Fluid gets updated pretty frequently and I can’t test everything.
I am going to add a “tested good” version number to the Docs but I am looking for a few people that use their jackpot boards regularly and have no issue “flashing” them. Running $ss, keeping an eye on it running Typical milling, laser type stuff. I would like to have a few people try it out before I say it is good. I am typically only milling so that catches most things but maybe there are some edge cases I miss.
This is probably a once a week thing at most. So if you use your machine regularly, it could be pretty useful to have more eyes on it. No commitments required of course, just trying to figure out who likes to test bleeding edge and who rather wait for stable.
When a new version comes out load it up and give it a shot. Post your results below with something like “3.7.12, did a probe and some milling, also ran $ss, no issues” or “3.7.12, laser etching, no issues”. If you do find an issue put as many details as you can so someone can try to replicate it.
If there is a repeatable error one of us can go try and help the FluidNC Devs try to replicate it. That way we do not all go over there and spam them with the same issue and keep everyone over here updated.
Once I have my customs dealings and afterwards the built finished I’m also eager to participate, I run beta software on a lot of my machines and devices and usually like contributing within the communities that are fueling my hobbies
And I found that I personally learned the most about these things by regularly running into controllable problems on purpose and fixing them. Most of the time a downgrade/re-flash is easy so the risk can be mitigated. Keeps the brains fit and brings (me) joy
How would this work- you notify a group that a build is available to test, which is then the trigger for the beta team to flash it and do some test work?
This, of course, would mean I need to get off my butt and build a few more machines to test with.
Having only the JL1 laser at the moment running a jackpot.
I am thinking just put up a notification (probably in this thread), and when 2-3 of us use it we can call it good. Preferably all three not just milling. Then we call it good.
I don’t for see many crazy issues but I don’t like that some people are using the current releases day one without anyone testing them at all. There was one release that probing did not work right (if I remember right), it was quickly fixed.
I like that this thread is a thing. I’m not volunteering to beta test, but I am the kind of guy that would update to the current version once it’s okayed even if I’m not having any issues. I’ve updated a couple times now and it’s awesome how easy it is.
I am on 3.7.12 only because I updated trying to resolve some config changes I messed up. The web installer defaults to the latest version so I just went with that.
With that default, we should probably have a group that follows along with the releases, because users will stumble into problems as they’ll always be loading the latest.
For the people who aren’t interested in bleeding edge, having a “stable” version that only changes every 3-12 months would be ideal. It seems like there is a lot of churn at the moment and getting v.10 or v.12 is important, but once a set of features is reached and stable, I would strongly suggest freezing and making it very clear to stop at that version and then have stable updates less frequently.
Ubuntu has stable versions come out April of even years. The October and odd year releases work, but they try new features and experiment with UI changes in those non LTS releases. I usually don’t even install the stable releases for 18 months, just to let them get the really rare bugs worked out in the point releases. OSes are not the same as firmware. But it helps to have a point of reference that several months to a year is not crazy for a stable release.
If the stable version has solid wifi, memory and TMC comms, then maybe it isn’t a big deal to wait for the dongle or new laser features or something.
Absolutely. I am starting to understand enough to know what is important and what isn’t. At first I still think we are going to be a little more often but it sill quickly stabilize even more. One more major feature and I think we will be at a great place.
I was having some issues this weekend, so went to look at what updates were available and saw that there is a pre-release of 3.7.14-WebUI 3 test. I was curious, so after a little pain, I got it working. I have only homed my machine (successfully), so far, but will report back when I get more time on it. An issue I ran into is the WebUI needs to be uploaded separately. I think this usually installs automatically, but maybe they don’t want to have that happen with a new WebUI. It took me a little while to figure out why it was connecting, but I couldn’t see anything in the browser window.
Two things I noticed, but did not have time to look into. First, the old macros are still stored on internal memory, but they do not show up in the macros tab. I think I read somewhere that macros had been reworked, so I will have to figure that out. Second, there are not separate buttons for homing axis individually. There is a home all button, but if it homes z my Primo will crash, so I did not try it.
Both of these are configured in the settings on the “ESP3D Interface” tab, or whatever it’s called.
The v2 files are not compatible with v3.
The preferences.json, and all the macro configuration needs to be removed. You should only have the web file and your config.yaml on there to start.
The new interface has a lot of options so you have to dig into them a bit to get it set up how you want until someone shares a good “starting” preferences.json for how we tend to use it.
All of the panels, order of panels, etc. is all configurable.
You need a CSS file named in the format “theme-”. This file can be compressed or not, but best being compressed for Flash storage limitations.
Here is a light theme I made as an example. If you unzip this file and place the theme-FluidNC.gz file on your flash, there will be a theme selection available. theme-FluidNC.gz_01252024.zip (11.8 KB)
I have not tried to create a dark theme.
If you’d like to try your hand at creating a dark theme, then I posted this on the FluidNC Discord previously
