Asset Browser - why still so WIP?

I am sure this must have been said, but just in case it hasn’t,

JuuuUUUUuuuust imaaAAAAaaaagine just for a single moment in time…

Things might be easier to have a .blend act as if it brings everything together and have the assets saved out as different files with their related data blocks. For example

.blendMaterial
.blendCamera
.blendGeometry
.blendCurve
. … and so on.

So a .blend file would become a simple exposed folder structure at an OS level. Is there a reason why a .blend file can’t have everything stripped out except for one particular item such as only that geometry, curve, cameras, lights, or materials … containing that material and its related data blocks with just a file extension name changed? almost link everything is linked? Perhaps everything could be in its own 3D space as well?
A saved .blenderProj would put everything together or a Save Project as zip could be an option as well.

I am sure it’s more complicated than creating different file extension names for related objects and their data blocks.

JuuuuuuUUUUuuuust Imagine, you know, just for a moment assets would be a lot easier to pass around.

At this point I just wish for Blender to have python API for editor space addons, so someone can make a proper Asset Browser in form of an addon instead this one.

1 Like

Unfortunately its not Blender Dev’s Failure that the Asset Browser implementation was weak and didnt succeed upto the user expectations. Just amongst the Asset Browser there are so many other things as a backlog to be fixed/improved by the devs but the Fact is there are just not enough devs available/hired. As we all know the alredy handful of team is working tirelessley on multiple projects so it’s just not possible without hiring more devs in future to speed up the development. As a user, we shud Donate to the development Fund to help !

3 Likes

Jokingly or not, best not to propose violence as a viable option.

Keep it civilized!

9 Likes

Well, there are asset browser meetings every other week (see: 2022-06-02 Pipeline, Assets & I/O Meeting), so if any of you has any ideas for possible improvements, you can join the discussion. Ranting won’t solve anyone’s issue, feedback does.

7 Likes

unable to delete few assets together. So yea, «very WIP, much raw, so earli»

1 Like

I found it interesting that Juan Linietsky (one of the main original developers of the Godot Game Engine) wrote that Godot and Blender currently seem to have inverted problems.

Godot has many potential developers wanting to help but they don’t have enough budget to hire them. While Blender currently has a very solid stream of budget coming in but not enough people to hire for specific tasks. Wonder how correct that assumption was. ^^

Hm. I’m not too sure about this. There are Test builds for features that are still being worked on. You have to strike some kind of ballance and Blender decided that this is the one. Unity keep all their tools in a constant state of “this is beta don’t use for production or on your own risk” over many many years lately and I honestly find that a lot worse. I’d rather have a stable release with a handful of features and then iterate on it than either nothing or something that is on a constant state of instability or flux.
Ultimately it always helps a lot more to offer alternative design ideas and improvements or even actual programming help if possible. Saying what’s bad is easy. Making an actual proposal and waiting for others to shoot it down is hard.

3 Likes

The problem here is that the Asset Browser in it’s current state doesn’t offer even the most basic features, which would make it worth utilizing in any kind of a real project.
It is currently almost the same like working with libraries, just with drag & drop (unusable due it’s limits) and bugs.

1 Like

It can store assets, allowing you to drag and drop them, and allows you to create an asset directory for better navigation… that’s about it,

1 Like

It can store assets, but only Objects, Materials, Poses, Worlds, and Blender versions 3.1 and 3.2 respectively added support for Collections.

Other data blocks are planned, but some of which require to actually change the way Blender handles them. Such as brushes:

The way brushes are managed in Blender is in dire need of a complete overhaul. The plan is to tackle this is a separate project, so it is not something to expect in the Blender 3.0 release. Nevertheless the Asset Browser is designed to work as a basis for better brush management. Just like it acts as the basis for the new pose library system.

Source: Asset Browser Workshop Outcomes — Blender Developers Blog

  • Merging all brushes into a single tool with brush presets
    Any types of brushes would then be available as a list of brush presets in the brush settings.

Source: 2022-8-24 Sculpt/Texture/Paint Module Meeting - Blender Development / Meetings - Blender Developer Talk

Non-data-blocks are also planned, but don’t seem to have prompted more than broad, long-term thoughts so far:

Non-Blender assets, such as image or audio files, will likely be supported in a future version. For such files, asset metadata is then stored in XMP sidecar files, similar to what other software is also doing. Importers (USD, glTF, FBX, …) could add support for their file types as assets this way too. Furthermore, it should become possible to enrich an asset with a Python script, which can then provide code to be run when the asset is used.

Source: Asset Browser Introduction - Design limitations — Blender Manual

Usability changes are also planned. Such as:

Asset Pushing

Asset pushing is a way of getting assets into the asset library, where you are working on a file and want to copy the asset from that file into the library.

Cross Blend-File Editing

(…) Blender itself is not allowed to write to other blend-files than the currently open one. This rule helps to limit complexities; for example, it is hard to reliably implement an undo system when manipulating other files. The rule does get in the way of mass-updating assets when they are stored in various blend-files.

Since there is already tooling that can manipulate blend-files outside of Blender itself (see Blender Asset Tracer), it’s possible to also create an external tool for doing such edits across blend-files. Such a tool might even be implemented via Blender’s application templates system, or as an add-on; (…)

Milestone ?: Usability

Once the asset manager is working for the basics and the UI can do its job, we can start to make things nice. This includes better handling of thumbnails, re-configuring the ID browser to allow for asset browsing, a nice UI widget for managing tags and tag-based filtering.

A key part of making the UX nice, is to make it effortless to drag assets into the scene. When dragging in objects & meshes, they should be able to snap to surfaces. When dragging in materials, the underlying target objects should highlight on rollover. You should be able to add multiple assets at a time.

Milestone ?: Project Repositories

Up to this point the asset manager will handle only the use case of appending files from a local or online repository. The next step would be making it works also for projects.

I stop the listing there, but there are quite some things to read here and there. You can follow the current development state on the Asset Browser project’s tracker page.



While I do feel that the Asset Broswer was made publicly available too soon and its does seem to have lead in fair chunk of the users feeling mislead into using a tool that wasn’t ready for their expectations, and maybe even today it should be put aside under the developers extras, I can’t tell for sure that it isn’t usable. Factually, you can work with it, so surely there are people who already use it in productions just fine.

What is definitely sure about it, is that it is not finished but is still rolling up the hill. The BF is working on so many things at the same time, and the Asset Browser is such a big piece, it’s unrealistic to expect it to be done and polished this quick.

So, let’s arm ourselves with patience, wait and see, give feedback to devs, and hopefully we get a nice asset browser soon enough.

4 Likes

It would be nice to have an option to display the full names of the assets in the asset browser.
Sometimes I need to name them with more than 2 words and I like to have them in small size thumbnails.
Maybe we could split the text on multiples lines (3 maximum?) and give more spaces between the thumbnails to fit it, if the option is enabled.

Show full names for assets

7 Likes

At this point wouldn’t it be reasonable to deprecate the asset browser and remove it from Blender if there are no developers working on the basic UX? It’s very risky to keep it in Blender as non-beta feature. Every time I attempted to use it I just ended up with lots of wasted time.

1 Like

There is nothing I would want less…

For me personally, the asset browser has become indispensable; I use it to store the models that I use, which makes building scenes so much faster when every asset is a click away. I use it to store common or high level node groups, which is incredibly useful when working with shaders and geometry nodes.
And a bunch of add-ons I own are also integrated with it (Botaniq, Scatter, etc.), and the difference between using the asset browser vs the default addon interfaces is simply night and day.

I can’t express how much time I have lost to that append dialogue, and how much has been regained from replacing it with the asset browser…

It has sped up my workflow immensely, and while it does take a fair amount of work to set up and get used to, to remove it simply because of that would be ridiculous. I agree that there is still a lot of work to be done, but improvements will come, and being unhappy that it’s not perfect immediately won’t help anyone.

3 Likes

I think removing the asset browser due to non-developed features would be met with many angry users; despite the limitations, many people regularly use it in that limited way.

(I am not one of them. The “18 step workflow” and lack of parent/child support is among the many reasons the browser is a tool I’ve just removed from my layout.)

1 Like

Dude what is this mentality all about ? Today alone I read multiple comments of you constantly complaining how bad Blender is here and there. Then pay for other software or contribute constructive comments/ development .
The Asset Browser is far from perfect sure, but even in its current state it makes working in our studio so much easier already.

4 Likes

In short term maybe. There are more features than just asset browser that end up in really poor unfinished and barely usable state, such as mantaflow. In long term, if no more development resources are spent on it, it does more harm than good.

Please tell me, how are you utilizing the asset browser to save your time? I am genuinely curious since I just don’t see the amount of time it takes to properly create an asset which is reusable between blend files paying off by having it available in the asset browser.

If you are claiming that asset browser makes working in your studio so much easier, then there’s a significant chance that a non-asset browser solution would make it even easier.

I genuinely do not know WTF are you talking about here. You are saying you are reading my comments but it doesn’t appear to be the case. Just in this thread alone, I’ve wrote following.

How much more specific would I have to be for it to count as constructive? Also, if there was a way to pay someone for Blender patches which would be guaranteed to make it into the master so it doesn’t have to be maintained separately, I’d have done that in a blink of an eye.

almost all of the suggestions you made in your comment down below are already possible. Sure a new UI Popup could save in some cases a few clicks, but we spend many minutes to hours per asset to prepare the real asset data( apply transforms, create/ optimize materials/ mesh data, generate custom thumbnails) and maybe 30 seconds to setup the “metadata” you are complaning about like choose the right cataglog, blendfile, filename.
So I really dont get your point. Yes with a more refined workflow/ UI we could save maybe 20 seconds per Asset,but this is completely negligible compared to the time that must be invested to prepare the actual asset.

we invested a ton of time to build up a library of multiple thousand ready to use assets and as I said more than 95% of the time we spend on preparing the actual asset data. This was an investment, but we benefit every day from it and already reached the break-even.

We have a huge hierachical structure of thoudsands of indivdual blend files just containing one Asset, therefore we have no need to choose a target library, the blendfile already lives in the right domain. After the real time consuming process we just click-> mark as asset → choose the catalog → choose the thumbnail and save the file, all done in a few seconds.

In my opinion real improvemts could be done for example in Versioning, better Variant management, smarter Caching and so one. But this is already on the horizon : A bird’s-eye view on the asset browser system - YouTube

1 Like

This is exactly the problem. You are conflating asset creation and preparation with the asset storage. Those are two completely different things. Asset Browser is not a tool for asset creation or preparation, only for asset storage. What you are describing as a bulk of your workload is something that was already being done way before Asset Browser existed. People were preparing assets and organizing them into libraries of blend files. The whole thing Asset Browser was supposed to make simpler was the asset storage and reuse. And while it did marginally improve the asset reuse by creating a new editor with visual thumbnails, it did not improve the asset storage a single bit, despite asset storage being exactly what Blender lacked the most.

You are claiming the reason asset browser is good enough is because most of your asset related work you specifically do in your workflow is not related to asset storage or browsing. That’s hardly a valid argument.

Here’s what’s so frustrating about this:
Even before the Asset Browser was conceived, most people had rough idea of what the asset reuse workflow should look like, because there were already a great examples. Game engines for example. It was not some nebulous territory that had to be researched and explored. It was already quite clear and solidified, so many people had reasonable expectation of what asset management, storage and reuse workflow should look like. Yet, even after a year of being out of beta, the most basic requirements are still not fulfilled.

I’d compare it to having a big, flooded basement with couple of meters water, and you having to use an old bucket to scoop and throw the water out. You decide it’s not feasible, because it would take you weeks of manual labor to remove the water just with the bucket. You hire an engineer to solve this problem. You have at least rough idea what the right solution is… some sort of electric compressor pump with hose to pump the water out. After half a year, the engineer finally unveils his solution. It’s a bucket with a handle. Oh look, it has a handle now, so you can carry the bucket with the water easier!

That’s exactly what the Asset Browser situation is right now. The frustrating part is that the shiny thumbnails aside, it just doesn’t provide much of improvement over manually making libraries of .blend files and manually appending from them.

On the one hand you may think removing it may be a bit too radical of a step, but after Mantaflow, this is yet another feature that establishes a precedent it is okay to add a feature to official main branch of Blender, get it only 80% of the way towards usability, and then stop the development and let it rot.

I guess the asset Browser wanted to give faster acess to blender assets, organize, preview and filter them. And this is exactly whats possible right now. The storage of assets itself is another complex issue (see video link)

My argument was, that even if all your suggestions would be implemented, you would safe a few seconds per asset. The primaray goal of the asset browser is browsing, but you complain the asset browser to deliver a complete concept of asset storing.

For example this is under current development and allowing for better filtering assets #105841 - WIP: Assets: Implement Traits - blender - Blender Projects. Then a few other developments could be done but then the Asset Browsing is really mature.

After that the next big topic is Asset Management/Storing, which also includes Versioning, maybe online services. But this wasnt the goal of the Asset Browser at all in my opinion but you complaning about this…

Be patient the developers will adress this whole other topic in the future