Yeah the N panel sucks. It’s the #1 reason why I don’t buy addons… that shove things into the N panel that I feel should go elsewhere. I’ve actually bought a few addons that I’ve never used just because they prove how nicely things can be placed somewhere other than the N panel.
There’s been a few single button addons in the N panel that I’ve paid someone to put in a more sensible place. I probably could have written the python for that myself if I wasn’t so lazy.
Addons are so important to the ecosystem I boggles my mind that they haven’t dedicated more thought and time to improving the N panel. The online addon repository is nowhere near as important as just fixing the N panel and educating addon developers about how they can put their tools in other places that make more sense.
Educating addon developers.
God knows why most of them silently default to having whatever UI their addon needs in the 3d view N-panel. It probably really just boils down to laziness and lack of thought or care by addon devs.
I mean, it’s not like the api, the core devs, the users or anyone whatsoever forced the decision to put an addon’s UI into the N-panel onto any addon developer.
My point here is: If the N-panel is crowded, it’s 100% the fault of addon dev’s ignorance/laziness. Blaming the core devs for it is absurd, and so is demanding a solution to this problem from them.
How much of a difference is there supposed to be?
Blender addons have to be GPL compliant after all. This implies their devs have to give you the source code, specifically of all the python stuff which directly interacts with Blender’s python api.
It follows you can look through this and you can (and are allowed to) make changes to that very python code which constitutes the (existing) addon’s UI.
But yes, I get it. You don’t want to deal with all this effort, which is sure a lot more than if you just could click-drag on or rightclick onto some addon-panel in a Blender-session to detatch that panel and have it hover.
I can understand that.
I still think most people overstrech the definition of ‘bat-guano-insane’ (whatever that means) the topic of this thread implies by far (and make an elephant out of any little itch).
There’s a pretty big difference: one requires you to learn python and the Blender API - or at least enough to hack at it to get the desired result, the other is a basic familiar feature that works out of the box in many software without having to write custom code. And, it would be a nice feature.
I say this as someone who has written a small addon and a few utility scripts when needed to add functionality I wanted to Blender: I love how friendly Blender is to manipulation via the API and that we have a large community skilled programmers, but not everyone who uses OSS knows how to code. Not everyone is comfortable with code. Some are downright intimidated by it. And that’s okay! They’re not lazy, they’re just normal artists. It’s understandable they might be frustrated being told the solution to their problem is to go learn to code and write the fix themselves.
I take it for granted that the ‘bat-guano-insane’ to most people means ‘whatever annoys you personally the most.’ For me it’s Multiresolution Apply Base: logically it should move the base model’s vertices to the corrrect position relative to where it needs to be when subdivided, instead it deforms the mesh.
That and the Asset Browser would be a hell of a lot more useful if it could save assets. As in push them from the current file into the asset browser. But that would be “asset management” says the blender devs, we don’t want that they say. It’s an intentional design decision that blender is one of the only 3D programs that refuses to let you save assets from your current scene to your asset library.
If i were a project manager I would want no less than 6 people working full time on asset management features for the indefinite future. So, I can understand why we won’t get it anytime soon. Most likely more than 1 or 2 studios will eventually release an addon they were using internally. But that’s probably 2 years away also.
I never find this “well, you can just change the code yourself, it’s open source after all” response particularly useful. Not everyone wants to learn programming in order to use a piece of software, and in that case that response has got to be really annoying. It annoys me, and I am a programmer – but that still doesn’t mean I want to deep dive into the Blender code or even the Python API at this time; that’s not why I am here. Yeah, I am tempted – though I don’t know Python at all, I wrote my first tiny add-on this week, and it was a bit of a struggle session (that’s ok, to be expected on a first foray), but ultimately satisfying [*]. I can easily imagine how overwhelming that must look to somebody who doesn’t have any of that skill set. Nothing wrong at all with not wanting to embark on a massive learning effort just because one would like to have a certain feature in Blender.
[*] And I admit, I put two buttons into the N-Panel. Just to figure out how to do it. No worries, I am not releasing this travesty into the world at large. I also agree that N-Panel crowding is a bad thing, and I wouldn’t want to contribute to it.
Yeah, I agree with this; I doubt the lack of floating windows gives anyone sleepless nights. But it’s clearly a thread for kvetching, and hey, most of us like somewhere to blow off steam when things annoy us. I try to get over how entitled some people sound; maybe they’re just having a bad day. :wry grin:
Yes. For now I am using Fweeb’s Export to Blend add-on for that, which has been a life saver during my recent reorganization. But contrary to what you said about the devs’ attitude towards that, I think I read somewhere that this is one of the things planned – but I’ll need to dig for it, and it’s bedtime (could also be that I am confused). What I want as much if not more is to be able to organize my assets into catalogs without having to constantly switch files.
Nothing in life is truly free, not even Blender. I’ve paid pythod coders from $5 to $50 just to move a few things out of the N-panel and into other areas.
That’s a difference between “being able to do it (rip some panel out to make it a floating panel) right in Blender’s UI” and “being able to do it by modifying the addon’s sourcecode”.
It’s not a difference between being able to do it for an addon you code yourself from scratch and one provided by a third party.
The latter was what @SoundDifferent refered to, to the best of my knowledge/understanding.
So I debated the claim modifying the code of an existing addon was all that different from making it work if you start from scratch.
Me neither, but @SoundDifferent very much claimed it not to be possible to do via the python api:
On another note I have to wonder how bi*ching about this or that in a BA-thread is particularly useful to a greater extent than telling people they can code it themselfes as it’s open source.
There are addons for that available now, it’s just one of those things I’d like to see as either a core feature or a built-in addon if we must make the node wrangler compromise. @piranha4D linked to a free one I hadn’t heard of yet.
Thanks! I hadn’t heard of that one. I know there are addons but it’s one of those things I’d like to see either built-in or in a built-in addon (like node wrangler)
If their attitude has changed, I’m glad to hear it. They definitely had been insisting that the Asset Browser was intentionally designed not to manage assets in their various write-ups of the asset browser and it’s intended function before release (I followed development closely at the time).
I wouldn’t be surprised if they’ve changed their mind because the vanilla asset browser is exceptionally unpleasant to work with and I think most users would like to be able to save their assets to the asset browser from the file they’re currently working in, regardless of Blender’s philosophy on the difference between asset browsing and asset management. For me, it’s been easier to just keep appending things like the old days before the asset browser.
I spent over an hour trying to make a simple chair with two parts (base and seat) have a rotatable seat as an asset. I tried every different combination possible, ranging from simple to ridiculous. Once I realized there was no way to do this and still have it be a single asset, I gave up on the asset browser altogether in complete disgust. I won’t be trying it again any time soon- it works well for poses, and that’s it. It’s entirely useless for anything beyond a pose library.
They still don’t have collections in the Asset Browser? If so, lame. I wouldn’t know since I gave up on it a while ago, I still hope to find a way to use it though.
I could find it useful to have even just materials in the asset browser. I’d prefer it be able to someway handle parented objects as assets too though.
They have collections, but there’s no way to make that kind of change within that collection. I tried everything- even adding a custom property to the collection and using that to drive an empty, which in turn drove the rotation. The custom property ceased to exist when adding the collection as an asset (even though I made it Library Overridable, etc etc).
It looks like dropping a collection asset from the asset browser drops it in the form of a collection instance. I tried doing the same with a single object, it works fine : its the same as the source, custom properties included. What’s weirder is that I would expect being able to create a library override on it to edit its contents but I can’t.
I’ve never had a custom property or driver survive linking or appending… But I haven’t tried since 2020.
A few years ago I had a working spotlight asset assembled via geometry nodes. The base rotated 360 and the neck swiveled 180 and it had an ies light.
_ fun fact, when you link a collection that contains a light and scale the instance up a little your light will disappear inside one of your meshes because I guess there is no concept of scaling light objects.
When I’d link the object and then library override the object itself only, I got access to all the controls on the geometry nodes modifier. It worked great.
However a while ago I tried to use it and when I did the library override it made every object used by the geometry nodes network appear in my scene.
Scene management when it comes to reusable things is a frustrating mystery to me. Nobody has any tutorials I’ve found showing how best to handle hiding the source objects used by collection instances and geometry nodes. Whenever I open an old file where I used such features I first have to struggle to understand where I put such things.
It might be a pipeline issue because the systems in place now (overrides, linking, assets) help make things straightforward. I’m linking my characters in my animation files and haven’t had an issue with properties disappearing. The materials not being overridable is a long standing known issue, but that’s about all I can complain. What is it exactly that doesn’t work for you ?
@piranha4D I cannot find that property… where is it ? what’s the ALO ? such mystery
I’m not getting that either, does it have to be linked in for this to happen ? I tried with the asset browser to drop an instance of a collection containing a cube and a point light, and scaling seems to work ok
