[NEW PATCH] uNature - GPU, Interactable Grass. Interactable Trees, MultiThreading and more!

Uhmm go to the foliage category on the UNTerrain, is the height toggle on?

 

Hi Elroy, Is there any tutorials/documentation for calling functions on uNature at runtime? Changing the colour of the grass for dryness, growing when it has rained etc,,,

 

"Update foliage on height change" is checked

 

Hey,
No official API, but to edit the prototype settings you simply call ->
uNature.Core.FoliageClasses.FoliageDB.sortedPrototypes[prototypeID(Shown on the foliage window)].... (configure settings there)

 

Oh I see, probably a bug on my end then. Will check tomorrow!

 

Hi,

When i tried to build for "UWP" i'm getting this error / having the same error with some uConstruct scripts too :

"
Assets\uNature\Scripts\Core\Terrain\TerrainData\UNTerrainData.cs(12,36): error CS0234: The type or namespace name 'Formatters' does not exist in the namespace 'System.Runtime.Serialization' (are you missing an assembly reference?)
"

There is any UWP support for uNature or uConstruct ?

Thanks

 

Hey!

That's a good point, I forgot that UWP has some issues with the Formatters.
I will look into a fix for the upcoming version, please wait for now if possible

 

Hi Elroy, My grass seems to only first populate the game when my characters moves. Grass is invisible until movement. Anyway to get the grass to spawn in before player moves.

 

Uhm never heard about that issue but I will look into it in the new release, a few workarounds that I can suggest (Also is your player position 0,0,0 ?):

1) When the game starts, disable the UNSeeker and enable it (Might fix it)
2) Before loading screen ends move the character a few units forward and on the next frame return it to the original position (If you have a loading screen)
3) If none of the above works, let me know and I will try to get you a faster fix

 

Update regarding the new version:

Starting to update the trello right now with the new gameplan. Feel free to comment on the trello/ forums with suggestions!

Note that it's going to be quite big so not sure how much time it will take, but it will take uNature to the next level. I am not sure about upgrade or not as the upgrade from TMS was indeed free so I am thinking about:

1) Update for free for those that bought it in the last 3 months (will hurt early supporters)
2) Release an update that fixes all of the issues with the current version and then make the upgrade an option
3) Have no upgrade at all (Which might hurt my ability to support the asset in the future, which I don't want to happen)

Upgrade price will probably be very low anyways, and I prefer option 2 so even if you don't want to upgrade you will still have a fully working package and I will not hurt early supporters.

The reason I am thinking about upgrade price is because the upgrade from TMS was free and it hurt my sales a bit (not blaming any one for this, it was my decision and I am happy with my ability to help the community) and I really want to keep uNature going. And considering it will be a huge update I hope it will help me and maybe one day it will even be picked up by unity

Please give me your thoughts on this, I really want to make sure every one are happy and I will not do any upgrade version unless the current version works perfectly.

Btw trello - https://trello.com/c/Td88AQwP

 

My two cents: I'm very close to abandoning uNature because it still doesn't work with MapMagic. No interest in upgrades or anything until the version I paid for works as was/is still advertised. Maybe the procedural generation side of MapMagic isn't used by everyone but you've promised me big fixes for months now. IF that was working I'd be fine paying for an upgrade if it was truly that, a big step with new features. If it's just the "new version" people aren't going to feel good about paying for it twice.

 

Absolutely agree, as stated -

Making the current version working is first priority of course

 

I have had uNature for a long time but not used it.. until just recent.

I am a hobbyist only and only got real serious after i recently got my oculus rift a few months ago. To be able
to be inside my game is a huge incentive. I only re-picked graphics as a hobby a 2 years ago..
was in my 20's it was way to expensive on win 3.1 XD ...

Long story short. I would not mind a reasonable upgrade price. I think as the developer for all the
work you have done.. your not being unreasonable at all.

P.S I do not know how to code at all, but i think if you fix the map magic problems ( I don't use it, but own it as well)
but once it is fixed for a specif version of MM, if it changes in the future and it breaks, then I don't think you will be duty bound to fix it.

 

I would not begrudge a reasonable upgrade fee, no doubt you have to make it worth your while as a developer to support and update the asset. There is a huge amount of promise in uNature.

Make it pay and deliver the goods I say. Too many asset develops do not support the asset or communicate and you are certainly not one of them, you deserve my confidence.

PS add trees

 

Thank you very much guys... It really makes me happy to see that

"P.S I do not know how to code at all, but i think if you fix the map magic problems ( I don't use it, but own it as well)
but once it is fixed for a specif version of MM, if it changes in the future and it breaks, then I don't think you will be duty"

Absolutely! I want to make sure everything is perfect first! Actually the main reason I am doing this huge revamp is because the new approach makes integrations really really easy and should fix the MapMagic integration one and for all but will make sure its working on the current version too!

PS: Trees stay of course! I won't be touching them Only grass refactor for now

 

Hey,
So, can you name any timeframe when / if uNature will eventually fully work with MM generation ? I also bought it for that purpose alone month ago and wasn´t able to use it once since then. For planning reasons I will now start looking for alternatives because I cannot depend on promises alone anymore.

 

Good question - honestly I can't say right now and hope it will come out in the next week or two but dates aren't really my strong side. The issues with map magic aren't really map magic related and its more of how the latest uNature system works and the need for it to use maps (Which is why you get those spikes) So I am now trying to find a way around it which is pretty much the highest priority.

Note that in the new version the architecture is completely different and it should adapt to any terrain system automatically without the need for me to integrate it. (First will get a fix for the current version though) [That uses the unity terrain of course ]

If you bought it only for that purpose I would recommend looking for other systems right now and if you want I will gladly refund you

 

sounds mega cool!

 

Good point but probably fixing MapMagic will take 1 week of rapid work which is what I am focusing on now, afterwards I will start on the new version as I don't want to leave uNature's integrations broken for the time being

 

I know your right... hehe , a week won't be that bad in the long run either.

 

Hi Elroy,

What is the timeline on the new version. I'm looking at release dates for my game in the next couple of months. Will it be with the current version or will the new one be ready for then?

Edit :- is there a plan (or already) a way to expose specular/smoothness values or be able to create a wet look to the grass?

 

Honestly can't tell at the moment :/ still planning out the scope of the update as I want it to be absolutely insane and stacked up with some new features

Uhmm will add!

 

Ok some more updates regarding the new version ->

Regarding the map magic fixes:

Been looking into the best ways to workaround the issue and seems like I might have it fixed very soon! I have some more ideas if it fails so I think I am close!


Regarding uNature 2.0:

Been working on some of the pure foundations like the surface management, patches, lods, prototypes etc and was able to get the new patches generation to work and darn it's so much nicer! (Both code wise and less buggier)

Note that its still in very early stages and so I don't have anything to post yet. I hope to start a WIP thread in the future and gather some testers and hopefully this time the release will be much smoother!

 

@ElroyUnity Count me in!

 

Hi @ElroyUnity, is there a way to make grass really dense (i know that will not be performant)?
I mean near that terrain can't be seen, I'm trying to make low and very dense grass.

 

Uhmm some stuff you can try:

1) In your prototype you will see "Max Generatable Density". Increase it to 20 and try to paint again while setting the density on the paint to 20.

2) On your prototype, increase the width and the height (A hack that a lot of games use in order to achieve dense grass while not rendering too many blades. Of course it might not look as good as using lots of blades and could cause overdraw)

Note that just like the unity terrain, there's actually a way to set the detail resolution but its broken. I might get a fix for it in the upcoming map magic fix or just leave it for now as it's already in the new 2.0 version .

 

Thanks for the fast response!
Well, one more stupid question this may be in docs, so maybe I missed it.
How to create new prototypes?

 

Hehe np!
And just select your textures/ mesh in the project and drag and drop it into the prototype area inside the foliage manager

 

Could you do some tests with more than a few trees, I have about 70 different trees and it just freezes unity for like 20 - 30 mins everytime I load the scene or stop playing the scene. Or as soon as I add the script it will hang too.

Thanx

 

Uhmm interesting, will give it a shot tomorrow!

 

Hi again, may I leave here feature request? It would be great to get LAM (Landscape auto material) integration.
Thanks your grass is the best .

 

Hey - Not sure to be honest if there's even a user base left for the asset as its kind of not active any more. I will make an integration poll once the new version is fully stable and only then will look into new integrations

 

The most activity goes via email and they too prepairing lam 2.0.

 

hey, so we've had unature sitting around for a long time, but just started testing it now. We have simple mesh grass, which is all working except that vertex normals are not applied...?

Using unature 2.2 and unity 5.6.1f1.

I did think of outputting normals to a texture, but it would be somewhat tedious.

Also, I outputted vertex normals before CalculateGPUVertex, which revealed it was clear before then...so I'm a little confused as to when the info is lost in the first place. I'd guess you're recreating the mesh and not copying normals?

Have to say though, pretty impressed with everything. Wind and touch bending worked well, and it seems you've done just about everything imaginable as far as performance goes. The UI could better though, in that it would help allot to have your notes in the manual as tooltips on more stuff.

 

Hey!

Uhmm interesting regarding normals as it should be copied perfectly. I assume it might be something with the shaders not applying the normal orientation (Will have a look).

Regarding the UI and stuff - Actually working on 2.0 now which is a new version thats taking uNature to the next level
It means -> full rewrite, new ui and much more features

 

Hi Elroy,

I am looking for a solution that allows me to add and remove trees at run time without the HUGE fps drop that occurs when calling SetHeights. I would like to be able to remove potentially hundreds of trees from the terrain. The solution doesn't have to remove all the trees in an instant; removing the trees over a minute or two would be satisfactory as well.

Am I right in assuming your API document is only available once purchased?

Thanks

 

Hey!

Yes, but uNature won't help you in that case. It uses the terrain trees so when you remove hundreds of trees it will still have the issues you noted above.

Maybe in the future I will make my own solution for trees similar to the grass aspect of uNature but atm its quite far away

 

@RoloJo critias is something of a tree version of this. At least, it is in that its for increased performance due to unity's poor culling system. I've experimented with it back when I was working on a game where we'd attempt to have (potentially) a couple thousand speedtrees and a few more thousand unity trees for a forest vibe. The results were modest, but worthwhile. Don't know if it has Add/Remove tree functionality but it'd be trivial to do.

That said, I'd be interested in any other performance related stuff you do, Elroy

 

Hey, sorry to bother you again.
Small feature request, can u add touch bending trace (grass not going up (to it previous position) after being pressured), like this.

 

Hey,
The current touch bending is a lot different and so it won't work like that, but I will work on it for the new version

 

I just bought UNature and so far so good. But I do have a question regarding the touch bending. It doesn't seem to be working with objects or even just the player. Not sure if there was something I missed setting it up but the "touch bending" box on my grass is selected and I even tried playing with the strength and there was nothing?! Any ideas? And yes I'm new to unity and I code like a potatoe.

 

Can you get me screenshots of:

1) The touch bending object on the editor scene window
2) The touch bending component settings (inspector)

 

Ok so I forgot to put the script on the player.....my bad.
Thank you for your fast offer of support.
Regards

 

1. How do I determine which version I am using? I looked in the editor window and inside some of the scripts but could not find it. If I had my way, the uNature version number would appear at the top (or bottom) of the Foliage Manager window, which I keep in a tab when working with uNature.

As far as feature requests, the most important feature for me is core function stability. If I can add your asset to my scene, use the tools as described and get great looking grass, well, that just makes my day every time. I don't mind paying for an upgrade to any asset that works as advertised and keeps getting better.

Final thoughts:

2. You might revisit the grass appearing in row feature, which is much improved but can still be observed at times.

3. I once tried to apply grass to the entire scene, but got patches rather than complete coverage. I might not understand that process, yet, so this one could be on me.

LOVE uNature.

 

Hey!

1) Go to window/uNature/Settings -> should be on the top
2) Yep! working on the new upgraded verison of uNature with lots and lots of changes, will fix it for sure. Note that I will revisit it in the current version too so its not unfair for people that don't want to upgrade.
3) Uhmm what do you mean? did you get any errors? are you using the latest version?


And thanks

 

Found my version number, 2.2, right where you said it would be. Thanks.

I tried to research what I had done when I thought I was applying grass with uNature to all terrain in a scene. I reread the documents, but I do not know what I did, or how I did it. I do not recall any errors, and I get no errors from uNature during play.

However, when I pause play, the grass disappears from both the scene and play views. When I end play the grass usually returns to the scene view. If I cannot see the grass in the scene view with play off, I can get the grass back by selecting the paint brush, a paint pattern, and a grass. Note: I never uncheck "enable " in the Foliage Manager. Not seeing the grass during pause is an inconvenience, but not a major problem, because I can always turn play off to work on my grass.

I use Megasplat. Does this mean "use color map" will not function. I would love to use color map.

I do get one error in edit mode scene view, when I mouse navigate. FYI it is:

 

Hey ->

Will look into the error but regarding the pause, yep expected, will make it render on pause mode in the new version

 

Hi, elroy!
I'm using uNature in runtime to generate the trees' pool from runtime generated terrain tiles for tree colliders.
It is now very laggy and the delta time can raise up to thousands of milliseconds when a new terrain tile is under processing.
Any ideas about how can I optimize this function?

 

You can't... afaik there is no async way of doing that. However, we have a (slightly) similar issue where we Instantiate hundreds of assets in our scene generation... so we just use a loading screen and wait a frame between each instantiation, so the framerate doesn't turn into MS PowerPoint