipper you need a Pi.
You can go back to my 300 version of firmware, I am pretty sure they worked fine then.
Hey guys, I’m suffering from the same problem, right now running 414. Wanted to give the 300 firmware a try, but I am unable to find any other version than the 400s on GitHub. Can somebody help with an “old” version. Greetings
I too saw issues with 414 and arcs, and 415 cleared it up. The latest Rambo is 417 which Ryan has put effort specifically into good behavior on arcs. I’d recommend trying that first because it has fixes for other things and is less likely to produce confusion around different behavior on those other things.
Thanks for the heads up. I forgot to say, current setup is a RAMPS 1.4 board T8 16T LCD 32step DualEndstop, just for reference. I’ll give it a try, and report back.
There is work currently underway to automatically generate all the flavors but it’s not live yet.
In case you haven’t seen it, there’s another thread here that contains a lot of our experiments and thoughts around getting arcs working correctly.
So I tried porting the 417 to my RAMPS Board. Made a dry test run with the gcode stuttering before. Version 414 needs ~32 min, Version 417 is down to ~17 min for the same gcode. Now trying to run the code with motors etc. attached. Only having problems with the display (12864, not properly showing all the characters, looks like a timing issue, but not sure yet). And my x2 is not getting powered, I’m currently investigating if it’s the cable or a pin thing in marlin.
Be sure you don’t have the E0 fix and the EXTRUDERS=0 set. You need one or the other.
Seems something got loose while doing all the firmware testing 
Now x2 is working with EXTRUDERS=0 setting. Thanks for the hint. I still need to figure the display out. I had my MPCNC running today, with no issues so far. Everything seemed to be quite smooth, no stuttering, I could even ramp her up to 200% Fr, still no stuttering. Need to test some faster speeds as well, right now my movement is limited by the initial settings in the firmware.
To update on this topic, I have found a solution to G02/G03 arcs being slow in Marlin!! Read the last post in this topic:
It’s quite an easy fix and will probably be implemented in the next master version of Marlin (I hope).
Let me know if this worked for you
Thanks, I’m definitely interested to compare this to see why it has the effect it does.
Also, there was quite a bit of discussion and tweaks regarding arcs in this thread:
The Marlin editors have already picked up the change and are probably going to implement it in the next update. I am using my own converted version of Marlin since my machine is a DIY pen plotter and not an MPCNC. However, I understand you guys over here are using the same Marlin base. Let me know if you managed to fix it. I see you already lowered the values to 0.001 or something, that would come close to zero I guess. So indirectly you already found your own solution.
I saw this on the marlin twitter last night, it is merged. But now I do not have an assembled machine to do the arc tests…crap how am I going to tune/test it… I can’t update the firmware until I can check the settings.
So with this change it fixes the arcs settings and me and Jamie’s patches are not needed?
I have no idea what that zero is doing. It lets the last segment get calculated instead of forcing a value?