This was another problem with the old rendering. Now it is consistent: if you want to know if solder mask is exposed or not, check the solder mask layer.
In general our goal is for KiCad to have as few settings as possible. Adding a user-visible setting in the GUI is the last resort, when it is clear that we can’t find a standard behavior that works for most users. The advanced config setting Seth pointed to is not a “KiCad 8 mode” but rather a tuning parameter for the new mode, and its behavior may change over time.
Ultimately, any change to “how things were” is going to be disruptive to some people, we know that. These changes (hopefully obviously) weren’t made lightly or “just because”, but rather because we found issues with the old way that, after a lot of discussion, we decided the best way to fix was to make some changes. We try to focus on functional problems (such as the gold rings disappearing in some situations, which Seth pushed a fix for yesterday) rather than nostalgia. So, we will first have to be sure that there is no way to both fix the problems with the old system and also address functional problems with the new system (in other words, be sure that a system with no options can’t work for most people) before we would consider adding user-facing options for something like this.
Whilst I haven’t installed v9 yet so can’t comment specifically on this issue, I can comment generally though.
I have been somewhat critical of Kicad’s GUI and to a lesser degree to some of its visual in the past, and I do think the argument made above (by the developers I think) is a valid one - made for the right reasons! However, I think the way it was done was somewhat ‘harsh’ and the comment by @crafyjon inadvertantly comes across ever so slightly disengenuous.
Let me explain. Kicad has used and beloved by many many different users covering a staggering userbase over a timeframe covering many tears. Paid software just can’t command the level of affection or loyalty that Kicad does - ESPECIALLY, when users have had to battle deficiencies or quirks in past versions. This resulted in intimate knowledge and familiarity with the software in order to be productive. Reading the comments above it strikes me that the developers have perhaps p***ed people off by their ‘heavy handed’ (sorry can’t think of a better phrase) change to a portion of the software interface people have become used to.
I think, with hindsight and with any future changes that directly affect what users see and stare at for hours, there ought to be an option to revert back to the previous version - for the sake of a check box in a settings menu or toolbar if we ever get customisable menus.
May I also use this time to thank the developers for all their hardwork producing v9, and this is in no way meant as criticism just a suggestion moving forward.
Thanks guys
While I can understand why this idea sounds appealing, I want to also be frank and say that this is unlikely to happen. When we make changes like this, it is usually because something fundamentally has to change, which almost always means that it would not be practical to just add a “go back to the old way” checkbox, from a technical point of view. In order to spend our development time most wisely, we generally limit the amount of maintenance burden we carry as much as possible, and one way we do that is by not trying to support both a “new way” and “old way” of doing things, most of the time.
This is why Seth and I have been talking about changes to the new way, not a “going back to the old way” option. In general, having “one way” is the best way for the KiCad team, and so it’s what we are going to be doing most of the time.
If you intend this message to be viewed in an entirely positive light, I suggest you double-check the definition of the word “disingenuous” that you use above.
Actually (and admittedly off-topic so merely an aside), I completely disagree with this. For myself, affection and loyalty has absolutely nothing to do with whether the software is free or paid for. It’s only about how good it is.
And having to battle deficiencies and quirks NEVER endears me to a software product! Absolutely the opposite, in fact. Any software that requires such hard-won knowledge, or requires me to battle deficiencies or quirks, is absolutely never going to gain my respect, loyalty or affection. Quite the opposite.
Why on earth would I entertain such positive feelings towards such badly written software? It doesn’t make sense. My respect, loyalty and affection is reserved for the very best software that I use: logical, elegant, intuitive, frictionless, transparent, discoverable, with just the right functionality.
Sorry, I’ll go slightly off topic also……
When Altium CircuitStudio came out in beta I was invited (as a member of the public) to be involved by Altium, at no cost, and I contributed a good bit to its dev. What made it attractive was the immediate response to issues and the community of beta users where it felt like “our” software talking and discussing issues and fixes with the devs. The mere chance of ending up with a great piece of software tailored in part to our own ways and methods at the end of it was a huge incentive.
I’d have done the same whether it was free or not. Ring a bell with KiCad?
To be honest, it’s a certain kind of person that enjoys and can tolerate this, not for everyone.
Of course, in the end Altium abandoned us and the product!
All of KiCads code commits are public and there are daily builds of “Nightlies” that allow anyone interested to try out new features before they get locked in at release.
This pad colour change cam about due to the need of many professional KiCad users for complex pad stacks, which are a critical step on the road to dealing with flexi-rigids and various 3D RF structures and also importing from various commercial PCB packages correctly…
Discussions were held in this forum about the thickness of the yellow ring following the development news. At the time of the discussions Devs were open to changes and the thickness was quickly increased. Final thickness was still to be determined. I guess it is still to be finalised after @Seth_h 's above comment.
I suspect this is something that KiCad users who have never done any serious professional (making a living from) software development will find hard to appreciate.
And when development time is given freely by volunteers it is even more important that their time is spend as wisely as possible to give the biggest ROI (Return on Investment) possible.
We should all, as users, constantly remind ourselves every time we use KiCad that it is largely due to these volunteers that we have such wonderful software to use.
Colors are important, cannot be taken lightly in GUIs. Color pallets are/should be adjustable.
But you knew this already.
Having more graphics layers is not a bad idea, KI has a few, more would be useful. Various manufacturing steps needs their specific info, as does engineering, design notes, and testing functions. These layers should be available selectively to specific needs. Adding it all into a single layer would be messy and confusing.
Being able to rename those layers would be helpful.
You can add a lot of user defined layers, if you want. These can be renamed however you want. They can also be whatever color you want. They can also be marked as front/back and paired.
