With the independent study I am doing of OpenGL Programming and the leadership skills, communication and dedication to the graphics world, I would like to apply myself as the project coordinator. I will not take the role of lead developer but rather stand graciously as leader and help manage and direct the peoples goals.
This would be a community project though, with a voting system so that we can vote upon how the project should take shape. I do have many proposals on how the development should remain equal, fair and to everyones pleasing. I wish to keep the modelling, animation, rendering, and gaming all in equal once the project becomes completely operational.
I will form a cabinet of some sort to recieve advice, and also take all emails of good, bad, and the occasional 'odd' blender community type mail. I am also open to the idea of sharing this position with another person and working with them as this project is about the community. So, as we begin putting all this together, I hope you all will take this post seriously as I do the position, and hope to hear from many/ALL of you!
Thankyou,
Sheridan Sellers([email protected])
go to www.rocket3d.com a blender dude who is making a modeller at the moment-- + he has done a character animation system aswell I think-- that looks like a foundation. Get in touch with him! – You should definately do the modelling and animation in one program!-- then maybe an external renderer. I can help, but not with programming.
Matt
someone should start up a temp website straight away-- not just blender – but global!
I would also be interested in helping out. I agree with the above posts for C/C++ (gcc for cross platform), and working on a modeler first.
Maybe it would also be a good idea for some of the resident artists to come up with potential screen layouts. Someone who has experience with various packages such as Lightwave, Maya, Blender could come up with a best of breed idea for the gui layout…if we are looking at coming up with something orginial.
Also, what makes sense for a storage format? Something unique or reuse a format from an existing package for compatibility?
With the independent study I am doing of OpenGL Programming and the leadership skills, communication and dedication to the graphics world, I would like to apply myself as the project coordinator.
Good!
I have done lot of project management, but not for open source community. I bet there are slightly different rules out there.
Playing with an idea of the project group roles:
overall coordinator
technical coordinator (e.g. arranging hosting for web pages)
I vote against Java. The speed is not there yet. In a recent (about 5 weeks ago) test for the company I re-coded one of our computationally expensive (logic, not math) transactions in Java and C++. The C++ version took 0.1ms on the average, Java took 50.0ms.
As for Cg for shaders, it is an interesting program but [so far] it is limited to a small number of platforms and graphics cards. My setup is supported but I bet that more that half the community would be left out if we built something that required it. (Though we’ll need some way to take advantage of this and other new GPU features as they emerge, especially in a realtime component.)
Someone mentioned 3 major components needed to be developed (in that note, as separate programs). The video sequencer should also be included (my opinion of course) and either a default renderer or an interface to popular renderers.
you would think a “lead architect of distributed programs for 18 years” would know that without having to do the experiment, much less do it himself.
Absolutey! But Java is being sold to management as the fix for all things, and we techie types have to continuously demonstrate reality. For those who have ever seen a Dilbert cartoon, we really live that life, including the pointie haired bosses (some bosses are smarter, but those don’t get promoted). Vendors have changed from selling tech to the techies and expecting us to convence the bosses that we need it, they sell possible savings to management and let them dictate direction.
As for doing it myself. I like to program. Architecting large systems pays well, but programming is where the real challenges, fun, personal rewards are. And (as I tell my boss) if I don’t practice programming, how can I design a system and expect it to be implemented? How could I chose the correct language for any particular purpose?
Finally much like an artist, there is ego involved. How many artists could sit by while some worker bee with an airbrush tried to paint their vision? “Who could do it better than I can?” “Send out for pizza, and give me that keyboard”
I don’t really like working. (I don’t think lots of 12-year-olds do.) I don’t know anything on programming. (I once tried C++, but it was too hard.) I think maybe we could change those icons (i.e. Lamp buttons is a light bulb, Materials is a red ball) so that they’re in 3D. I might be willing to do that kind of thing using the original Blender. (i.e. that light bulb? Actually looks 3-dimensional…)
An idea for improvement:
Insert sound in the Sequence Editor.
BTW: If we do make this, how much space would it take up? Would it fit on a floppy? Would you have to put it on a DVD? (Hey, Blender on DVD… not a bad idea, but most likely really complicated…) Would it be like the older version (open up 5 different sessions at once without resorting to Scenes) or the newer ones (open more than one, and one will die)? Also, what would it be called? “Blender 3.00?” “Blender: The Sequel?” Plain old “Blender?” Or, with those legal things, an entirely different name? “Freedom 3D?” “Unlimited?” Hey… “Blender Unlimited…”
Plus, if anyone knows how to do it, we should make “Blender DVD” so that you can design your own DVDs. But, as said before, it’s be really comlpicated…
Does anybody still want a Blender Game Console? (Blender Worlds… hmm… we could create a design, put in plenty of gigs of space, make it open-source…)
hi all
thought i’d contribute to the language war
i think i’ve gone on about ruby elsewhere, and i like its simplicity, and think that for prototyping its great, however there is no doubt in my mind that the whole thing will eventually need to be in C++. It’s the only OO language (if used properly) with the performance required, and most people out there will know it. (I’m aware about next big thing and Java and all this . . . . hehe, when will management learn?)
Points to look for in an implementation language
OO (maintainance and separation of code)
lots of people know it
multithreading capability
performance
platform availability
I think it would probably be a good idea if work in a higher level language than C++ were to take place (ie Ruby, Python and to a lesser extent Java) for prototyping to see if people can get their head around such a large task with a lower implementation burden. Then talk about optimizing slow bits into C/C++ could take place, eventually with the whole thing done, but in a nicely organized manner.
As an aside look at how good wings 3d is and thats in erlang! http://www.wings3d.org/ i think btw.
I didn’t see this brought up before, but instead of starting from scratch (which is hard — just designing the object and material classes and their interactions is a headache) have you considered joining K3D? They are pretty far along (no animation yet, I don’t think) in terms of modeling, and have support for external RMan renderers. It is written in C++ and looks to be pretty extensible. They have an active mailing list and are asking for help.
To get my film work done, I am trying Animation::Master. It seems incredibly feature-rich for $299 ($199 educational price), and includes things like hair, fur, cloth, dynamics, great character animation tools, and a good hybrid raytracing renderer. Its community is very helpful and active. For those of you who said “Boy, I’d pay $100-$300 for good modeling/animation tools” you might want to check this out. A possible downside is that it is spline-based, not polygon, but after seeing some incredible work done with it I am willing to give it a try.
This is a rough project. Modelling and Animation are not trivial. I will help as best as I can but don’t have a great amount of time or C++ experience (I use kylix). I can help with project specifications/design.
I agree with Green that FLTK and C++ is the way to go. Most of the special effects/animation houses that use a bunch of custom software have gone that route to port SGI software to Linux/Win and they seem quite happy with it… its fast and fairly small footprint.
I would say:
Look at all the current modellers/animation software either in development or dead for ideas and code (if GPL). Dont reinvent if you dont have to.
Keep it as simple as possible and get a result. Design a modeller/animator. Dont worry about a game engine/renderer/compositor/video editor. Biting too big a chunk is death for a new project. Those others are available.
Keep the team small and close knit at first. Committee’s are good at destroying a perfectly good idea with bloat and flame wars.
Design in the extension language (python maybe) from the beginning so that it is an integral part of the system and not a bolt-on.
Programs to look at are MindsEye, Nurbana(is this open source?), Wings, Ayam, K3D. In addition there is the Clay software that looks pretty nice as mentioned earlier in the discussion (will it be open source?).
One question that has to be asked is: Is it better to start a new project or join an existing project that has similar goals.
I’ve already started on those icons. I’ve finished 9 of them (for the “buttons” header). I’ve got a lot of work to do…
BTW: I’m not spending LOADS AND LOADS of time on these icons. They’re simple, just like the originals were, but mostly in 3D. They’re copies of the original icons (i.e. a lamp emitting light, a red ball, a purple Pac-Man-like thing, etc.), but as I said, mostly in 3D, saved as Targas. (Does anyone mind if some of the materials look flat because they’re shadeless?)
But what if we do go with those other people and help their 3D programs? I don’t know what we’ll do with these…
Hi,
i think you should set up a web page and collect the stuff you talked about (or even made already). Also you could make a list of people who joined your project and what they can do and want to do.
The idea sounds great, makin’ a new programm but i think it will take a long time (but i’m waiting ). Keep it up!
I have finished those icons for the Buttons header! (However, they are in 3 Blender files for various reasons, one of them–the Armatures–wasn’t made because the Blender version I used didn’t have it, and they’re relatively simple.) I’m going to take a break for today.
And, yes, I know, I wasn’t even assigned the job, but oh well, you’ll need them.
PS: It’s kind of fun creating the icons, so could I not program?
PSS: I made them at 80x80 pixels each. This might be a little too big, but the models & materials and all are saved so that if they need to be re-rendered, just tell me.
PSSS: I know I’m taking too long, but if you’re wondering why I’m going through all this trouble to make the icons look nicer… Have you ever seen “And Now For Something Completely Different?” (Monty Python, basically sketches from the first and second season of “Flying Circus.”) All of the animation in it was redone for the movie. I figure, if we’re remaking Blender, what could it hurt? Plus, how would we get the originals into our new one in the first place?
I love the enthusiasm, but I think perhaps you are getting a little ahead of yourself. There has been no talk of look and feel. Will it even have buttons like Blender. If it does, then one had better be sure that it isn’t too similar to avoid copywrite infringment.
I agree… but hey even if his buttons don’t work, he probably learned something along the way :). Plus if they don’t he might have a better feel for how to do better ones next time around.
but I bet that more that half the community would be left out if we built something that required it. (Though we’ll need some way to take advantage of this and other new GPU features as they emerge, especially in a realtime component.)