Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Simbuilder5]

Could Speedtree be used to cover the land in Outerra?

[ChookWantan]

That looks pretty damn awesome...

[BenDragon]

To be honest, I've seen this stuff in trainz... and it lagged the game by 10-20% depending on the type of computer you're running.

If speedtrees are going to get implemented, they should NOT affect the performance of how fast the engine runs.

A game is not just about looks, it's about how much lag it would produce too.

Please also keep in mind that this is a world engine which doesn't need a pre-created map of a tiny section of world. So the performance demands of it could already be quite high.

I really do advise caution when implementing these trees into the engine.

[ZeosPantera]

I don't think outerra needs to worry about trees...

[Steve.Wilson]

OMG.  Feels like a scene out of Avatar.  Nice video. Thanks, Zeos. No lumber worries here.  And I see a terrific spot for a cabin.... 

[giucam]

Honestly, i don't like so much the trees in that screenshot. Seeing them from above like in the video is ok but in the screenshot i can clearly see they are not real, unlike the other features of outerra.  Now, i know they're a WIP so i don't cry for that but imho the problem is that they come out of the ground without a minimal base. They seem to be light poles planted in the ground.

[Ozybolairy]

How new is this video and screenshot? I notice there is no grass on the floor which would improve the realism factor that guicam mentioned. The models could be improved at the base of the tree to show roots rather than going stright into the ground.

[cameni]

More than a year old, from Aug 2, 2011. We delayed the work on 3D trees to get the demo out.
Yes, 3D grass alone would help greatly with the problem giucam pointed to, but of course the tree generator will also produce the upper roots when finished.

As for the SpeedTree, it's not fast enough to be used in a huge world like ours. Even though it's procedural, it's not particularly fast because it was not made to generate models optimal for modern 3D hardware as well.

[Ozybolairy]

Which leads me to my next question... has work progressed on 3d trees since or is the current focus the importer?

Cheers,

[cameni]

Work on the 3D trees hasn't been restored yet, planned after the importer.

[Midviki]

Work on the 3D trees hasn't been restored yet, planned after the importer.

And my opinion is that it should not be restored.As a developer I would probably want my own kind of trees, cause I would probably want like 200 different species of trees.

+ that I have in mind to make them grow in real time(or faster probably) trough the animation and a *Brush* tool would be a tremendous addition.

And to prove my point... you have a island, very big island.It will take forever to place the tree's that you would want 1 by 1.And you would probably want palm trees near the beach side of the island.And more diversified trees when going up to the other parts.So a brush would be nice just to paint the zone where you want the trees spawned.And I think this would work with the terrain texture as well.Maybe rivers, lakes.Making the tides of the seas stronger and more aggressive on some very short edged shores.And more peaceful and graceful on the gulf zones.A brush would be nice. 

[ZeosPantera]

And my opinion is that it should not be restored

You high?

I can go into more detail concerning this statement but I feel Cameni can better answer the whole point of the engine and what it wishes to accomplish. Along with the future plans for the procedural tree generator.

[Juggernautz]

And my opinion is that it should not be restored.As a developer I would probably want my own kind of trees, cause I would probably want like 200 different species of trees.

+ that I have in mind to make them grow in real time(or faster probably) trough the animation and a *Brush* tool would be a tremendous addition.

And to prove my point... you have a island, very big island.It will take forever to place the tree's that you would want 1 by 1.And you would probably want palm trees near the beach side of the island.And more diversified trees when going up to the other parts.So a brush would be nice just to paint the zone where you want the trees spawned.And I think this would work with the terrain texture as well.Maybe rivers, lakes.Making the tides of the seas stronger and more aggressive on some very short edged shores.And more peaceful and graceful on the gulf zones.A brush would be nice. 

Trees can be placed in many ways. Using a procedural algorithm, foliage density maps or done via brushes/polygonal area modelling. It's not going to be a case of placing trees individually, unless you want to have a very specific unique decoration mesh.

[Midviki]

And my opinion is that it should not be restored

You high?

I was referring that it's not like an immediate issue.It can wait 2-3 yes.And if speaking of money, the issues that might matter most first are animation / collision / scripting pannel for objects / and the most important things of all is how to make Outerra sustainable on multiple servers but the same world.Cause at the physical dimension of the world.. splitting the people on different realms just because a server can't hold enough people on it(and I'm talking of maybe 100 milion people playing the same MMO in the same world).I mean look at the usual MMO's... World of wacraft , Guild Wars , Tera , Lineage... and the list goes on.They all split their people, spliting people is like not having a round planet, but more of a plain game as usuall that can't hold more than an usual game, it won't be any different or special if you can't fill it with milions in the same place.If those issues would be resolved. People could start producing content and Outerra would collect the fee.Then enlarge the staff and developers. Cameni and Pig and who else is on the team now would be C.E.O.'s and things would move faster.

[cameni]

Custom placement of trees has nothing to do with the tree generator or tree library used. We simply cannot use SpeedTree trees with the amounts this world requires, so that's why we need the generator optimized for it. You can still have a few trees as models though.

Custom placement will be done as everything else - by defining vector shapes that modify the underlying model that's supposed to generate the basic tree coverage from available data (real-world forest coverage, climatic models). The shapes can alter it by cutting or putting the trees, changing the type and other parameters etc.

Note that making a MMO work on a single world the size of Earth isn't one of the problems we are solving in any near term. Making a MMO brings a lot more problems than making other types of games, and OT still needs to address the basic needs that games/sims may have (also the ones you listed). There's a lot of games that could use OT as a world rendering engine, in singleplayer or with their own custom networking for multiplayer. OT initially won't provide its own networking as there's a wide array of game types, each having specific requirements, with their preferred networking libraries.
There's a reason why the big fish do the splitting, if it was easy it would be used already. But technically it's an orthogonal thing to the world rendering engine, it could (and perhaps should) be developed separately. Volunteers?