Search Unity

  1. Megacity Metro Demo now available. Download now.
    Dismiss Notice
  2. Unity support for visionOS is now available. Learn more in our blog post.
    Dismiss Notice

GAIA - AAA terrain generator, procedural texturing, planting and scene creation

Discussion in 'Assets and Asset Store' started by AdamGoodrich, May 21, 2015.

  1. Wylaryzel

    Wylaryzel

    Joined:
    Sep 13, 2013
    Posts:
    126
    Hi Adam,

    really like the fresh and different approach you are taking for bringing terrain to life in unity :) My current workflow with switching between different tools to get nice terrain is tiresome :)

    As you were shortly mentioning a launch - and I know you are busy with several other pieces of your like (like several of us) - any idea or rough ETA for an availability of the tool?

    Thx in advance
    Wyl
     
  2. AdamGoodrich

    AdamGoodrich

    Joined:
    Feb 12, 2013
    Posts:
    3,780
    Hi Wyl,

    Yeah, I had two full days assigned to it this week plus the usual morning, evening and weekend regime, but have had other business interruptions that have taken up my time.

    I am in two mind's. I could put a smaller release out there that is immediately useful and focuses on just doing a smaller set of things really well, and commit to a road map, or I could delay longer to make it a more rounded asset.

    You only get one chance to make an impression and I really don't want to cause a bad one by releasing too early. There are still some things that are deal breakers to sort out before release, but other features could be released over time and I am open to feedback from you guys as to which approach I should take.

    It's quite functional now:

    * Raw import
    * Terrain generation / terraforming
    * Terrain texturing
    * Spawner
    - design and real time vegetation (trees only at moment, grass by end this weekend)
    - real time monstor spawning
    - pool management
    * Simple vegetation growth
    * Resource manager (good if you want to load and save tree, detail and texture settings)
    - save assets from terrain
    - apply assets across multiple terrains
    * Various terrain utilities
    - smooth, stitch etc
    * Really really cool feature X that I will reveal on launch

    Not tested yet - multi terrain application of all of the above - should hopefully be pretty trivial but its a mandatory feature. Pixel perfection across terrain borders can be challenging to get right. Also, as I have built this, I have developed new techniques and these need to be applied back across the asset - implementation and test can be time consuming.

    So to answer your question - there is a lot done, and some of the more interesting and fresh stuff is still to do, and polish that needs to be applied (e.g. hiding the currently exposed debug settings etc). I also need to create a web site, do documentation, find a name etc.

    Speaking of names - I would like to put it out there to the community. I own www.procedural-worlds.com and have a whole suite of things I want to do under the Procedural Worlds banner.

    I was thinking of Gaia - but lets make a competition out of it - if someone comes up with another name and I choose it then they will win a free copy of the software when its released :)

    So in a long winded sort of way, sometime within the next month.

    Cheers,
    Adam.
     
    BackwoodsGaming likes this.
  3. Wylaryzel

    Wylaryzel

    Joined:
    Sep 13, 2013
    Posts:
    126
    Hi Adam,

    thanks for the comprehensive response. With regards to the decision of the release, I would suggest to consider it based on the unique features you have compared to other contributions in the asset store (and you have some strong ones). As soon as you think you have enough of them, you should release the asset and see how it is accepted. You can always build/incorporate further features afterwards. Not to few assets went that road and were well accepted in the community.

    Yep, multi-terrain application would be a must have feature. Another part would be if it could be used in combination with Horizon[On], besides RTP :)

    BR
    Wyl
     
    AdamGoodrich likes this.
  4. John-G

    John-G

    Joined:
    Mar 21, 2013
    Posts:
    1,122
    Horizon(on) support would be fantastic, especially since it is currently in the 24hr sale.

    Another name you could use since you were thinking of Gaia is another Greek god Morpheus - God of dreams.
    Dreaming of a breath taking environment, Morpheus will morph it into reality. ;)
     
  5. AdamGoodrich

    AdamGoodrich

    Joined:
    Feb 12, 2013
    Posts:
    3,780
    Nice name - I like that ! :)

    The system produces standard terrains.. I have not used Horizon yet, but can't imagine that supporting it would be that big an issue if it handles normal terrain.
     
  6. mwituni

    mwituni

    Joined:
    Jan 15, 2015
    Posts:
    345
    1) Re a product name :
    I've always found when choosing names, its difficult to find something that's quite unique and not used for zillions of other things online ... and thats good for google search!

    I've gone through this a while back for my projects and have a bit of a short list (eg. my username - mwituni means forest / wilderness) ... so I'm submitting some (unique) names that aren't too tongue-twisting but could be applicable to what you're doing ... from ancient African mythology where life started, and the most beautiful landscapes are ...

    Chuku
    is the God of creating land and responsible for the stunning landscapes. In some cultures Odudua or Oduduwa (goddess) is ... btw Odudua gets no google hits ... so is a prime name ... I should be selling it! :)

    The bushmen are the oldest tribe on earth. Their God and ancestor is Dxui ... i have it shortlisted as the "UI" part may be useful in logo design.

    Theres variances on the above names between cultures, but some difficult to spell not just pronounce :)


    2)
    wow, nice one. Pretty good fit for this project too.


    3) I mentioned it before, but need to again ... it sounds like you are targeting single player games. With a bit of thought you can include good useful support for multiplayer games without having each customer have to write a lot of code to support it. That'll increase your market too. I'm referring to thinking about things like consistency across multiple PC's - ie. trees spawning at same place and maturing at (similar) time on each PC. Growth rate etc probably can be a bit off, but terrain changes and trees placement etc would be important.

    While its a workaround to send many large messages up and down the network to synch clients, its inefficient. One efficient way is to generate your procedural content (and growth rates etc) from a generated seed value - like an integer, but also have the option of having it as a variable that can be set externally (say from a server message) - in your editor you may show it as a seed value with an auto-generate button, or allow the user to type any value in.


    4) I get your uncertainty whether to launch or wait, and I'm sure most of us would want to get started.

    I think as long as you have a reasonable product that users can get a decent amount of use from, as well as a clear and well documented roadmap of upcoming features with some sort of idea on timeframes to give buyers confidence, you should corner a similar market anyway. Specially as it sounds pretty unique. Often users might be drawn to products still in development as it means they might have some input into the end-product.

    Going soon in your case should also give you the advantage of being the "first" (with new type of product) and that also helps product success. If one delays too long there's a chance someone else could release something similar. Of course competition is healthy, but this is re being "first" and setting the trend.

    Another advantage , for both seller and buyer, is products released early / with outstanding features, tend to be discounted quite a bit ... so that allows those on tighter budgets (more hobbiests etc) to buy-in ... which you would have lost, as you can't drop price much when its complete without upsetting some buyers who licensed at full cost (other than initial few days etc I guess). So thats a bigger customer base for you, and larger part of the market, plus gives you lots of early adopter feedback while you are completing.

    Guess i'm saying ... hit us with it.
     
    AdamGoodrich likes this.
  7. Archania

    Archania

    Joined:
    Aug 27, 2010
    Posts:
    1,662
    UTerrain.
    no idea actually.
     
    AdamGoodrich likes this.
  8. nesis

    nesis

    Joined:
    Jan 22, 2014
    Posts:
    6
    Hey, I came here from the SGDA Facebook group you posted this in. Just thought I'd say I'm loving the quality of your terrain engine and the screenshots you've produced. It's cool seeing a fellow Aussie dev doing great work :D
     
    AdamGoodrich likes this.
  9. AdamGoodrich

    AdamGoodrich

    Joined:
    Feb 12, 2013
    Posts:
    3,780
    Thanks mate :)
     
  10. AdamGoodrich

    AdamGoodrich

    Joined:
    Feb 12, 2013
    Posts:
    3,780
    Thanks @mwituni for your input :)

    One of the reasons I started this forum before release was to get input, so I really appreciate the time you guys have taken to share your thoughts. There are a number of cool ideas in here I hadn't thought of :)

    One thing I will clarify is that the system is fully seed driven and deterministic, so that should help with multi player games. The spawner has a seed, and uses the .net random class rather than the Unity random classes so that we can get replicable sequences, and can also run multiple instances without interference caused by sequencing issues.

    Another pet peeve of mine is games that only release on one platform.The resource manager supports assigning mobile and desktop assets. There is no reason the spawner couldn't be extended to spawn mobile versions on mobile, and desktop versions on desktop. This could be applied to randomly placed buildings as well.

    Keep the ideas and the names coming :)
     
  11. BackwoodsGaming

    BackwoodsGaming

    Joined:
    Jan 2, 2014
    Posts:
    2,229
    One problem I always run into is adding additional terrain tiles after I've already been working on my terrain. I've tried stitching programs after generating the additional tiles, but none of them seem to work well. I'm wondering if your generator might have or eventually have the ability for me to add a new terrain tile to my scene, specify neighbors, and have it include stitching to match borders with neighbors as part of the generation.
     
    CaptainMurphy and AdamGoodrich like this.
  12. AdamGoodrich

    AdamGoodrich

    Joined:
    Feb 12, 2013
    Posts:
    3,780
    I have a stitch method in the utilities component. It works out where each tile is compared to the others and calls the unity functions that join them together. It doesn't yet handle fixing pixel imperfections, but I will add that as well as it is useful, and will make any imperfections i create when i add multi terrain support easy to fix.
     
  13. CaptainMurphy

    CaptainMurphy

    Joined:
    Jul 15, 2014
    Posts:
    746
    Our game is designed for the single player experience and also uses the Tenkoku/Suimono combo. I would love to see this system working on our setup since we are developing an open world that covers the ENTIRE Caribbean. What we were looking into recently was a voxel system that would let us create heightmap tiles and matching biome and splatmap tiles (RGBA textures) that would define things like the terrain and vegetation that would grow there. The trick we want is to be able to select the scale at the game start time so that a player could choose 1/10 scale or 1/4 scale when they start their game. From that point on their world would render at that scale using the same texture tile data regardless of the output scale. Being able to define some parameters for the scaling would be great, like the terrain max height/width scale so that a larger scale would not necessarily make the mountains taller, or smaller scales may not make them shorter.

    As of now we are fighting with TC and WC for this and are going crazy.
     
  14. AdamGoodrich

    AdamGoodrich

    Joined:
    Feb 12, 2013
    Posts:
    3,780
    Another thing, the system is written in C# and the code will be provided, so you will be able to extend it if you want to. I have tried to minimise dependencies as much as I can. Also, some of the core classes are completely unity independent, so it lends itself to real time / multi threaded use. If there is demand I will make a real time generation capability in a later release.
     
    CaptainMurphy likes this.
  15. AdamGoodrich

    AdamGoodrich

    Joined:
    Feb 12, 2013
    Posts:
    3,780
    Sounds interesting. Need to digest this :)
     
  16. BackwoodsGaming

    BackwoodsGaming

    Joined:
    Jan 2, 2014
    Posts:
    2,229
    Personally I don't really need the real time for players, but I could definitely see where at least the editing functions would be great for development. I can't count the number of times I've been running through a terrain only to discover something. Would be nice to be able to edit it right there when testing. Of course it might be nice to have a subset of scripts available to use for things like adding basements/houses to have the script level off a spot in the player's plot or scoop out their basement and then one of Cobus' buildings from his new Exteriors Kit could be dropped in.
     
  17. AdamGoodrich

    AdamGoodrich

    Joined:
    Feb 12, 2013
    Posts:
    3,780
    I have't gotten it working yet, but I have the concept of both an additive and subtractive approach to interacting with the terrain, and I mentioned to Cobus at some point that I would like to create a set of templates for his buildings so that you could do exactly that :)

    Imagine.. - you peruse your gorgeous environment and find a nice spot by the sea... so now you want to build there but the slopes don't work - its tedious and time consuming to modify and then texture the terrain to do this and even then you aren't happy because the terrain tools that Unity provides are just not that precise... so you choose the scoop tool, select the template that best matches the basement you are building, choose the texture you want, and then just click to apply them to the terrain at the location and rotation you want. A moment later you have pixel perfect and properly textured base right where you need it to be... now it's time to get working on that awesome mansion overlooking the sea... and some time later the view from the porch is ... breathtaking :)



    The capability at least will be there, and with luck Cobus will let me create some templates for his Exteriors pack :)
     
    BackwoodsGaming, John-G and S4G4N like this.
  18. S4G4N

    S4G4N

    Joined:
    Mar 13, 2013
    Posts:
    3,213
    @AdamGoodrich

    That is a very good idea, if you could have basic shape templates, then punching those required cavities into the terrain will be a breeze.

    Also wandering, would your tools be able to subtract and out/export for instance .fbx or .obj meshes from terrains created. I think specifically Mobile and also other users would love to be able to generate the desired landscape or small scene, export the mesh and then unwrap and paint before using in there Mobile/Desktop project.

    Cheers
    Cobus
     
    AdamGoodrich likes this.
  19. AdamGoodrich

    AdamGoodrich

    Joined:
    Feb 12, 2013
    Posts:
    3,780
    Added to the road map :)
     
  20. BackwoodsGaming

    BackwoodsGaming

    Joined:
    Jan 2, 2014
    Posts:
    2,229
    That sounds EXACTLY like what I was looking for, both for foundations for houses without basements and basements for houses with them. Here is what I'm hoping to get eventually.. Within my world I will have designated areas with land claims for homesteads. I want them to be untouched wilderness areas and be various sizes, anything from a small house to a farm. Being in a wilderness area, they might have trees, rocks, undergrowth that need cleared where the house would be built, which what you described sounds perfect for. Players could also harvest the trees and stone to be applied as building materials to build their home. I'm thinking something similar could be done for fields/gardens to plant, which once set as a garden would use that growth script after planted.. I'm visualizing speedtree corn and wheat growing already! lol

    At any rate, I think I'm going beyond the scope of what you are building a bit. But if it includes the basic terrain capabilities to clear and place foundations, or even select an area and lay down textures for a field for growing crops or pasture area, that would totally rock!
     
    AdamGoodrich likes this.
  21. Whippets

    Whippets

    Joined:
    Feb 28, 2013
    Posts:
    1,775
    Have a set of shapes that can be sunk into the terrain. The same code might also work for streams and rivers with a spline tool to made the winding shape.
     
    AdamGoodrich likes this.
  22. AdamGoodrich

    AdamGoodrich

    Joined:
    Feb 12, 2013
    Posts:
    3,780
    The latest incarnation of this asset started with me doing something similar in an additive way to generate mountains from ridges. Will add this idea to the roadmap. Its a variation on what i have already done with beaches :)
     
    Whippets likes this.
  23. Whippets

    Whippets

    Joined:
    Feb 28, 2013
    Posts:
    1,775
    Can't wait to try it out :)
     
    AdamGoodrich likes this.
  24. Wylaryzel

    Wylaryzel

    Joined:
    Sep 13, 2013
    Posts:
    126
    Btw - if you need a beta-tester, call me in... as powerful as TC is, it seems that I have not enough control over it and is whether suitable to completly randomize everything or utilize fully configurated setups defined outside (e.g. WM)... looking forward to your tool :)

    BR
    Wyl
     
    AdamGoodrich likes this.
  25. AdamGoodrich

    AdamGoodrich

    Joined:
    Feb 12, 2013
    Posts:
    3,780
    I reintegrated the beach tool today.. But then forgot time n had to rush out the door to dinner with friends. Still needs work, and probably needs to be shown in terms of before n after, but screen shots tomorrow :)
     
    John-G, S4G4N and BackwoodsGaming like this.
  26. Archania

    Archania

    Joined:
    Aug 27, 2010
    Posts:
    1,662
  27. Iguanapl

    Iguanapl

    Joined:
    Mar 11, 2014
    Posts:
    23
    AdamGoodrich likes this.
  28. CastleIsGreat

    CastleIsGreat

    Joined:
    Nov 10, 2014
    Posts:
    176
    this looks amazing.
     
    AdamGoodrich likes this.
  29. AdamGoodrich

    AdamGoodrich

    Joined:
    Feb 12, 2013
    Posts:
    3,780
    Here you go ... its the same location as the other images, but this time I ran a couple of iterations of the beach tool. The areas of beach that run into the bay are the remnants of previously raised areas of land.

    I run a localised smoothing algorithm when making a beach and its a bit heavy because imho the beach is too flat and I would like the beach to have more definition... also modified Suimono settings to be more bay like :)

    The thing I love about powerful tools like these is that you can play, and in quite a short amount of time you can completely change the feel of an environment.

    Note - all of the images I post here should be treated as WIP images. They don't represent completed work, and are generally put together quickly to show off different aspects of what I happen to be working on.

     
    Baldinoboy, mwituni, Archania and 3 others like this.
  30. Whippets

    Whippets

    Joined:
    Feb 28, 2013
    Posts:
    1,775
    Damn if that's put together quickly, I hate to think what's possible wit hsome time and effort. Once again, that's incredible. I need it now.
     
    AdamGoodrich likes this.
  31. Whippets

    Whippets

    Joined:
    Feb 28, 2013
    Posts:
    1,775
    How are you getting the density of grass/foliage and showing at such a distance without grinding to an fps halt?
     
    BackwoodsGaming likes this.
  32. BackwoodsGaming

    BackwoodsGaming

    Joined:
    Jan 2, 2014
    Posts:
    2,229
    Damn! I wish my WIP terrains looked even a tenth that good.. lol Love the beaches! I WANT this tool! I'm about to go insane trying to get my terrains the way I want them to look in TC... lol
     
    Wylaryzel and AdamGoodrich like this.
  33. Whippets

    Whippets

    Joined:
    Feb 28, 2013
    Posts:
    1,775
    Yep, I'm just not bothering with terrains until this is out. Waste of time.
     
    Teila and AdamGoodrich like this.
  34. BackwoodsGaming

    BackwoodsGaming

    Joined:
    Jan 2, 2014
    Posts:
    2,229
    I only started messing around with them to play with Cobus' Exterior Kit.. Hopefully I'll be able to fix them with this when it comes out and do all my production terrain with it... heheh
     
    AdamGoodrich likes this.
  35. AdamGoodrich

    AdamGoodrich

    Joined:
    Feb 12, 2013
    Posts:
    3,780
    I haven't been running any perf stat's on this as I am more focused on providing the tools to create the environment at the moment. Am making sure that it generates vanilla Unity so that you can do your own thing with regard to optimization. The general rule of thumb is that the richer you make it, the more it takes to render it.

    I have noticed that lots of things seem to contribute to performance, and when i was building the last largish terrain I tended to try something - observe - try something else - observe. Don't consider myself an expert here and would love to know what happens under the hood in Unity so that we could play to its strengths.

    Here are a couple of things I noticed :
    * Running a smooth on the terrain drops the tris significantly with a small performance boost
    * Changing the terrain detail distances makes a big difference - but I love lush grass and my system can handle it
    * The more trees the bigger the hit - but SpeedTree just looks so good :)

    I find that Unity 5 generally seems to handle rich environments pretty well.

    The Spawner will have the ability to apply the trees into the Unity terrain, or to treat them as separate game objects. It would be interesting later when treating them as game objects to explore culling them completely based on distance from the player. I don't know what the impact would be but will try that at some point.

    Another technique I have done that seems to work well, and I may include this, is to dynamically control the detail and quality settings based on frame-rate. I have a script that monitors this and then makes the changes. It is quite a simple but powerful technique.

    Later on I will explore performance optimization and put together a document on what seems to work for me.
     
    BackwoodsGaming likes this.
  36. Whippets

    Whippets

    Joined:
    Feb 28, 2013
    Posts:
    1,775
    All good :)

    Have you got a rough timescale for this tool? weeks, months, summer, winter?
     
  37. AdamGoodrich

    AdamGoodrich

    Joined:
    Feb 12, 2013
    Posts:
    3,780
    Hoping to have first version out in the next month.
     
    Teila and BackwoodsGaming like this.
  38. Whippets

    Whippets

    Joined:
    Feb 28, 2013
    Posts:
    1,775
    Cracking. Can't wait :)
     
    AdamGoodrich likes this.
  39. BackwoodsGaming

    BackwoodsGaming

    Joined:
    Jan 2, 2014
    Posts:
    2,229
    That is what he said earlier.. But I'm assuming if he is able to release that soon, it will be a release like Cobus did with Exteriors Kit where he plans to add other stuff based off of a roadmap.. At least from the sounds of it he still has a lot to do that sounds like it would take more than a month. Personally, I think it is looking pretty good as is if I'm able to create terrains like the screenshots with it! lol
     
    AdamGoodrich likes this.
  40. BackwoodsGaming

    BackwoodsGaming

    Joined:
    Jan 2, 2014
    Posts:
    2,229
    Heheh.. I'm such a wordy poster... So.. Yeah.. What Adam said... heheh
     
  41. AdamGoodrich

    AdamGoodrich

    Joined:
    Feb 12, 2013
    Posts:
    3,780
    Hehe - yeah - I will release earlier, but only when it is polished enough, and will have a roadmap that will get built over time.

    It's funny how we get caught up in the details - I have a spawner that is quite powerful - and a texturer that is quite powerful but has some features that the spawner doesn't have - and then I have my object DNA system that adds another level of capability again.

    I am very conscious of keeping it simple and coherent - and have been cracking my head on this from a design perspective over the past few days. I am waiting for the flash of inspiration that will allow me to bring it all together in a way that makes sense.
     
  42. John-G

    John-G

    Joined:
    Mar 21, 2013
    Posts:
    1,122
    Terrain composser got some serious competition on its hands. I'm really excited to get my hands on this when it's released :D:D:D
     
    Teila and AdamGoodrich like this.
  43. BackwoodsGaming

    BackwoodsGaming

    Joined:
    Jan 2, 2014
    Posts:
    2,229
    Sounds like you easily have several different tools that could be produced separately but then also offered in a bundle.. Similar to how Nathan did with Sectr or the Probuilder guys did with their tools.. Definitely agree with John.. Very looking forward to it being released.. Now just hoping I have money to grab it when it is! rofl
     
    AdamGoodrich likes this.
  44. AdamGoodrich

    AdamGoodrich

    Joined:
    Feb 12, 2013
    Posts:
    3,780
    You might also consider this as a replacement for a number of tools that you currently have to purchase separately :)

    I think your idea is a good one though. That way people can choose to purchase just the components they are interested in.
     
    Teila and BackwoodsGaming like this.
  45. Wylaryzel

    Wylaryzel

    Joined:
    Sep 13, 2013
    Posts:
    126
    Hi Adam,

    I'm throwing money on the screen with each new pic but it falls back on my desktop :)

    Anyway, may some few suggestions:

    a) I like the idea of different 'agents' for doing the job. As everything is more or less based on the basic unity terrain it should be easy to swap in and out tools for particular cases. Like I only want to use the spawner component based on a already made terrain -> use only the spawner. In that way they should be more independed and you can clearly define a start and end scope of the tool. Additionally, it may keep the agents simpler.

    b) I really like the idea from shawn to have this modular approach - also with regards to the asset store. You could sell the individual 'agents' - if somebody only needs a specific one, or offer a bundle . The bundle could also raise in prices as you add additional 'agents' to it but giving a discount compared to the individual 'agents'. I think it would be a win-win situation for the users as well as yourself?

    Coming back to UI and handling:
    c) Maybe you would consider a 'Suite' window as a home for all the 'agents'. In there, all the different agents may could be plugged into it and have a central home for modifications / etc. To be honest, I like more the approach of TC or maybe this tool: https://www.assetstore.unity3d.com/en/#!/content/14318 compared to the approach of Sectr to attach scripts to differnet GO. But I don't know how complicated it would be.

    d) If possible, may consider a sequencer module. What I mean by that would be that we as a user could control when which 'agent' should do his calculation. To give an example:
    - First agent would be the rough landscape generator (may again splitted into different ones like generating mountains / beaches, etc)
    - Next would be the agent responsible to generate the splatmap
    - Next kick in the spawner agent for object placement (e.g. rocks, houses)
    - Next kick in the spawner agent for foliage (trees, gras)
    - Next again maybe again the agent for landscape generator for flatten terrain around objects if needed (depends a bit on the options that the different agents could have).
    In that way we have control when and what should happen. For example in TC I have to visit the heightmap and splatmap controller twice -> First for generating the terrain and again after placing the objects to flatten the terrain.
    Hope this makes sense.

    C) + D) maybe could be combined into a node-based approach like shader-forge or other node-based systems (came to this idea while screen the asset store and found this : https://www.assetstore.unity3d.com/en/#!/content/13418, looking at the 4th pic in the asset store)

    Was a idea in the morning - may some of the sounds to complicated or are really out of scope. But you know, as long as the tool is not available I can dream :)

    Thx in advance
     
    AdamGoodrich likes this.
  46. TechnicalArtist

    TechnicalArtist

    Joined:
    Jul 9, 2012
    Posts:
    736
    Good work,its work on mobile device?
     
  47. AdamGoodrich

    AdamGoodrich

    Joined:
    Feb 12, 2013
    Posts:
    3,780
    As much as any Unity terrain works on a mobile device. Yes :)
     
  48. TechnicalArtist

    TechnicalArtist

    Joined:
    Jul 9, 2012
    Posts:
    736
    Thanks for reply.best of luck
     
  49. Teila

    Teila

    Joined:
    Jan 13, 2013
    Posts:
    6,932
    I see one problem with this. Generating a terrain is nice but the terrain is useless until you have a proper splatmap generator, spawner for rocks/trees/etc., or a way to modify the terrain around objects. So what will happen is that someone will buy the terrain generator, and then sit for a month for the splatmap stuff, and then another month for the spawners. Or, more than likely, they will use some other option, like TC or Landscaper or several other terrain "decorating" options out there. Once your terrain is done, the last thing you want to do is buy more assets and redo the job you spent so much time on...especially with TC since it is such a bear to use. :)

    In my opinion, better to release the entire package and then break it up, like Cobus did. That way, we can buy the finished product and complete our projects. Or if we only need part of the bundle, we can buy that separately.
     
    AdamGoodrich likes this.
  50. AdamGoodrich

    AdamGoodrich

    Joined:
    Feb 12, 2013
    Posts:
    3,780
    I will be waiting until we have at least a minimally fully featured product :)

    In other cool news, yesterday I came up with a design approach to the Spawner that I was pretty happy with. By default it will come with a useful and working set of placement tools.

    However the placement logic is in c# and derives from a base class, so it can be extended and modified to suit your own tastes. For those of you that are coders this is very powerful because you can leverage the underlying framework and still easily customise the way it works as long as you conform to a simple api.

    Any prefab that has a placement script added to it can then be managed by the Spawner.

    The goal is to use a similar approach to anything that can be placed... from mountains to textures to trees and grass to things like buildings, beaches, or even composites e.g. placing things that have dependencies in one go such as buildings on top of bases, grass and textures around trees etc.

    Also Spawners can be spherical or square, meaning that you can place them nicely side by side if you want to.

    They are deterministic, so that you can Reset them with a seed and they will play the same way every time.

    And finally, and most importantly they enable localisation, so they will support the bigger goal of enabling interesting, rich and extensible biome support :)
     
    Last edited: May 31, 2015
    chiapet1021, John-G, Archania and 2 others like this.