Slashdot Mirror
Blender Community Rescues Sources
Christoffer Green writes "Today the Blender funding campaign went through the 100k limit,
sufficient now to pay for the ransom fee needed to make Blender Open
Sourced. The Blender Foundation aims to have the deal signed before
October 1, do a pre-release for donating members only at October 5,
organize a Blender Conference in Amsterdam October 11-12-13, and make
the official CVS release on October 13 for everyone.
This doesn't mean that you should stop donating though. The foundation
still depends on your contributions to cover costs that have been made."
as far as I can see, E100K has actually been paid. Another E8455 is 'pending', so they already should have some funds for their expenses.
Great news!
So, blender is what? I gather its some product that we're buying from its stockholders. I've got to say, its a really neat idea, buying off a broduct to make it open source.
Mod point free since 2001
Hopefully this will be a tend of things to come. Esentially this will allow for more users in the future, on more platforms. It even allows the possibility of selling the compiled project with the helpful additions from the OpenSource communitity, with the exception of making sure that the source is still available (under GPL).
-- Never monkey with another Monkey's monkey
I could swear, these things cost much less than that
The company who owns blender burned what, E10mil?, in the last 2 years... on what? I still remember the loudmouth directing the company telling everybody who wanted to hear it that blender would be so profitable... What did he do with all the funds he raised? Will the people who invested in the company be thrilled the source is sold for 100K? I don't think so.
Besides, blender the product, is ok, but the interface is so darn goofy it takes a hell of a lot of time to get used to it, especially when you compare it to the big boys in 3D world: 3DSMax, lightwave and Maya
Never underestimate the relief of true separation of Religion and State.
Great question! Viruses? Heh. OK, seriously...:
DVD-Video creation.
From capture to encoding to muxing the bits together with navigation, and burning. There are many such tools available for Windows.
I know `dvdrtools` works at burning pure-data DVD's, but to burn DVD Video you have all these other steps before it, and the toolchain does not exist in Linux.
DVD Video is an area that will lag on Linux for a LOooong time? Why? Because the software comes FREE with DVD-Recorders, so there's an incentive to dual-boot. If you don't like that gratis authoring package, many people will (like it or not) grab DVD Maestro or something, off Gnutella or Kazaa networks.
So there is not enough DVD authoring on Linux: not for commercial packages (if there is even one DVD suite on Linux, it certainly ain't sub-$500), and not for the glory of being the first GPL toolchain. The specs are scarce, development is hard and it's too much for one developer looking to provide us with a solution, no matter how much glory there is in doing so. It's a brutal, team-based development project spanning several domains of expertise.
I'm glad to see this succeed, and I'd like to see this new type of "market" compete with both commercial software, and the pure-free stuff we already enjoy. The competition will give us what we need, and may the best team win!
yes, and in fact the OSS community (in this case, myself and a small handful of others) already do!
ardour is my own contribution to this issue.
3 years of full-time unpaid labor, funded by income from amazon.com, tested in a commercial recording studio, aimed squarely at the high end market with low end costs.
its massive, its complex, its very very very hard for a novice to build, its only available from CVS at this time. do you think it will get better? you'd better believe it! package releases coming up within 6 weeks, v1.0 hopefully within 12 weeks.
Remove the chain in that image awwready. It's 'Free'd.
My photolog
Just thought I'd mention that due to a generous money donation by a private sponsor, the LinuxTrade software was converted to the GPL on 08/30/02.
This is a great trend, IMHO.
Promises are not easily forgotten. Who is the lucky person who gets the T-shirt?
Blenders back end is amazing, but the interface is based upon enabling a slow and inefficient style of modelling that is no longer usable for comercial production.
,x/y/z/xy/yz/xz separation of input, etc would be good too. This and the ability to save and easily modify these active planes (for input) would make blender much more powerful and allow work to be done in a single perspectival window (maybe with small orthographic views for newbies who don't dream in wireframe and can't see it in perspective)
... I need pictures. Code the point/control point location as an equation taking things like remaining tangential to this point, and maintaining a right angle between these two sections and remaining within a certain part of the length of this line in a way that is dynamically updatable. Do that and make it stable and easy to learn and the modelling world will stop and praise you!
It has many things that commercial packages do not have. What it lacks is predominantly in the interface. Yes I have used it, and much more advanced (and expensive) packages.
It needs to move away from the three orthographic views for modelling, one perspectival view for visualisation mode of design, where people use ten moves in three windows to achieve what should have happened in one move with ABSOLUTE ACCURACY using object snaps. The people who do this in front of me, then tell me that they are saving time. They continue to say this when they are in living hell later on when they need to use boolean operations or anything advanced with their mess of a model. I would find it funny, were it not for the human tragedy (DON'T THINK THAT'S A JOKE)
In order to move away from this interface mode, Blender will need to separate the viewport from the active plane (the co-ordinate system being used for input and editing of objects) and implement GOOD snapping for endpoints, midpoints and center points as an ABSOLUTE MINIMUM. snap to face
The other enhancement NEEDED is an improved HEIRACHICAL layer structure. The present collection of little buttons that pass for a layer structure are humerous if you don't actually try to use them. A layer structure with grouping, toggleable visability, snapability, selectability and lock status is part of modelling. If this could be used to facilitate object selection, apply heirachic object propeties according to group membership, and be extended to transparently allow for the division of the project into blocks (separate files, I think this is practically done) that could be used simultaneously by a range of designers on different tasks, then Blender would be up there with some of the best editors in existence. (Moving the configuration stuff into dialogs and/or running it vertically would help the interface a lot too. The basic layout of the buttons is very pixel hungry)
These things are not big additions compared to the amazing stuff already in there, but I haven't seen and no doubt wouldn't understand the code involved. I know nothing about it's language and the developers have been too defensive about their interface to be worth approaching.
To be the worst nightmare of EVERY commercial 3D/4D modelling/rendering program around, here are some non interface related suggestions:
It should improve the granularity of it's sub-object editing. Selection, deletion and insertion of points, lines, curves, faces, subfaces, control points etc, and their simultaneous selection at a range of levels (select different points, lines, faces and objects and move them with a single operation.:-) This will bring blender up to spec with some of the most efficient and intuitive modelling tools around.
To take a leaf from some of the work in development at microstation (I am not from microstation. Sorry microstation, you should have continued your support for Linux) They are working on some seriously cool new tools that TOTALLY BLEW MY MIND. I would leave unix forever for this.
Ready, They are working on something like a GUI integrated development environment for the back end scripting of models as part of the standard modelling tools, so that you can use a GUI to tell a point to remain at the
Anyway that's probably long enough.
No warranty of any kind is offered as to the quality of this post.
A lot of people have been talking about the usability of the Blender interface.
Now obviously I am not the first one to admit that it's not incredibly strait forward. Usually if I have taken some time away from blender, it takes me a while to get back into the groove of things.
However, one thing I have noticed. Once you actually DO get used to the interface, everything starts to feel natural. You stop thinking about how you use its features... you just USE them.
Blender is not your typical mouse interface. To use blender properly you need both hands.
However... is starting to get a bit dated compared to the many of the other 3d modelers out there. But this is why we are getting the source!
I hope a lot of the people who actually DO work with the source, understand just how uniquely useful the interface actually can be.
A good in-program tutorial would probably go a LONG way toward the usability problem. I should be able to say... I wish to perform this task... show me how to do it!
Now someone mentioned the widgets being a little weird... Yes I agree that's true. Take up too much space? They are 3d widgets! Zoom out some. (Although yes, you do waste some space on the sides when you do this... =/)
I actually kind of like how the widgets work. If you need to set an absolute value, shift click and enter the value you really wanted.
Again, I hope that when blender actually does end up being hacked to pieces (think mozilla)... that the developers take into consideration that quite a few of the interface features actually do work... and are fairly comfortable to use.
So yes... Blender is starting to show its age. But think of it this way, right now Blender is kind of like Netscape 4.x. Give the community a year or two with the source, and you might just see some amazing things done with it.
Now... I'm probably not the first one to think that Mozilla was over engineered. Although I hope the same thing won't happen to blender... Mozilla eventually did turn out alright. And who hasn't been accused of over engineering? I know I have =)
Luke
Support will start at $5K/year. For that, you will get dedicated 24/7 service from a set of the developers, accessed via a single number. you will have to run ardour on a system we build for you; if you run it on your own system, support will cost more. if this bothers you, consider that protools for windows is certified for only a single intel-based system, built and sold by IBM. run it on any other system, and there is NO support available.
let me know when you want to sign the contract. i suggest you at least wait till version 1.0 comes out, but don't let that stop you.