Now that Unity has level building tools built it does it have any destruction tools yet?

Unity has adopted Pro Builder for generating levels, so it has mesh generation code, but does it have any in built mesh destruction tools/APIs yet?

 

@Arowx can we just not?

 

It doesn't have "level building tools," it has "greybox tools."

This is, of course, a standard "Arowx makes an unreasonable request" thread.

 

Unity makes games, but why can't it make coffee? As a creation tool it's clearly lacking in this area.

 

Sure. You can destroy any asset with a single press of the "delete" key.

 

It won't be long before we finally have DOTS Destruction!

 

Well with a physics engine called Havok you would expect it to have some kind of destruction system.

 

Just because *some* people use it as level editor, it does not mean it was intended as such. ProBuilder is admittedly a prototyping tool.
Just like the Standard Assets, and people made some games with them, it does not make them intended for production.

It had one: https://cargocollective.com/ysalmi/Havok-Destruction
But obviously the trademark does not mean the same as back in the days. And, surprise, Nvidia does not sell physics accelerator cards anymore.

 

I'm aware of that, BUT saying that PB doesn't have "level building tools" is - as you can clearly see - not quite true.

 

Well, the original statement wasn't about PB. It was said that Unity does not have level building tools. It has grey-boxing tools. Also don't forget this is an open forum, all kind of people get all kind of wrong ideas when this kind of misinformation start to spread. It was the same when people started to ask for advanced behavior for the new Standard Asset Third-party character controller for prototyping.

 

Now that we have a game engine, why can't we have an anti-game-engine? /ArrowX threads in a nutshell XD

 

Whats the deal breaker? No UV2 generation? I might look to pro builder to change some aspects of a modular asset we have for our game. (Its at a wrong scale for VR so need to scale it down but then some things look too small like door ways so need to make new door way assets). But if I cant make it a prefab and reuse same mesh or no UV2/lightmapper support I can understand why it cant be used in production-

 

UNITY hang themselves by asking which game to emulate, got answer of mario 64 and zelda BOTW, which is implied an advance system, and then people were discussing about "hooks" to the system to allow custom upgrade. To be frank they didn't present it as just a prototype system, but as a strong basis from which you can build upon.

 

There is also a difference. A character controller is a very large portion of the feel of your game. If you want generic asset flip feel use the ready togo controller. If you want custom feel, build your own.

Its not the same with a editor time tool like pro builder.

 

In theory that's true, that would be ideal if every game was original masterpieces, but a lot of the feel is also not the whole character controller, but the tweaking of its parameter, a lot of game are not action game with fancy movement, so a generic character controller, from which you can set acceleration and speed, and have basic function like swimming and climbing (or at least hooks to build your own) is mostly sufficient for more than half the third person game, especially RPG like.

The thing is frequenting thread of asset character controller, the demand are mostly always the same since the beginning, basic movement, platform, swimming, climbing, and hooks to make everything else. People are asking for a solid base, mostly because making a competent character controller is actually hard.

 

I know its hard, I have made one, for VR never the less. I would still argue to make your own controller, but sure you can make it your own with tweaking. The point still is though, pro builder is a editor time tool, and not the same thing as standard asset stuff.

That said I have no clue if the pro builder tool is production worthy, I guess I will find out the hard way

 

Yes, the intention is that it's a prototyping tool, but just like Unity has evolved beyond being just a game engine ProBuilder will eventually evolve beyond being a prototyping tool if it hasn't already. Intentions don't mean a whole lot in the long run.

 

What is destruction of a 3d object, a way to slice, cut and fracture a mesh. If you can cut and break up probuilder meshes then you have the basic tools to create a destruction system for creating holes, breaks and fractures.

 

It does have destruction tools, more precisely - wallet destruction tools, they're called IAP

 

detach face, Alt U, Alt X, Backspace, Alt B

 

Havok Destruction system demo circa 2008, it's 2019 now come on Unity!

 

Havok is a physics engine, not a game engine. The reason it exists is to extend things. You're more than welcome to license a physics engine with more advanced features.

 

Indeed. You have the basic tools, so you go ahead and make the specific advanced tools you require. Or use one of the many third-party solutions out there.

It is entirely foolish to assume that the mere existence of some specific advanced tool mandates that a similar advanced tool intrinsically has to be provided for you.

 

@Arowx here's the thing - I'm not seeing your actual use case. Are you working on something that needs this right now?

I appreciate theory as good as the next person but you're just turning up with a question and leaving it, I'm not sure if anything at all occurs as a result of this kind of discussion.

 

@sxa it's a built in feature of Unity's main competitor.
@hippocoder Use case fun and realism (see video), with games throwing around bullets, rockets and bombs in HD surely we should also have great looking destruction physics.

After all with all this processing power DOTS releases we should have something to show for it in lower Unit count games. Unless you have tens of thousands of Units DOTS is kind of idling twiddling it's thumbs. Destruction could fill that gap with amazing dynamic fun gameplay.

 

That's not a meaningful use case to have Unity do something for you.

 

Yeah well what is destruction to you? is it mesh cutting in realtime? is it physics or spawning?

I suspect you are really saying "lets generate new convex meshes and colliders at runtime" ? cos that's the only thing I can imagine people ask for when they ask for "built in destruction".

Since the physics are already handled (using havok, not unity physics cos unity physics won't stack much).

 

I definitely would not want something like that, but if you can't brainstorm there is always NVidia Blast, it's an extension of PhysX that actually handles all the physics and colliders and even fracturing for you, and it's open source.

Why doesn't Unity seems interested at all in GameWorks middleware?

In any case I would rather built-in destruction or fluids over built-in ProBuilder.

 

Possibly licensing costs or GPU affinity issues, a lot of consoles use AMD GPUs.

Note that the Havok Physics engine has a Destruction focused toolset or did circa 2008. Unfortunately the Havok Physics engine does not appear to publicly document it's API so we don't know what Unity are using and not using from the available tools the Physics engine provides.

At least with Nvidia's open API approach you could look for features that were in the Physics engine and ask for them to be added to the game engine. e.g. Origin translation for large world support.

 

I guess I should say this again but:

You can license and/or implement these things yourself. These are not general purpose tools and do not fit with the idea of what a kitchen sink engine does.

 

Because it's totally realistic to be carrying around the amount of explosive power needed to do anything in that video.

 

So Unity is a Kitchen Sink engine and Unreal is a AAA engine, I thought Unity was in direct competition with Unreal therefore any feature that gives one engine an advantage should in theory be adopted by the other.

It depends where in the world you live, in Syria right now along the Turkey Syrian border it's probably the kind of fire power you need to have to survive. Or yes in any military conflict scenario anywhere in the world that kind of explosive power is what will be carried, although more often mounted on vehicles and aircraft. What you see in the demo is more akin to what an A10 warthog or modern armoured vehicle could do (although I suspect they would run out of ammo in less time than it takes to run the demo at that rate of fire).

Also some places in Texas but I'm not sure what's so dangerous in Texas...?

 

Both of them are kitchen sink engines, but Unreal is constantly putting in things specifically geared towards genres of games. Stop being willfully obtuse for once and actually think about your posts before making them, please.

If you are so desperate for UE4 features just download it.

 

Last time I remember UE didn't have DOTS. But apparently, Unity is not like trying competing ...

 

Some developers have implemented their own ECS architecture in UE4.

https://forums.unrealengine.com/dev...ng-ecs-architecture-in-ue4-giant-space-battle

 

Yeah I remember that post. Unity Asset store had also some ECS frameworks. And some massive nice GPU effects, before Unity ECS was loudly announced. Weather they were competitive back then that is different story.

However, while whole destruction concept is cool, I think ECS / DOTS has wider target audience, than destruction feature. Most of time destruction is mimicked successfully with explosion and particles. Alternatively with premade models. So is not like end of the world, without it

 

So one company (A) did limited purpose destruction and another company (B) decided to change how games are developed for the next decade for all developers and hardware. I think I'm picking B.

B also gave the tools to solve destruction. I'm pretty sure you can find a decent convex splitter on github. Port that to DOTS and bobs your uncle's stepsister's wife.

Or did you expect me to do it? : P

 

I wonder if I can somehow derail this thread towards being disappointed that Caronte got deprecated (because it was pretty awesome).

 

Well, the Ragdoll wizard, tree creator, Unity Recorder and all the XR stuff aren't so general purpose either, on the other hand we don't have a tween engine, spline editor, a pool manager or even UV viewer... and the list goes on...

 

How do you want me to explain to you that games are more likely to have trees and characters that fall over than destructible assets?

Or gameplay footage for marketing and communication reasons?

Or platform support?

There's a billion tweening assets out there, many of them free, and they're not even hard to implement on your own thanks to the animation curve system.

See above.

Different games have vastly different pooling requirements and none of them are difficult to implement on your own.

You know what does? Your modelling software. If it's for stuff you're building in editor, you'll never guess what ProBuilder has.

 

OMG, we don't have "make my game" button in Unity. What should we do now? :O


(BTW, I'm fairly sure that there are people who are taking a look at some kind of destruction system at Unity, at least on the demo or proof of concept level. Because they just usually do that.)

 

Well guess what? you can also make trees, ragdolls and marketing material with 3rd party specialized software.
But when you can't review the UVs of your imported model in a "3D" engine, like every other engine on the planet, I think that's some really displaced priorities.
What happens when you iterate 100s of models and textures or don't have access to a modeler, what's the point of having Swap UVs when you can't see your UV channels?

I'm not dissing Unity for their hard work, but I'm perplexed how can you downplay the importance of certain essential tools like Tween and Spline and other utilities that we constantly use and must download an extra plugin for when you praise the integration of useless gimmicky things like the Sprite Assets Creator in Project Window that makes you a life-saving 4px by 4px empty masked sprite in various forms like squares and hexagons...



Come on give me a break.

 

Do you mean they adopted a batch based simd approach https://forum.unity.com/threads/could-batched-vector-ops-speed-up-unity.253924/

 

I see you're a member of the Arowx school of false equivalence. It's not just that they're available as third party software, but the breadth of their use cases.

Wild how everyone else seems to get by just fine, with Unity being the most popular engine on the planet. I literally deal with 3d models all day and have for ages and I have had no problem here because the UVs typically don't change. This is an edge case.

No, you give me a break. You're acting like these are monumental things that need to change but it turns out that all that really needs to change is your workflow.

 

If you think I mean faster than C++ cross platform performance done far easier than ever before? If so yeah.

 

Yeah, we all know that. So what? Its barely been in the full release for a month. And yet umpty-million games managed to get made without it.

Are you trying to argue that Unity has to feature-match every single thing that its competitors have, just because they have them?

I guess so. After all, that 'main competitor' keeps up with all the features Unity has, doesn't it? Right? Right?

 

@Murgilod Dure I'm just following your logic that's all.

I'm not saying those are make or break features but if the UVs of your models don't change it's either because they're dead simple that you don't need to iterate and optimize them, or you get 'em ready from the artist guy that does all the UVing and baking.

So this is not an edge case, the only edge case here is that hexagon sprite, did you ever need a static lowres hexagon sprite that never changes in your game?

At least Microsoft thinks different than you and gave all its Windows 10 users (game devs or not) a built-in 3D Viewer with UV view AND Vertex Colors AND Material IDs AND PBR channel overlay views.

NONE of which Unity provides, and they don't even have access to FBX SDK source.
So why didn't any body convince Microsoft that it's not their job to do it.

Listen this is not X vs. Y kinda thread, we're only here to give feedback to ameliorate and plug the holes in our favourite 3D engine, so dial down on your enthusiasm.

 

I posted my attempt at porting elasticsea's chaos-like-destruction project to dots over in the dots forum https://forum.unity.com/threads/destruction-with-dots.761162/

 

These aren't holes though and, again, most games don't need to iterate hundreds of models and textures without access to the modelling software. This is not something loads of people need. It's actually such a niche thing that there isn't even an asset for it.

You're acting like this is a more common issue than it is and that's especially ridiculous when there are actual issues that need attention.