When will BGE be able to handle nature scene

I think if BGE can create good-looking trees, terrian and water, it will be more popular.
I hope it’s API can be updated, which means it’s srong enough to create games for next-generation.

The BGE will never be able to create such things. You need to create them. That is the whole point.

But I guess you are more referring to methods that makes it easier to show them in a game environment (e.g. LOD, data streaming, optimized particle systems).

Yes.In fact, I think BGE can support many features like SSAO with scripts and add-ons, but it’s lack of efficiency. That’s a trouble.

It already can. Just very few people have the staying power to make a nature scene. Because, well, there’s a lot in one.

If you want proof, go youtube ‘blender game engine nature scene’
Heres a selection from the first page:

That’s true. But it still takes time to develop before it’s performance become nature

I think if BGE can create

BGE doesn’t create anything. It just plays whatever you create.

good-looking trees, terrian and water

all of that is currently possible with BGE, you just need to create it but that will take effort always. EDIT: Nice post, sdfgeoff :slight_smile:

I hope it’s API can be updated

You mentioning API here makes no sense to me and I suspect you aren’t exactly sure what you are talking about. API is only the specification for what are the parts of the engine and how they interact.

which means it’s srong enough to create games for next-generation.

Thus why it doesn’t mean that. Engine documentation does effect how well it can be used but engine processing speed and features are the thing that translates to “strength” some next gen graphics might require.

All I can say the current gen novice developers have totally unrealistic expectations of game development. They set their standards according to some multi-billion dollars game that was developed for years by hundreds of people. They require to be able to create such level of graphics and gameplay on their own by drag-and-drop in no time at all. Not even that will suffice, they want to create “next-gen” games that look even better than any of the commercial games.

There is no such game development tool, even Unity/UDK can never be as fast and efficient as a dedicated engine that was specifically designed to do one specific thing it’s supposed to.

Novice developers going for bleeding edge graphics first is completely backwards anyway. If you ask the (other) professionals they will tell you the best way to start game development is by remaking some of the classics and starting your own projects small. If you’re starting by trying to make your dream game that will never happen because you don’t even know the basics.

Isn’t it amazing that God was able to render all of creation in only 6 days? LOL He must hell of a badass renderfarm.

Please do not bash on naive expectations to much. We discussed that several times before.

Game development is like move making or magic illusions -> an illusions.

It looks like something real, it behaves like something real … but it is not.

I can remember as I saw the magicians as a little boy … woha they can repair a shredded newspaper or even burned money. They can find my secret card in a mixed card deck. I was amazed and I wanted to do this too.

Now I know a lot of the tricks they do, but I never developed the skills to present it very well. The reason was mostly you need to train, train, train and train even more. If you do not train enough you will not amaze the audience and you will destroy the illusion.

I can see some familiarity to game development. You can easily create a game with the tools available. But if it is poorly designed it will not amaze your audience. The chances to fail are higher on large illusions then on small ones.
The major difference to magicians is … game developers do not live from the secrets of their illusions we like to share them here(at least not we lonely wolves ;)).

Back to the topic. Realistic looking nature requires a lot of skills and even more work. And yes it can benefit from build-in support to ;).

As far as I know the time was manipulated :wink: (magic).

Please do not bash on naive expectations to much. We discussed that several times before.

I always write as kindly as I can. But that kind of expectations need to be corrected. Bashing BGE for something that is about the user and not the software is just wrong. The topic alone is very harmful in itself and I don’t want anybody strolling by thinking BGE couldn’t somehow “handle nature scenes” already.

No intentions besides setting that straight. Cheers :slight_smile:

Why don’t we gather up all the recent relevant threads and create a sticky titled: Why Can’t the BGE do ABC like XYZ?

It won’t eliminate all these types of threads but should help reduce them.

All video game creation is a compromise. No engine on earth can handle millions of trees with 1000’s of faces each. If you look at the most popular games like Skyrim and the like you will see that their trees are just hexagons with a few png planes for leaves. You can make a wonderfully detailed scene but who’s going to enjoy it at 1 fps? Google URU. It’s an older game but still the artwork is stunning imo. Instead of ultra real they go with a style that is so perfectly uniform throughout the game that you forget it doesn’t look real. Stay consistent in your style. Refrain from making anything more detailed than anything else. It breaks the illusion and pulls ppl out of the experience; reminds them that its all fake.

This is very true. 10 points from me ;).

This isn’t what is needed; a viral tech demo is needed. Why do people say, "Game Engine [X] from [X] game is ‘so much better’ than Game Engine [Y] used in game [Y]? "
Because it has examples of how great it is (in most cases the actual game developed from it.). CryEngine 3, for example is glorified over most game engines, including unreal engine 4, even though unreal engine 4 is capable of much more than CryEngine 3, it hasn’t had a game developed in it yet that has become famous… CryEngine, on the other hand, has released many AAA titles.

This is why Blender invests money into creating short open movies. It is a tech demo showing that Blender is capable and shows off what it can do, which therefore increases the popularity of Blender.

Basically, the day a developer using Blender sells a top 100 game is the day it becomes recognized for what it can do.

I may be wrong, but I’m under the impression it’s actually easier for BGE to handle nature, outdoor scene due to the lighting with only 1-2 lights for the whole scene, contrary to the interior scenes where much more lamps are needed to achieve convincing look without baking/vertex coloring and other tricks.

1 hemi lamp in the whole scene and 1area lamp in each room would make perfect indoor lighting.

About being able to create terrain, trees and water, this should be easier than ever because this mainly a 3d modelling and texturing program with a built in game engine. if you want it pre-made made then just search in the resource forum or use applications that create it.

Not to mention that in one day He’ll wipe out this universe and start anew.

Isn’t it amazing that God was able to render all of creation in only 6 days? LOL He must hell of a badass renderfarm.

I was under the impression he modeled and textured it in six days, but that it’s still rendering a frame every 5.39106 × 10−44 s It must be a real great renderfarm to do that. Especially in such uncanny realism (but then, how would we know if it was unrealistic).
Also, reminds me of this comic

Anyway, back on topic:

This isn’t what is needed; a viral tech demo is needed. Why do people say, "Game Engine [X] from [X] game is ‘so much better’ than Game Engine [Y] used in game [Y]? "
Because it has examples of how great it is (in most cases the actual game developed from it.). CryEngine 3, for example is glorified over most game engines, including unreal engine 4, even though unreal engine 4 is capable of much more than CryEngine 3, it hasn’t had a game developed in it yet that has become famous… CryEngine, on the other hand, has released many AAA titles.

The problem is that no-one has yet had the staying power to make such an awesome game. Krum has that chance, if it ever get’s completed, as does torakunsama’s project (COS).
But I don’t think they will actually help unfortunately.

For some reason, people assume open source/free programs are in some way inferior. I think someone not only completing a graphically intensive game, but also putting it on the market is necessary.

I also recommend taking a look at the game Rage. I don’t know much about it, but I know they used a totally different approach to building a highly detailed scene. I believe they used prerendered scenes and a sort of environment map to display them. Here I’ll post a video explaining it.

…Well I can’t seem to find the video I saw a couple years ago. It explained how all the graphics were prerendered and not rendered on the spot just stored as images or something. This also reminds me of games like Riven which was an old point and click adventure game in the Myst series. (Can you tell I like Myst?) =) Well that game is nearly 20 years old now and still the gfx are stunning. They ofc accomplished this by having prerendered images for each step. This is back in the 90’s before the days of fallout and games that allowed you to free roam highly graphic worlds. Anyway, as I understood it Rage was the evolution of ideas like those seen in Riven.

If you guys are interested in game design I highly recommend downloading URU from http://mystonline.com/en/play/ . It is 100% free though in 03 it was $50. It is the best lesson in gave art I could direct anyone too. Great graphics and wonderful immersion.

Could God create a scene so complex that he himself could not render it?
Could God create a game so difficult that he himself could not finish it?

On a more serious note…:stuck_out_tongue:

The BGE does not come with batteries included compared to other game engines. As a result, the BGE is perceived to be inadequate. It has the capabilities of any other modern game engine short of cutting-edge and experimental features. This means you must create all the tools and assets yourself. This aspect of the BGE is slowly growing such as the various templates, libraries and shaders. So really what the BGE needs is more organized resources, and essentially somebody needs to put it all together and create a flagship game that the community can get behind on, a la Half-life modding style.

Rather than teach somebody how to make a game from nothing brick by brick, it’s sometimes easier to disassemble and reassemble something to understand how it works. Of course, both techniques are needed. Blender the modeling tool provides both techniques in the form of tutorials and sharing blend files, etc. Not to mention, modeling and animating techniques and skills can be transferable between 3D creation programs compared to between game engines where the API and tools are completely different. However, the BGE lacks more in the latter where there are no polished games for others to study. What comes to mind is Dead Cyborg which is closed-source, and maybe Sintel which is still alpha. There are also others showing promise.

In the case of the BGE and game creation, the latter works better because it provides an adaptable learning curve. For example, an easy introduction would be where a person can change a single variable that controls the speed of an object, or for the highly ambitious, a person can replace the entire character and give them more functions. Personally, this is how I learned to understand game creation and engines.

Naturally, the BF focuses more on Blender the modeling tool since that is its original purpose and most of the professional users are in the film industry. However, the BGE segment has grown and continues to grow (hmm, we need to compile statistics on this). Ton’s post before nicely shook up the community :p.

Yo Frankie was more of a featurette, like those included when you buy movies, than a complete game. So yeah, it’d be cool if the BF sponsored an open-source game, multiplayer would be a plus or an extension by the community? The creation of a game would also benefit and showcase Blender the modeling tool. Seeing how Valve has been paying attention to Blender and heavily investing in open-source, now seems like a good opportunity.

@Jackii - That’s exactly what I mean - 1 hemi lamp and 9 area lamp produce around 65ms rasterizer and 15fps, with only one solid material. That’s tested on 2 PC’s which can handle with no problem modern games. If you know some method how I can produce playable framerate with 10 lamps, please do share.