2007 was probably The Year Of Free Graphics: AMD/ATI’s specs, a totally new Mesa, output hotplugging via XRandR and the announcement of new shiny OpenGL specs. While this all was truly great, the OpenGL releases never happened, and there are no updates on the topic.
Graphics in Free Software were always a difficult topic in former years: specs for many hardware bits were missing, proprietary drivers often didn’t support new hardware for several months, the graphics API (OpenGL) was seriously falling behind Microsoft’s competing API (DirectX) in terms of features and hardware support and last but not least the main free implementation of the API, Mesa, was hopelessly outdated.
But this all changed, and many important changes were in 2007:
- XRandR made output hotplugging possible
- Mesa was released in version 7.0, finally bringing OpenGL 2.x support to free graphics drivers
- It was announced that the OpenGL API would be released in a cleaned up version as well as a total new version pretty soon.
- ATI/AMD made an impact by releasing proper and detailed documentation for their graphics hardware
While this list is indeed impressive, there is one big gap: the announcement of the new OpenGL specs. These never happened. Not even a cleaned-up version of the current OpenGL API ever came up anywhere. And, although it was said that the communication with the community would be improved there aren’t any information available about the current state. In fact, all possible channels are deserted. The only bit of information available is a post by Barthold Lichtenbelt (OpenGL ARB Working Group chair) in the OpenGL forum dated end of October 2007 stating that the development of the next spec faced difficulties and that the release will be delayed:
The OpenGL ARB found, after careful review, some unresolved issues that we want addressed before we feel comfortable releasing a specification. […] None of these issues we found are of earth-shocking nature, but we did want them discussed and resolved to make absolutely sure we are on the right path, and we are. […] The ARB meets 5 times a week, and has done so for the last two months, to get this out to you as soon as possible.[…] We don’t want to spend time fixing mistakes made in haste.
[…]
More details will follow soon in an upcoming OpenGL Pipeline newsletter.
Of course unresolved issues can turn up. But in such cases a vital part of maintaining a community – and also a stable customer base – is to deliver more information from time to time. And it is important to state schedules as soon as it is possible. If something is difficult, than that must be stated. Communication also means to say something from time to time just to show: we are alive, we care.
But the OpenGL team has failed in that regard: the mentioned OpenGL Pipeline newsletter is totally outdated and therefore a joke, and the forums don’t contain any other official post explaining the situation.
I’m pretty sure that there is work going on behind the scenes – there are many different parties with a vital interest to ship a next generation OpenGL. But the total lack of communication, the inability to handle basic communication principles in the right way is depressing. OpenGL is not only a technology, but also a product, and if the product is managed that bad it can’t be good for the technology.
Also it leaves the sad feeling that, in respect of the API, the free graphics ecosystem is still back in the old days.
/home/liquidat 
Thanks for this overview!
In February this year, Keith Packard has done some very promising talks about the future of X.Org:
– http://www.radeonhd.org/fosdem-2008/fosdem_2008_intel.ogg (from http://www.radeonhd.org/?page=fosdem_videos)
– http://youtube.com/watch?v=-oFxhqYn-g0
Hi Robin, thank you very much for these links! I checked for Keith’s talks some time ago but there was nothing, so you did me quite a favour 🙂
Hi liquidat, nice article, I’m an admin of the video game website JeuxLinux.fr (the french biggest website on linux games) and I would like to know if I can publish an translation of this article ? 🙂 And nice blog, there are many good things, 😉
julroy67, the licence of all my articles is clearly stated on the left side: cc-by-sa. So of course you can translate this article (and others if you want to) as long as you link back to the original and keep the licence 🙂
Oh I didn’t saw it x) But i don’t know, is the GFDL compatible with it ? Because all the website is under GFDL for me. I think it is 🙂 if not are you ok ?
I am not a lawyer, but as far as I know they are not compatible due to some law reasons. However, the next version of the GFDL will be made compatible to cc-by-sa. This is because the GFDL is debatable and very difficult, especially outside the US, while cc-by-sa learned from these mistakes and does a pretty good job.
Anyway, enough about the law stuff: I give you, julroy67, the right to translate this article and to republish it under GFDL. if you need more rights or whatever, drop me an e-mail 😉
I wrote a hobby games engine in directx/c++ a few years ago but have taken a two year break from programming when I switched to Ubuntu.
I’ve been itching to get back into it and have been holding off learning OpenGL because the new version was on the horizon.
Where should I start? Mesa, SDL, glut? I’m reluctant to invest all my energy in new skills that will be out of date soon. If you can just point me towards the right technology I can do the rest.
I have an 8800gts with proprietary drivers and would like to try all the stuff that my old GF3 couldn’t do such as parallax bump mapping 🙂
BobCFC, unfortunately I can’t tell you anything because I just don’t know the current situation well enough. I did hope that one of the OpenGL people would read this post and actually post an update somewhere, but it doesn’t look like it atm.
However, since you have a Nvidia card: the OpenGL core of Nvidia cards is the same for Windows as it is for Linux. So I would, for a start, go over to the Nvidia forums and ask the developers what they would recommend for cross platform developing with your card in mind. I’m pretty sure that they can give you recommendation.
This is what happens when you allow a critically important technology to be put under the control of a single commercial entity.
After the OpenGL ARB relinquished control over to the Khronos Group, there was all sorts of “big talk” about what ‘they’ were going to achieve.
SHOW ME THE MONEY KHRONOS
I distinctly remember having read of an “OpenGL rendering backend for Vista DWM [Aero] that will perform faster, be more stable and more efficient than the Direct3D implementation”. Which ‘they’ were going to provide, because ‘they’ themselves said so.
SHOW ME THE MONEY KHRONOS
Can you smell the vapours? I sure can…