Universal 3D File Format In The Works 464
telstar writes "The Register is reporting that more than 30 companies are working together to define a new file format intended to serve as a universal 3D file format. The new file format will be named the 'Universal 3D Format', or U3D. According to the article, they hope to make the new format as standard as MP3 has become for audio, and JPEG has become for 2D images. Interesting that they would choose two lossy media formats as models for comparison."
Re:Lossiness? No, try patents (Score:3, Interesting)
Uhm... yes... it'd create a jittery effect that could make a mess of things when it comes times to convert the rendered output to an MPEG.
What about VRML (Score:5, Interesting)
Is it not scalable or something?
I was always under the impression that it was as open as html.
One forming defacto standard... (Score:5, Interesting)
Blender support (Score:5, Interesting)
But...
One restriction I wouldn't mind, however, is the same sort of compatibility requirement that JAVA has: If something follows the format, it MUST follow it exactly and have no proprietary extentions. This tripped up MS when they tried to hijack JAVA for their own nefarious purposes.
Just my views on this...
Re:Really bad examples to pick... (Score:5, Interesting)
In the first line of the article, it says that Microsoft is involved with developing the format. Maybe I'm cynical, but I have little to no faith that this will come out as an open standard. We all know about Microsoft's SOP with respect to actual open standards that they've "enhanced". With them in on the ground floor on this one, I think it's doomed to be proprietary.
Re:What about VRML (Score:5, Interesting)
I would like to know what's lacking in VRML. A lack of foresight (didnt plan ahead for programmable pixel shaders, funtional textures, etc)?
And if it's that sort of problem, how can this new format not fall into the same traps, since the authors likely don't have magical crystal balls that tell them what types of information GPUs of the future will want to store.
Could be good (Score:5, Interesting)
But how general will it be? If it can handle detailed CAD models, and open landscape, and UT2003 style maps, high polygon characters and so on, then will it end up being unspace-effective for all of them?
Is there a reason why right now 3DS seems to be the nearest to a standard we have, when it doesn't even have many features?
JPEG might be the standard for images, but it isn't used for everything: Sometimes PNG and TIFF are used for particular reasons. TGA and PNG for example support Alpha channels, while JPEG does not. My friend draws pictures, and sometimes she gets good compression with JPEG, but sometimes the quality loss is terrible. Sometimes GIF is better, or something PNG is. And then there are vector graphics.
MP3 is nearly a standard, but we use OGG for political/legal reasons, or a lossless format when that is important. Real is often used when the sound needs to be streaming.
So, really, how useful will this standard be? And how free?
As one person Reluctantly in the 3D industry (Score:4, Interesting)
about a year ago i started doing 3D animations, this year i wanted to buy a collection of 3D models but in the end i found that 1 the app that i use has terrible support for model importing (blender) and 2 there are two many different formats out there. someone previosly mentioned 3ds but thats the stupidest thing ive ever heard because 1 its proprietary 2 it sucks. the closest thing to a standard is the
and for the 3ds guy your modeling software sucks and is a POS.
Re:.3ds (Score:5, Interesting)
I fail to see how this will be successfull... (Score:5, Interesting)
As there are so many different ways to represent the geometric structure of a 3D object that tie to the engine rendering that object. The fact of the matter is that 3D graphics rendering is still a non-trivial problem which requires optimizations for the use in question. Just about any piece of hardware still in use can handle JPEG and MP3 without a notable performance hit.
3D applications still push the limits of the hardware they run on and are keyed for specific intents; 3D games sacrifice detail and accuracy of modeling the interaction of light on surfaces for speed, while povray and RenderMan go for full hardcore ray tracing to make sure each pixel on the screen is accurately representing a reflective light model to the capacity of their respective engines.
Sadly, I don't think this arena has trivialized to a one size fits all format yet.
30 companies? Good luck trying to get an agreement (Score:3, Interesting)
These guys want to get 30 companies to agree to one specific file format that would probably have an impact on the work they do???
Good luck!
The only question that matters (Score:2, Interesting)
IS IT PATENT ENCUMBERED?
All other issues are secondary.
HDF (Score:1, Interesting)
Re:It might be something like this: (Score:5, Interesting)
They originally tried to do this through the Web3D consortium (the owner of the VRML standard) under the guise of a CAD format. After a lot of manipulation of the members and several other very dodgy things, the consortium told Intel to get lost. It's now just popped up again under another guise. The laughable thing is that this file format is completely inappropriate for CAD requirements. It's somewhere between a scene graph file format and a programming API, with neither being particularly good. For example, it's not extensible and has a lot of hardcoded strategies. If you wanted to extend or change an iimplementation of one item in the modifier chain, it would require complete reimplementation of the entire system. For example, changing the humanoid representation to using shaders for rendering the mesh was impossible. The entire format is designed around CPU-based rendering. Video hardware accelaration is not possible for about 95% of the spec.
Nothing has changed at Intel since we were dealing with them for the last 2 years on it. Effectively this project is 2 engineers and one manager trying to save their arse and the code from failed Shockwave efforts.
An example - the press release says it will be an ISO standard. The ISO people have no idea what Intel is talking about as they've not been approached yet. It would fall under either SC24 or SC29 subcommittes (SC24 3D graphics, SC29 is programming and home to MPEG) and both of these committees already have standards that fullfil these requirements (MPEG and VRML/X3D). It wouldn't make it past the front gate at ISO.
Re:Really bad examples to pick... (Score:1, Interesting)
I'd say go with ubiquity--not only because it's right, but because it gives your sentence an almost poetic flow with "utility."
They pursued the crown of ubiquity
Worn by the JPEG and the MP3,
While others would prefer utility
as the humble OGG, or the PNG.
(Score: -1, Too Much Free Time)
Re:What about VRML (Score:3, Interesting)
Mainly, VRML kept trying to position itself as an internet technology, as opposed to a simple, standardized file format. This lead to a couple of nasty effects. First was the fact that it was pretty well designed to look like HTML. This, in and of itself, isn't terrible; I've always been a big fan of the taxt based .obj file. However, it meant that they tended to take a very text based approach to a very graphical medium. You could create sphere, cones, and cubes with a couple of short lines of text, but creating complex shapes out of polygons was a pain in the ass without serious modelling software.
Also, since it was getting pushed as an internet tech, there always seemed to be more focus on what could be reasonably run on machines in real time, as opposed to what could be rendered over the course of several hours. Thus, anyone doing serious graphics work wasn't going to save their document as a VRML file since it wasn't going to support features that they needed, like Shaders, NURBS, or UV texture maps (in the early versions, anyway).
Finally, there's already plenty of 'standards' out there. Darn near anything will read a DXF or OBJ for simple geometries and most serious software will read 3ds files. Thus, if VRML wanted to be the true standard, it needed to offer something that no one else did. The great chance that they had for this is in Animation. To the best of my knowledge, there's no good way to transfer animated scenes between programs (if I'm wrong on this, PLEASE correct me). Well, one of the big pushes in VRML was to add animation very quickly. Unfortunately, they decided to do this by simply adding Javascript. While I'm all for scriptable text formats (ie Postscript and LaTeX), this is about the equivalent of the MPEG committee saying "Why don't we just add Javascript to JPEGs?" It's got it's niche uses, but it's never going to be the maintstream standard.
Thus VRML gave up just about any chance of being a real standard format for 3D. Then a lack of quality browser plugins and the fact that the giant file sizes didn't mesh well with the 28.8 modems of the era left VRML to die a slow death by starvation.
Re:What about VRML (Score:5, Interesting)
key difference: vrml is for realtime 3d.
interesting note: more students have had success with using the unreal engine to model spaces. it is much prettier, and the navigation is better.
I was excited for a moment... (Score:3, Interesting)
So what's the point here? Will this enable me to model dancing hamsters and spinning thingies in Alias or Rhino and export them directly to Front Page and Power Point? Be still, my beating heart.
Re:I fail to see how this will be successfull... (Score:3, Interesting)
Yes, but can it interoperate? (Score:2, Interesting)
Adobe, Boeing, Dassault/Systemes, NGRAIN, Lattice, Microsoft, Parallel Graphics, SGDL Systems and Tech Soft
Where's discreet (3dsMax), where's avid (Softimate), maxon (Cinema4d), or alias (Maya), or how about newtek (Lightwave)? Maybe this can become the universal CAD format, but if those are their backers, don't expect this to become a standard in the high-end 3D arena. Someone mentioned Kaydara Motionbuilder earlier -- that's good, but proprietary.
I wouldn't give this much credibility as a UNIVERSAL format until they get some of those companies in on it... And if they are, and they aren't listing it in their FAQ, they're foolish.
already an iso 3d standard (Score:2, Interesting)
web3d's x3d [slashdot.org]
so if they think iso is gonna aprove a second 3d format they are being pretty silly
Re:Really bad examples to pick... (Score:4, Interesting)
Re:Why VRML sucked (Score:3, Interesting)
It's not committees that ruin concepts, but lack of a concrete agenda. Start with a solid goal, continue with cutting edge research, and round it out with a coherent standards doc. That's how you make a file format.
issues (Score:2, Interesting)
What happens when you lose data in a 3d file? in images you have less crispness and contrast between adjacent similar pixels. In audio you have the same effect between adjacent similar frequencies. What happens when adjacent vertices get confused? You wind up with corrupted geometry which makes the entire thing worthless. (this assumes that we're using specifying vertices and not just using mathematical primitives)
Then there's the companies involved. I can see intel knowing something about 3d. They make processors for all of our favorite stuff. But Microsoft usesd to own Softimage which was the industry standard for many years. But did nothing with it but let it sit and rot and eventually sold Softimage to Avid. Have you seen Adobe's 3d stuff lately? They ought to stick to 2d. Why did none of the other 30 companies get mentioned by name? Who are they? is Alias involved? Softimage? Newtek? Side Effects? Discrete? Kaydara? These are companies that I know have a clue as to what it takes to make a 3D format, they've already done it.
What kind of applications would this format be aimed at? the needs of an architect making CAD drawings are vastly different than those of an animator making character animations, which are different than those of someone making scientific visualizations. An architect doesn't need any dynamic simulation routines or an IK solver. But an animator doesn't need solid modelling features or measurement tools.
That's all my ranting for now.
Re:HEY! (Score:3, Interesting)
One of the great things about some of the open formats out there (like openGL and vrml) is that they are exensible... don't have constructs for CSG? Go ahead and make them.
Besides, I REALLY like having a format that, when I'm just playing around, I can make text file and with a couple of lines have spheres and lights and cubes and stuff. I've designed simple furniture, including a hutch for my guinea pigs, using OpenInventor (which is basically VRML).
Existing "Universal" Format (Score:2, Interesting)
I know 3D Studio MAX and Rhino3D support it and I believe SoftImage and Lightwave do, too.
Intel's Motive (Score:3, Interesting)
But what about Intel's motive? From the article:
Getting a chipmaker involved in a 3D file format committee sounds like a good way to ensure a very computationally inefficient format that needs custom hardware to encode and decode. Heck, why not get some RAM manufacturers, hard drive manufacturers and bandwidth suppliers on the committee to make sure the file sizes are huge, too?
Re:Control mesh noise with a slider (Score:3, Interesting)
to the IEEE-754 standard is perfect. It is not
real arithematic, but it is well-defined.
again ? (Score:3, Interesting)
Despite even a good deal of acception, such as the FiLMBOX format, I keep seeing people falling back to
- export and import
- export and import
- export and import through third party plugins
- in-house export/import routines
I'd be all for an XML format. Yes, I know, storage space.. but considering there's a limited datatypeset, a compression routine could easily be written that collapses the file to a tidy binary, which a decompressor could then stream right back out to tidy XML. But whatever