Migrating from emWin I noticed that LVGL is missing a “Hardkey Simulation” feature where you can add a skin to the simulator and define physical buttons within that skin. The buttons can then be hooked into the application to emulate actual physical buttons. This is very helpful for simulating devices without a touch screen.
There is a similar feature in LVGL inspired how oscilloscopes’ UI work. See here: Input device interface — LVGL documentation
It’s very important for us to understand what features are to most important for the users in order to know what to focus on. So could you tell why have decided to use LVGL instead of emWin? 
I was actually using STemWin which is the ST licensed version of emWin. I’ve been pretty happy with emWin but looking at alternatives due to the following reasons.
- STemWin is provided as closed source binary.
- It is locked to STM32 microcontrollers.
- The version is pretty old and ST has no plans on releasing newer version.
I’m still evaluating LVGL and was attracted to it because it is open-source, has a large community and provided some features I was looking for such as double buffering, animations/transitions, small footprint…
I’m also looking into Microsoft’s GUIX. I couldn’t find any detailed comparisons online, and the evaluation is pretty time consuming. I would appreciate any pointers you might have to help with my decision.
Thank you for telling this It’s really unlikely that these will change in (ST)emWin.
I know about GUIX but haven’t tried it yet. To be honest it looks good, but I wonder why it didn’t get any traction once it’s developed by Microsoft. Anyway, I saw that Renesas is using it.
It has a UI editor, but finally LVGL also will have one end of this year. You can try the beta version until that if you are interested.