 |
|
|
|
|
|
||
|
|
||
|
|
||
|
|
|
|
|
|
||
|
|
||
|
|
||
|
|
|
|
|
|||||||||||||||||
|
|
 previous story
next story  |
|
"OpenGL turned over to Khronos Group" Discussion Page: 1 2 »» |
|
 |   _mycoplasma_
Member since: 12/11/2005 From: Sterling, VA | |||
Posted - 8/1/2006 9:33:29 PM | ||||
| does this mean that khronos alone will have the final say as to which extensions become part of the core specification? | ||||
| ||||
| PhilMorton
Member since: 7/10/2006 From: Sydney, New South Wales | |||
|
Posted - 8/1/2006 11:17:29 PM | ||||
| yep. Or a new review board run by them will anyway. | ||||
| ||||
| medevilenemy
Member since: 5/29/2005 From: Vestal, NY | |||
|
Posted - 8/1/2006 11:36:08 PM | ||||
| I dont like the way things have been going in this industry. Way too many acquisitions are taking place -- way too much centralization. | ||||
| ||||
| AdamGL
Member since: 3/19/2005 From: Mississauga, Ontario | |||
|
Posted - 8/1/2006 11:44:40 PM | ||||
| Do you think this change in leadership will affect OpenGL's place in the graphics api race? | ||||
| ||||
 | phantom Moderator
Member since: 12/15/2001 From: Ipswich, Suffolk | |||
Posted - 8/2/2006 12:36:31 AM | ||||
| Yes, it will and I stand by my statements in the thread we had in the OpenGL forum a few days back that this is a good thing. Khronos Group have proven they can walk the walk with OpenGL|ES, combine that with NV and ATI's new "lets work together" point of view and things should start to move forward; viva OpenGL3.0  | ||||
| ||||
| Anonymous Poster | |||
Posted - 8/2/2006 5:55:33 AM | ||||
| I agree. This is a very good thing. Khronos have been doing good work with Collada and this tranfer will speed up OpenGL development. | ||||
| ||||
 | MARS_999 GDNet+
Member since: 3/6/2001 From: Sioux Falls, SD | |||
Posted - 8/2/2006 7:06:56 AM | ||||
| I hope so, for as of right now, I may end up coming back to OpenGL and don't want to see it die off if I do. | ||||
| ||||
| Toji
Member since: 9/20/2004 From: Salt Lake City, UT | |||
|
Posted - 8/2/2006 9:40:05 AM | ||||
| Personally I think that Microsoft is actually going to be the number one booster of interest in OpenGL. With the additions that Phantom has talked about OpenGL will easily be up-to-par with anything DirectX 10 can dish out BUT it won't require the upgrade to Vista, which means anyone looking to reach a large audience but still wow them visually for the next few years are going to be OpenGL junkies ^_^ | ||||
| ||||
| dreddlox
Member since: 10/20/2001 From: Auckland, Unknown Region | |||
|
Posted - 8/2/2006 9:41:30 AM | ||||
| Excellent. Now, if only they could release a revamped API so that it rivals/beats the coding ease of DirectX, whilst remaining Linux compatible. At the moment, I dont even touch OpenGL because I know that starting even a small project will involve several menial hours of making a translation layer so that I can access all the extensions which I normally take for granted. That said, anything that brings COLLADA closer to industry standard is a great thing in my books. | ||||
| ||||
| Xai
Member since: 10/15/1999 From: Nashville, TN | |||
|
Posted - 8/2/2006 10:39:12 AM | ||||
| your realize dreddlok that you could simply write that layer once as a library and use it on all of your projects just as easily as you link to DirectX today right ... Obviously it is easiest to have it already written by someone else, but I just don't buy the whole "starting a small project" part. You add support for extensions to your library as you encounter a use for them, and then they are forever more part of your easily accessable reusable library / framework. | ||||
| ||||
| RichardoX
Member since: 8/18/2004 | |||
|
Posted - 8/2/2006 11:10:12 AM | ||||
| This is what I feared. Khronos is after the money in the 3d industry. Their main concern is not to make a better graphics API, they want to see the green. I fear that Linux support may be dropped, the API will be screwed and OpenGL transforms from a nice, compact and well defined graphics api into a monster that requires hundreds of lines of code just to get a triangle on the screen. I'd like OpenGL to remain as a simple library that draws triangles damn fast. Collada model loading, scene graph and stuff like that can be done with libraries built on top of OpenGL, so the core API is simple and efficient as it is now and other stuff goes into other libraries. Please Khronos, don't screw up my favorite graphics API. -Riku | ||||
| ||||
| Anonymous Poster | |||
Posted - 8/2/2006 11:18:50 AM | ||||
Quote: Now WTF are you talking about? - opengl doesn't compete with directx, it competes with direct3d genius. - opengl is easier to code than direct3d (although it boils down to personal opinions). - you've never touched opengl yet you "know" you need this "translation" layer (whatever THAT is) for a "small" project? WTF? - there are extension loaders genius. | ||||
| ||||
| LucidIon
Member since: 4/1/2004 From: United Kingdom | |||
|
Posted - 8/2/2006 11:32:08 AM | ||||
| Many of the things that have come out of SGI seem to be similarily good, eg., OpenML. It's not because SGI have been first to do it or because they're just really smart. It's because there's a number of previous versions, that aren't so awesome. There's a stack of previous video APIs before they did dmBuffers which became OpenML. They all seem to provide a function call interface and a bare minimum of abstraction - there's probably something to be learned here. Collada sounds a lot like OpenGL Inventor (and friends), so hopefully, they'll see the obvious path and build it on top of OpenGL. (crosses fingers) | ||||
| ||||
| Bloodypriest
Member since: 5/29/2005 From: Chicoutimi, Quebec | |||
|
Posted - 8/2/2006 12:03:32 PM | ||||
Quote: Collada is already seeing industry-wide adoption. Although, my personal opinion on it, is that it try to be everything for everybody so it's really too big for it's own good. But, still, supporting Collada in your project doesn't mean that you have to support all of Collada. Just support the subset that you need. IIRC, it was designed with this use-case in mind. Quote: I fear that you do not know what you are talking about, my friend. OpenGL, as it stands now, it quite clunky and already requires too much code to get a single triangle on screen. The plans for OGL3.0 are about simplifying the API (see ATI and NVidia GDC presentation about that), making OpenGL faster, making OpenGL compete with Direct3D feature-wise and making the drivers easier to implement. Of course, they do realise that there are a lot of OGL1.x and OGL2.x apps out there so they are gonna provide a compatibility layer which they are gonna call OGL3.0 Full Profile (compared to OGL3.0 Lean and Mean which will provide only the OGL3.0 cleaned up core features). | ||||
| ||||
| Anonymous Poster | |||
Posted - 8/2/2006 5:46:46 PM | ||||
| I'm sorry, I really haven't touched DirectX Anything in... 7 years, so this is strictly a question and not a judgement. But since when was OGL more bloated than DirectX? I mean I still use OGL and I don't see a difference in what it takes to make a spinning triangle from 7 years ago until now. Is this strictly speaking about loading extensions and shaders or are we talking something else? Past that, I believe I last used DirectX 7 and the ONLY way to start an app in that was to copy and paste the startup code becuase it was PAGES of code from what I recall. Has this changed so drastically? If someone could please provide a semi-detailed non-flammatory response or perhaps a link to a comparison. I'm really STRICTLY asking and not passing any judgements here. Thanks in advance. | ||||
| ||||
| Heewa
Member since: 8/27/2001 From: Seattle, WA | |||
|
Posted - 8/2/2006 6:08:19 PM | ||||
| DirectX has improved a great deal since then! It's actually pretty easy to get set up. You do need to follow a tutorial or something the first time, but that goes for OpenGL, too. The worst part, really, comes from doing Windows related stuff, but if you use something like SDL, it's very easy. Here's a GameDev article about it, in fact: http://www.gamedev.net/reference/programming/features/dx9sdl/ In fact, if you write your own shaders, there's very little that you have to do in the way of Direct3D calls to get stuff onto the screen, since you're skipping the entire fixed function pipeline. | ||||
| ||||
| Anonymous Poster | |||
Posted - 8/2/2006 6:18:39 PM | ||||
| Nice, thank you, that is actually very impressive from what I remember, of course leaving out the windows part simplifies it but as is the case with OGL, yay SDL! Anyways, so what is going on with OpenGL that makes it such a mess? It looks like they're more equal now? Extensions are the only issue now or is there something else? | ||||
| ||||
| V-man
Member since: 3/2/2002 From: Montreal, Quebec | |||
|
Posted - 8/2/2006 7:31:08 PM | ||||
| "Anyways, so what is going on with OpenGL that makes it such a mess? It looks like they're more equal now?" GL carries on the old stuff. If you don't bother with the old stuff and just use shaders, VBO, FBO, ... you will do well. It also supports FSAA, texture anisotropy, automatic mipmaps, ... So if you are targeting GL 1.5 and above, you are fine. Yes GL and D3D are on par. If you want to support old video cards, then you may have to use vendor specific extensions and display lists, ... Also, there are too many ways to do the same thing and GL drivers are bloated. They are planning to make a "lean and mean GL" | ||||
| ||||
| zedzeek
Member since: 7/8/2000 From: nelson, Unknown Region | |||
|
Posted - 8/2/2006 11:42:09 PM | ||||
Quote: wtf  these are the members of knronus, (sure i admit they all want to make money) but now who exactly are they gonna get money from + how much? remove the tinfoil hat | ||||
| ||||
| Anonymous Poster | |||
Posted - 8/3/2006 8:26:52 AM | ||||
| khronos group sucks. It seems like a wanking off contest among wannabe graphics programmers. nvidia should control opengl because they have the $$ and are waaaay more intelligent than khronos group. OpenGL needs doers not talkers. | ||||
| ||||
| Toji
Member since: 9/20/2004 From: Salt Lake City, UT | |||
|
Posted - 8/3/2006 9:31:36 AM | ||||
Quote: Much like this discussion does not need opinionated and poorly informed anonymous posters who use words like "wanking?" Seriously though, I'm sure you would have much rather the standard stay with a group who was extremely active about it, like the ARB, right? Or even better, was taken over by a private company like Microsoft! I mean, look what they've done for DirectX! It's a little early for griping about inaction just yet. Give them a chance to actually DO something with it. | ||||
| ||||
| Anonymous Poster | |||
Posted - 8/3/2006 11:25:56 AM | ||||
| As long as OpenGL doesn't become a object oriented, Windows specific, closed licence API then I'm OK with it. Well, seriously, I dont get the consequences of that aquisition, I'll just sit and watch the happenings. | ||||
| ||||
| mikiex
Member since: 7/13/2000 From: United Kingdom | |||
|
Posted - 8/3/2006 2:39:58 PM | ||||
| This is a good thing, too many cooks before | ||||
| ||||
| Ravyne
Member since: 2/26/2007 From: Kirkland, WA | |||
|
Posted - 8/3/2006 4:15:33 PM | ||||
Quote: Khronos sucks and spends too much time 'wanking' and never really does anything? Please! Just look at the rapidity with which they made OpenGL|ES the dominant standard for embedded 3D graphics. There are now several software and hardware acceleration options for this API and, having read the spec, I'd say they moved it in the right direction. No more 5 ways of doing things, just 1 way, the most efficient way. What they've accomplished with GL|ES is exactly the kind of cleaning up that needs to take place for OpenGL to remain competetive with Direct3D as far as the Windows platform is concerned. They're not going to bulk-up the API by including ancilary things like collada in the core. Their various other standards will have no more to do with GL than DirectSound or the DX math utilities are integrated with Direct3D: They're designed to work together, and nicely at that, but not forced upon you. These are smart people. Many of the primary people in the Khronos group were members of the former OpenGL ARB. This is only good news for GL and its supporters. | ||||
| ||||
|
Page: 1 2 »» All times are ET (US)  | previous story
next story |
|
