Standard Assets 2018 - let us know what you think!

oh yes and one last thing. Great to hear about what's behind the curtain in the source.

If that makes it easier for us to understand and its so close to Nvidia's code anyway why not just open source that bit (after a tidyup).

I think source licenses are beyond the means of indie devs nowadays.

 

As part of the review I will post the latest state of play information I can find so that anyone can jump in and say 'But ahah you do not have the latest information'. It will be a sort of mini FAQ like my Android advertising analysis.

According to this tutorial the input.touch system is still the right way to access touch events. Oddly this says checked in 4.6 !!! wow that's old. 2014 is the date on the video.
https://unity3d.com/learn/tutorials/topics/mobile-touch/multi-touch-input?playlist=17138

I must say that this is not even enough code to describe making a touch joystick. So in the AngryBots code we see how to script a joystick. We rewrote it in C# but Its still slow to be doing it this way as I think it should be at the C++ layer and function like a normal joystick.

Next step is to re-evaluate the gui rendering and check if there is the same problem with the new UI being too slow/wonky and find out again what the issue is again.

 

I don't really see how this is something unity would have a "position" on. Whether or not to use rootmotion is a question of how you prefer working and what kind of end result you want. It's entirely up to you. Both approaches have their reasons to be used.

I often end up using both in the same project, depending on the specific animation task to implement. For example my character movement will have no rootmotion, but my "climbing over obstacle" animation might have it

I'm having trouble understanding what you mean by "it can't do top line 3d animation", and what you want unity to explain exactly

 

Hi Phil,

Thanks for your comments- I will try to answer as best as my limited understanding allows.

Rootmotion is not used in UE4. At least as far as I could see. There is a switch but it didn't work for me so I converted them using Blender. The controller is fluid and reactive and the joysticks are responsive on a touchscreen. Its just works.

If Unity was to have a standard cc like UE4s it would not use rootmotion. A standard set of animations either would or would not use rootmotion. I'm really talking about the movements that are usually coupled to the joystick, turning of the character being the main bugbear in my case.

As to being entirely up to the developer. I don't see how they can do that without the appropriate examples and documentation.To my knowledge there is no detailed discussion of each technique, the physics of it and the UI controls that drive it. Also keep in mind that the more plumbing we have to do the more we aren't making a game.

UE4 succeeds in having a biped system that works due to standard skeletons, and a standard controller. Then you go from there.

I think its pretty important for a baseline cc to assume one or the other in its design. In the past some of Unitys rootmotion based stuff has led to jittery animation.

Also the rigid body is important when using rootmotion from memory. I believe the issue is does the system drive the position of the character using physics or does the character controller govern it more directly (non root).

Its important because when we get to the input control a non root system supposedly can control the character position exactly because its non reactive.

Regards *IF* Unity can not do top line 3d animation. Simply look at Unitys own example videos. Even recent ones show the state of the animation. Its jittery jerky and certainly not like a pro console or PC game. My own problems showed root motion to be difficult. Many people have stated that its harder to get the fine control needed. If you recall the example with the professor on the space ship. The graphics were sterling but the animation was not. If Unity can do top line animation then perhaps even their own teams are not in on 'the secret'.

I'm at the first stage in my journey to find out why this is all so foggy. I've been round the loop many times before but hope this time it might be a more informed effort. Join me

If I'm a numpty! great!. Then someone can point out my numptyness (but please be precise).

 

Actually Phil reading your 1st post again. you've already stated much of what I say here.

 

Thanks Will your comments at the top regards controller are most welcome and confirm an important part of our issue. That's very helpful for me.

Regards community run events, I personally don't think that's quite the same thing as a company run event. Informal events are great but I think the nature of what was being discussed for GDC sounded like in depth access to key developers which could only happen if it was a special professional focused event.

Personally I've found the UK a very hostile place for Android developers so tend to shy away from the club type sessions. People do love their iPhones here. That's really why I went with the subscription route as I thought it might give me a professional interface to Unity.

The forums work for me because they tend to be more balanced and you can choose what is important.

Anyway thanks for the comments. I appreciate you have your timeframes budget etc.

 

https://blogs.unity3d.com/2018/02/07/feature-highlight-cloth-physics-in-unity-2017-3/
Not perfect and you've probably read it, but it does go some way toward showing ideal setup.

Can blend quicker to other motion or use callback to intercept intended motion (adding info for readers).

@Anjin_nz not an attack but from your posts I can see you probably missed some of Unity's finer points, and that's something Unity would like to address, maybe more docs etc..

(gonna butt out thread as it's meant for Unity's info gathering purposes :/)

 

No problem HC. Our CC is an asset store solution that I have hacked to bits. Its one of the suggested options (by Unity). I'll contact Will and dive out of this.

 

Agreed.

I also want to see the packages be as clean as possible. A character controller doesn't need anything complex like Cinemachine. Basic movement with a decent blend tree, and a camera that collides with walls should be enough.

I wouldn't mind them going further by making generic positioning/rotation scripts and cobbling it together using them. That would allow different camera modes without requiring yet another script.

 

With Unity's push to video generation and OTOY do we also need Standard Assets that will quickly get people up and running making Hollywood level movies e.g. Standard Sets and Actors across a range of genres?

PS Procedural face/hair generation a la...

Will probably be needed.

 

So far loving all the standard assets, really on point..

 

No. If someone is making a Hollywood level movie they will want Hollywood level quality and a standard solution that tries to meet the needs of most people won't be that. Besides anyone who is aiming for a Hollywood level movie is going to have a Hollywood level budget. They won't need a solution that is basically Daz.

 

Yet a lot of Hollywood movies are set in California and use the same famous locations, so should there be a Unitywood or Unityland virtual town that can be used as set locations for games and movies?

e.g.









The fun thing is Unity could even ask the community to help build Unityland or Unitywood



They already have the beginnings of a Viking village, a Robot detention centre, Part of San Francisco, A space station.

It could even be more along the lines of a Hollywood movie sets or Westworld theme parks?

 

Why would a concept like Unitywood or Unityland have any value whatsoever to Hollywood?

They don't have the beginnings of anything. The state those assets are in is their final state.

Who is your target audience? If your target audience is the typical person who starts from a standard asset then your target budget wouldn't be sufficient for the quality an actual Hollywood movie would expect and if your target audience is the big budget industry they wouldn't start from the standard asset in the first place. Basically there is no audience for this.

 

It's value is to game/video developers using Unity.

And yet some have been reused and are allowed to be re-used by other developers. Look at what OATS studios is doing with their work. They are putting it out there for other makers/developers to use.

So once Unity start providing a go to VR/Game/Movie toolset with Cinemachine and OTOY the creatives won't want good sets and 'actors/characters' to try out there ideas and concepts in Unity.

 

I'm not denying that there is value to the concept for some game/video developers. I'm denying that there is value to the target audience you've been mentioning this entire time. Your statements have been about using the assets to allow people to "quickly get people up and running making Hollywood level movies".

Standard assets simply will not enable you to create "Hollywood level" movies any more than they would enable you create a "AAA" game. Neither would they be practical for an actual Hollywood director who can simply hire someone with their hundreds of millions of dollars budget to create a quality effect completely from scratch.

Where are you getting the impression that a studio releasing shorts for YouTube is "Hollywood level"?

Prototyping is an entirely different matter. Using them for prototyping wasn't mentioned in your posts though until just now.

 

My idea is that Unity should have assets or a world/land that has playable characters/npcs and settings that allow people who want to make games/media/video a set of props that allow them to do that quickly and easily.

Most creatives start with little more than their own imaginations... Hollywood studios have locations and entire back catalogues of sets, costumes and props.

So why doesn't Unity adopt the Hollywood Studio approach and provide a good range of sets, characters, costumes and props and the tools to use them as standard?

They could even buy in good asset store content to build up a standard studio.

 

If creatives want to have a large library of characters and environments to pick from, they should invest in making them themselves. However, as a general tool, Unity should provide acceptable prototyping tools which we're seeing them do more and more with things like the Pro Core acquisition. These are what the standard assets should be, not ultra high production value assets.

 

I just filled out the survey - would have been sooner had I seen it.

I added in the comments that I also use the standard assets for quick access to terrain textures, grass, and trees, and water (forgot to write water there, though).

I guess there's not much more to say, as I left my comments in the survey, but I like the idea of templates, and a library of useful assets for users to use.
Anyhow, looking forward to seeing what happens.

 

Unity standard shader that supports vertex colors.
Amplify has vertex color support but does not have a Detail Texture.
How hard is it for one company to support both?

 

.. I think you already posted this, in the thread.

 

Yes I did.
I've posted it in several others too

 

I would separate things, because things are getting mixed up:

I would make the
- standard assets for prototyping and show off what kind of things can be put into the engine (one or two human model(s), generalized, one car, one flying vehicle, etc)
- ready templates for people who start to make specific projects and they don't want to setup their project from the ground up or they want to learn how to do things (like we have nowadays, FPS template, puzzle template, platformer template, endless runner, tower defense, etc...)

 

I think Default Character Controllers would be a great thing to have for various styles of games to show different physics possibilities (such as 2d jumpthrough platforms/slopes) and interesting camera and character movement styles people can use in their games.

Also I think it's a great idea to split up the Standard Assets. However, one thing to consider is how hard it then becomes to finding a specific "Standard" asset in the Asset Store. -- I tend to forget the exact names of the Unity Assets at times when searching for them (such as "Post Processing FX"), so it might be a good idea to have them prefixed with a particular tag to where the name reads: "Unity Official - Name of the Asset Here" or something of the like so that we only have to search for "Unity Official" to grab relevant /official/ standard assets from the Asset Store.



In regards to the "Tower Defense" thing... D:

I may be a little biased here considering I don't like the idea of ripping-off game mechanics, but the only reason why I can see "Tower Defense" as being such a popular "template" people are currently asking for is that new users with very little programming experience are wanting to use it to do just one thing -- copy the success of the "Clash of Clans" concept.

I, for one, think this one should be put WAY further down the list (despite the number of users asking for it), because it will lead to even more clones of a single game type on the mobile phone space whereas legitimate Unity users would be better served with more general-purpose programming patterns. This is, of course, my own opinion. How to manage the enemy and player units that could be put into a tower defense game would be a LOT more useful than a simple "re-skin" of an existing game concept (which is almost constant nowadays on mobile devices, independent of whether they are a game or an app.) Personally, I feel like you guys would be doing the world a favor by subduing these kinds of cash-grabs.

 

https://assetstore.unity.com/publishers/1

 

Dude, it's already out.

What I'm still not clear on is how all of this is going to be delivered. Through the hub? Though the asset store? Through the upcomming package manager thingy? If it's the third, does that have an interface? Is it searchable?

 

Turns out the package manager is really well documented, so that's nice!

I'm a bit confused how the templates and standard assets will interact with the packages. From how I understand it, packages are meant to be handled in an npm-like fashion - they have versions, and you use the manifest file (or the manager window) to make sure they're installed or not.

But a template is something you will do heavy edits to - so uninstalling it doesn't make much sense. You're probably not only going to be adding assets and code to what's already there, you'll also want to edit the template's assets and code. What's the workflow meant to be when you want to start making a game that diverges from a template? Is that a workflow that novice users will handle?

Is there some kind of "make this package not be a package anymore" button?

 

Wow. That goes to show how far behind the curve I am! -- I was expecting a few months! D:

Apparently I've got nothing to complain about then lol -- maybe I'll just stop my moaning and check it out! n__n

Geeze -- you guys have been busy! -- I just looked at the roadmap yesterday and I swear I didn't see anything about this on it lol -- Perhaps I need to start following the Unity Blog more often? No idea how I missed this. Dx

Thanks -- I suppose I never would have guessed Unity was publisher number 1.

But here's the real question -- If Unity was publisher number 1 .... then who was publisher number 0 ........ ???



https://assetstore.unity.com/publishers/0

Spoiler

Even Unity doesn't know....

 

seeing unity dedicating a team for improving and updating standard assets will be great otherwise it's just a waste of time

 

I have an idea, because usually the same people are my patient zero as well, the famous, the non-replaceable, the one and only: TestUser.

 

https://assetstore.unity.com/publishers/0

I guess that explains why clicking on that link does exactly what one expects from the ultimate and unequivocal "TestUser" of our beloved Unity website...

I can't help but feel somehow "meta" after all of this... D:

 

Sorry to be posting this here, but I have no idea were to...does anybody know where/how to contact the Profi Developers people? They used to have some great asset packs, currently deprecated, and I would love to fond out how can I get copy of those files!! Thnxx for any help and also thnxx in advance if you can point me to where I should have posted this question!!! )

 

@willgoldstone

Has anyone at Unity thought about Standard Assets "templates" for Writing Editor Extensions and Tools?

One sorely needed example is how to deal with input across multiple editor windows and how to handle global user input. I currently have to hack this for some tools (such as being able to let the user type when I use a single key global shortcut they need to be able to disable). I think it would be really nice to see an example of how you guys would write a multi window editor tool that also acts in the scene view and uses global shortcuts that can be disabled when the tool is no longer in use.

Anyone up to the challenge?

This would be a great standard template to include -- I don't think enough people write enough tools that can really get down and dirty with workflow enhancement because much of this stuff is so obtuse. It's like black magic to many users who try. Nowhere near enough documentation on it either.

Spoiler

In fact, when I started working with Editor Extensions the first time, the very fact that Unity had TWO entirely-different systems of "input" blew my mind. I later find out the GUI input system was very buggy and didn't work how I expected, but for some stuff, it didn't work at all (tooltip and cursor issues, setting a checkmark on the File/Edit/etc. menu had to be delayed a frame before I could initialize it with an EditorPrefs variable, and all sorts of other quirky/wonky things that come with this Unity Editor scripting (such as serialization issues; missing keycodes; unpredictable order of events) The very least we could get is a few templates to show how to work with things like: the filesystem, keyboard controls, multi-window editorwindows that talk to each other, single-key shorcuts that can be toggled on/off depending on the tool mode, setting the tool mode, /invisible/ editorwindows, persisting data, how to check for key/keyup/keydown events, how to get mouse input, how to actually style editorwindows, how to simply lay them out properly, etc. -- Essentially just VERY useful template stuff to help people make their own tools more quickly. The "easy-to-make" game tools was the initial reason I came to Unity in the first place -- since then, I've found it to be a bigger headache than I ever imagined. Did I mention that the documentation didn't do much to help me? -- Good. I just wanted you to know that someone else probably has these issues too.

Also -- regarding the "tower defense" tutorial and template feedback -- I just wanted to say "Well done!" I'm really impressed at how much detail you guys went into with it on general gameplay programming and design! I apologize for my hesitation for it at first. My skepticism was clearly unwarranted. It did a great job at teaching generally instead of specifically, so my hat is off to you guys! We need more stuff like this! Great work to all involved!

 

Sounds good, one point I would add is making this stuff super easy to find for new users as well as old. Old and new need different directions though - a common mistake for a company is to assume that existing users will find things the same way. They will instead become entrenched on what worked for them one time.

So really, the hub or whatever is used should have "advertising tiles" or a similar concept so when the new user loads it up, they see all these template projects and a nifty looking char controller... wow that's what she wants! click click!

As for existing smelly people like me, blogs, usual modes of discovery. But new people need a different "ad" each time to drip feed them important things to help them try something new or discover something new over time without overwhelming.

/butts out again

 

Could you update std asset characters for HDRP

Code (CSharp):

1.     PlayerSettings Validation: Requested build target group (20) doesn't exist; #define symbols for scripting won't be added.
2.     UnityEditor.PlayerSettings:SetScriptingDefineSymbolsForGroup(BuildTargetGroup, String)
3.     UnityStandardAssets.CrossPlatformInput.Inspector.CrossPlatformInitialize:SetEnabled(String, Boolean, Boolean) (at Assets/Standard Assets/Editor/CrossPlatformInput/CrossPlatformInputInitialize.cs:127)
4.     UnityStandardAssets.CrossPlatformInput.Inspector.CrossPlatformInitialize:.cctor() (at Assets/Standard Assets/Editor/CrossPlatformInput/CrossPlatformInputInitialize.cs:22)
5.     UnityEditor.EditorAssemblies:ProcessInitializeOnLoadAttributes()

 

I made it to the bottom of page one and your first response before my ADHD chimed in lol.

I believe what the overall thought of Templates from what I read was something similar to a "Third Person Shooter" Tempate.. Or a "RTS" template. Now don't get me wrong, I really do love and have used "prototyping" some standard assets. The water is cool, the FPS thing was whacky .. The Standard Assets really don't offer anything 'substantial' in terms of visuals... I do understand that the main goal is to provide C# examples, shaders etc.. But it truly is / was lackluster..

I've taken a break (.. as breaks go ) from C++ and doing my thing to watch some videos on what Unity has done with Unity 2018. Maybe incorporating both "template" style games (RTS, TPS, FPS etc_) would be a decent idea.

 

Let me come in very late and say the standard assets are for prototyping, and they do a good job of it. I want to continue to see standard assets, because prototyping is the bulk of the work I do. Finalizing models is the very last step. Code runs so much, and changing an asset can break code. Prototyping assets serve a purpose and I often even end up creating my own for prototyping purposes. Keep the standard assets.

 

Ok where Exactly are the default Character controllers now???? There is no Package marked Character anywhere in the Package manager.....

 

Download these: https://assetstore.unity.com/packages/essentials/asset-packs/standard-assets-32351

 

I'm late to the party but my #1 gripe in Standard Assets is ALWAYS compile errors on standard asset scripts, especially Obsolete APIs. There's no way the average user is going to know how to fix the APIs which are changing, or even know what functionality is lost by the various broken subsystems in all those scripts.

It's frustrating when I import a Standard Asset and it pops up with the "I need to attempt to update this old clunky code with the latest APIs. Have you made a backup?"

It's even more frustrating when this attempt at updating APIs fails on Standard Assets. The whole point of a Standard Asset is that it should work out of the box. Users can freely tweak them afterwards if they're not exactly what the user wants, but the original unmodified script should not be actually broken.

I should never see “Some scripts have compilation errors which may prevent obsolete API usages to get updated. Obsolete API updating will continue automatically after these errors get fixed.”

 

I can totally understand your frustration on this. One thing to keep in mind (and it was hard for me to wrap my head around it until just recently) is that just because Unity is a huge gigantic corporation with tons of employees, it doesn't mean they have enough employees to cover everything they're trying to do.

For this, they'd essentially need to either hire a "miscellaneous-tasks" guy OR assign an (existing) employee to monitor the standard assets packages for each major release of Unity. That doesn't sound like a lot of work, but when the Standard Assets are likely to get deprecated in favor of a new system of providing assets, I can see their hesitation to actually hire someone for this role specifically, and it's really bad for organization to have a "miscellaneous" drawer to put things in -- at some point, that drawer will start overflowing and things will get out of hand.

That said, I'm not sure what their excuse was to not "assign a guy to it" before, but that's definitely a learning point for them -- and one they seriously need to take heed in.

The one danger Unity has right now is "becoming another Google" by putting their eggs in more baskets than they can actually carry. Google at least has infinite money to hire enough employees to carry those baskets. Unity, I think, is probably having to come back and make a few extra trips (which is why it takes them so long to handle seemingly basic stuff sometimes.)

How far off the mark am I on this, @Unity or @willgoldstone? -- Am I missing anything?

 

Constantly updated Templates with each version of Unity you place into the public. Templates being "Third Person Shooter" or "First Person Shooter" or "Real-Time Strategy" ... Pick your poison but the post above isn't wrong.. It makes Standard Assets, Useless Assets. (No offense intended and the more I understand the engine, the more I love it..)

 

If I want to download std assets from https://unity3d.com/get-unity/download/archive , it says File not found."

 

I have first person Rb std asset and it works in HDRP only it gives warning messages in console.

 

I have to admit I found this post to be a very odd one. I can understand not having it enabled by default (and it's not by the way) in its current state but to not make new and very powerful functionality available to developers just because an older collection of assets is not compatible would be bizarre.