Linear advance

[mhackney]

Hey Peter, I've been doing a lot of research in this area. One quick comment, the only "feature" I'm aware of that was removed with the introduction of KISS Preload was the ability to have different retract and advance rates. This feature was originally implemented at my request several years ago. I was, like you, an early tester of Preload and noticed this was missing. I don't think it is a major issue but there are times I would still use it.

As for mixing Preload and PA or LA (depending on firmware) - I would expect all sorts of bad things to happen and so far that has been my observation. I am not willing or able to invest time and energy into researching new firmware like Klipper so I can't speak to it. For the last couple of years I'm all in on Duet RepRapFirmware for many reasons that I won't get into here. I do, about twice a year, look at the other major firmwares/controllers to see if any new advancements are worth investigating. Klipper is working to solve a problem I don't have with Duet and RRF and KISS Preload.

cheers,
Michael

[pjr]

Hi Michael and thanks for your comments/observations

As for mixing Preload and PA or LA (depending on firmware) - I would expect all sorts of bad things to happen and so far that has been my observation.

Where Linear Advance has been implemented in a limited manner (basically, improperly), I too would expect and have seen bad things happening. But, if you think about it, Preload G-code is still just a sequence of movement commands with extrusion. There is no reason why a properly implemented LA firmware would have any problems. The firmware is there to do as it is told; if it can't handle what it is told to do, then clearly it is faulty.

With regard to the omission of items when preload was introduced, it may be that only one basic primary setting was omitted, but there were also many underlying changes which screwed post-processors (Cubeitmod and my own, for example), and that is not ideal.

Peter