“Asset Browser” is kind of a vague concept. There’s actually two kinds of asset browsers. The project asset browser, and the artist asset browser.
For managing the assets for a particular project, like, say an animated film, or a game, you want your assets linked in, and indeed changing one should result in all of them being altered in all the files in the project.
However, this would actually be actively bad behavior when you’re not dealing with assets that are common to a project. When you’re making something one-off, or at least are using assets that you expect will be baked into larger assets, what you want to do is directly copy them into your files so that you don’t have any dependencies on your particular library. Imagine you made a car, and you linked in some wheels that you have in your library because you make cars often. Your car would then be dependent on the wheels being in your library and you definitely don’t want that when you ship that off to wherever you’re supposed to ship your work off to.
From the scant documentation, it seems to me that the asset browser is currently intended to solve the latter use case rather than the former. I imagine tackling project asset management will come later.
Hi guys, I don’t know if this is the right place to ask this question, but since this is the theme.
because they removed the asset browser in the later version 2.92 Alpha, for example in the Beta and in the Candidate version it was removed. Why?
The magnitude of problems solved by geometry nodes are small compared to the magnitude of problems solved by the asset browser. I’m kind of annoyed they decided to prioritize geometry nodes over the asset browser. I don’t know anything about coding, but I imagine developing geometry nodes probably takes 100x the time it takes to develop the asset browser. I really don’t care that asset browser branch isn’t perfect. It’s totally functional & already usable. We NEED an asset browser so much. Why is it not in the Blender 2.9.2 candidate? I’m so disappointed.
P.S. Please don’t take this the wrong way. I love Blender & it’s my favorite software, but I value being honest about my thoughts. I probably could’ve sugar coated this more. Sorry.
The Asset Browser is largely handled by a different set of developers, so it is not getting robbed of a lot of resources with Everything Nodes also under way.
Do you think the BF hired 20 devs. with the intent to place all of them on the same project? That is not how software development works typically.
In addition, I notice the date of the Blender Today video, the Asset Browser has since been made a priority for the first quarter of 2021 and is available in the 2.93 dev. builds.
It wasn’t ready enough for a 2.92 debut, so it was bumped to 2.93.
From what I understand, Asset Manager is going to get development similar to Geometry Nodes starting in the next few weeks. So there should be a lot of rapid improvements week by week soon.
I am working on a long term game project, so potential instability is something I want to reduce to a minimum. Unfortunately, I’ll probably stick to official releases.
Personally, I haven’t had any problems using experimental versions, except more crashes than usual, so maybe it’s safe. But I don’t know enough about the bugs to warrant taking the chance.
I tend to use them interchangeably, and download experimental builds every week or so or when a new feature I want to try is added. Like you say, there isn’t usually any major problem other than slightly less stability, and I tend to switch back to the last release if I notice an experimental build is crashing more than usual. Never had any file corruption with Blender, but I would imagine it’s a possibility.
Just been messing with this the last hour or so. Great start. Going to be very useful. However, I’m wondering is it possible to add from any file to a directory? I’ve created some test directories and populated them with a few assets.
Example:
I have a master materials directory set up in prefs. I’m working across different projects doing lookdev. As I work I create materials that I want to save out for later to the master material library and just continue working.
From what I can see though, this isn’t possible as I would have to stop working>save the current file>open the master material file>append the material from the file I was working on>resave>reopen project I was working on>continue working… or I’d have to save a duplicated/stripped out version of the .blend file with those materials I want into the master materials directory…Is this the case?
I’m just trying to get my head around the principles of how it works. I’m familiar with the basics of Blender’s append/link system as I’m traditionally a 3dsMax user, so I’m assuming the Asset Browser is based on that? Sorry if I’m missing something obvious.
This mode is very useful but It lacks option to link without unnecessarily duplicating materials/shapes.
When you have a huge collection of props that are sharing the same materials, Inserting those on fresh Blend will quickly increase the size of .BLEND because of duplicated materials.
This system needs to have an option for Linking when object is drag&dropped.
System also needs to have a way to detect rigs and importing them via link&make local.
Edit:
It would be great to have an option to import multiple assets.
Drag & drop is not 100% working… sometimes It’s selecting more than one object which Is annoying.
Edit 2:
Asset Browser needs a way to properly refresh objects marked as asset.
I’ve noticed that some objects are facing wrong direction when they’re previewed in Asset Browser, despite of the adjustment I’ve made in .BLEND file.
The aim seems to still be Blender 3.0. They had a workshop last week about it. So it appears to still be receiving attention, despite slipping official releases.
Well, it’s very important new addition to Blender. The first iterations were quite bad. I see no point adding broken design to Blender. I’d rather wait untill it’s really ready for release and most bugs and design issues are ironed out.
I am happy to see the BF is making more use of their code blog, since all of the insight they give really helps to head off speculation and hype (which can end up causing them a lot of grief once the feature formally goes live).
It also seems like they are invested in making sure the first iteration of anything is not overly-complex and thereby running the risk of black-boxing things (where some users won’t have the flexibility to make it work for their needs). To include everything to start with also risks the possibility of having to rewrite it much sooner than planned.
