Blender Foundation to Create Open Movie, Open Game

Eloquence writes "The Blender Foundation, which maintains the open source 3D tool Blender, has announced two new projects, codenamed Peach and Apricot. Project Peach will be a new open source movie, following in the footsteps of last year's Elephants Dream project (which was initially codenamed Orange). Apricot, on the other hand, will use Blender in conjunction with open source 3D framework Crystal Space to create an open game, thereby showcasing both technologies."

Comments

Not the Blender game engine?

[Animats]

It's too bad they're not using Blender's own game engine. Blender has an integrated 3D animation system and game engine.

The trouble with the Blender game engine is that it doesn't scale well. The Blender game engine can be used "without programming", but what that really means is that you have to draw connection diagrams with hundreds or thousands of connections. Then you get to debug the wiring. For a non-trivial game, it's painfully difficult to debug.

It's an occasional fantasy of programmers that wiring visually functional blocks together is easier than programming. Engineers who wire up real hardware know better. That's why we have VHDL.

Fortunately, you can extend the Blender game engine in Python. Unfortunately, it's CPython, which is 60x slower than C. This isn't a hit you can afford in most games.

Re:name ?

[Anonymous Coward]

Actually, it is an Open Source movie. The contents, creation files, final products, etc. are all Creative Commons (or maybe Blender Open Content, not exactly sure) licensed, meaning you can use the actual production files for whatever you like, including a whole new movie.

What's not open about that?

Great!

[saibot834]

I liked Elephants Dream very much. BTW: You can download it here as Avi or Quicktime.

blender advancement

[Speare]

If Peach and Apricot follow the progress made in Orange, you will be sure to see a lot of very useful, general purpose improvements in Blender for everyone. The need for specific features in Orange really focused the developers (some of which were Orange team, some of which were in the general development community) on solving specific creative problems. It's the difference between "scratch an idle itch" and "remove a troublesome splinter."

CS Project Manager Perspective

[Jorrit]

Hi all,

I'm the project manager of Crystal Space. I'm of course very glad that Blender and us will do this project. It is an amazing opportunity to enhance both Blender and Crystal Space to actually make this possible. The plan is to make a really good looking game that can compete (graphics wise) with commercial games. Crystal Space can do a lot already but in some areas we still need some more work (more specifically things related to render2texture like HDR and others). Also for a game like this we need to work on a very good animation system. Having a project like this is of course the best motivation possible.

Another important goal of this project is documentation. As this game is fully open all sources (both code and art) will be made available. And we also plan to release a DVD with a full documentary on the entire creation process. Basically everything will be available (we would release the left-overs of the meals of the participants if we could :-)

And of course the idea is to make a nice and playable game. Six months is not that long but it also doesn't have to stop there. The nice thing about an open game is that it can be extended for a long time to come. Also the game logic will serve as a nice starting point for a new game based on a different setting. So lots of possibilities here.

Needless to say I'm very excited about this project. I think it will be a great thing. Both for us (Crystal Space) as for the Open Source community. If we succeed we will have a commercial quality game but 100% free and open!

Greetings,

Re:Why not Ogre instead of Crystal Space?

[Jorrit]

Ogre would have been a perfectly valid choice as well. It is just that we have closer ties with Blender at the moment. We have an excellent blender2crystal project and our connections with Blender developers are very good.

Greetings,

Re:This will be interesting

[bishiraver]

Doubtful. Last I looked, crystal space was only just achieving what mainstream graphics engines were about 3 years ago. People who are interested in eyecandy will gloss over it, while a minor group of linux users who are gamers as well as sysadmins will cheer and cheer and cheer and.. nobody important will care.

Not to mention, somehow I doubt a game can be crafted with an open-source development model. A game needs some very strong creative forces behind it to keep it cohesive - if a typical OSS app can't even keep their user interface cohesive [ever look at your typical gnome or kde app's config screen?], something tells me it will be difficult for them to keep their: storyline cohesive, user interface cohesive (like I said, most simple desktop apps have a problem with this), game mechanics cohesive and logical, art direction and style cohesive, etc etc etc. Unless it's just TuxRacer 2.0 (maybe make a version named TuxRaver as a DDR clone?)

Considering the typical absurd workflow logic that goes into most OSS apps I've used (beyond your bread-and-water openoffice, etc)...

You will start out as a penguin in the north pole, whereby you will get a letter from your family in new mexico telling you to come quick because your cousin has died during the construction of the great pyramids. You arrive in Sicily only to find that Siberia has been taken over by mutant cyborgs all named 'Bill Gates.' You must travel to antarctica and battle the chair-throwing monster Ballmer in order to gain the GPL of truth, increasing your hit points by 6.02x10^23 while slowing you to a barely managable crawl. Desperately you make your way to florida, where you successfully battle jack thompson in court and gain the Writ of Winnitude. You combine this with your GPL of Destiny (the one you found during the side quest back in new mexico involving the trout and the cliff wren - you did that, didn't you? it's not in any of the docs anywhere, but it's so obvious! - it's NOT the Gpl of Truth - that item is useless) and are magically transported to a land of rainbows and penguins and free software for everyone. Unfortunately, this paradise is shattered by an Ax of Servitude hewn from the yggsadril-apple. It's shiny, but it works good and is simple to use...

Right about here, most people get bored developing the game and it slowly dies, forgotten in the depths of sourceforge. A few people stop by every couple of months and say, "hey.. there's this bug.. is anyone still working on this?" A few people wax nostalgic about the groundbreaking progress they made on the pixel shader they used to put a shine in the main penguin character's eye (but is not used anywhere else). The controls were odd (you used t to go forward, z to turn left, i to turn right, and clicking your mousebuttons realigned your characters eyes with the directions he was looking) and most people who downloaded it wished pain and suffering on the user interface developers, who decided to go with a "Leopard-meets-vista-meets-diabloII motif."

Note: I love open source software, but really.. I can't see it working in game development :) I could see a group of people taking some open source solutions that are released under the LGPL and developing an indie commercial app with it, but you need a solid and non-transitive team to build a good game.

Re:This will be interesting

[Jorrit]

Don't forget that one of the goals of this project is to expand Crystal Space to do what we want to achieve.

As to Open Source not working. We are going to do this with a solid team that will remain for six months in Amsterdam (Blender studios). So while the end result will be Open Source the way to develop it is like any other commercial game (only with less money and a bit more time constraints).

Greetings,