Humus - Comments

More pages: 1 ... 11 ... 21 ... 31 ... 41 ... 51 ... 61 ... 71 ... 81 ... 91 ... 94 95 96 97 98 99 100 101 102 103 104 ... 111 ... 121 ... 131 ... 141 ... 151 ... 161 ... 171 ... 181 ... 191 ... 201 ... 211 ... 221 ... 231 ... 241 ... 251 ... 261 ... 271 ... 281 ... 291 ... 301 ... 311 ... 321 ... 331 ... 341 ... 351 ... 361 ... 365

Query FailedApp
Monday, August 10, 2009

Immediate mode is no longer useful for moderm gfx applications.

If you need immediate mode for learning purpose, why don't just use old OpenGL instead?
You can still use old version of OpenGL. Upgrade is not enforced.

Eosie
Monday, August 10, 2009

BTW, you all seem to want updated GLEW. You can do it yourself, partially. It's simple:

- get a working linux machine somehow
- use Subversion to "checkout" glew/trunk
- go to that directory and type "make extensions", which generates the entire GLEW source code from the OpenGL registry including all recent extensions

You can do that everytime a new extension appears in the registry. (Maybe someone could write a blog post about this?

)

Eosie
Monday, August 10, 2009

Keep in mind that Khronos "is" just a bunch of vendors.

Kyle
Sunday, August 9, 2009

Imo removal of immediate mode is silly. Yes its not the way gfx works, but its easy and handy.

What khronos should do here (imo) is creation of standard utility layer on top of core GL where immediate mode could sit (together with some other usefull things). Its a pity that it wont happen as they doesnt seem to have founds to do that.

Thats also a reason i think we will never see major overhaul of api (like lets say dsa, and other changes to object management) - khronos is too weak to preasure gfx vendors effectively

Jan
Sunday, August 9, 2009

Well, it is more than an hour of work, but i did write an immediate mode emulation library for GL3.

It's open source (including commercial use), you can get it here:

http://www.artifactgames.de/glim

I agree with the decision to drop display lists and immediate mode, both have made drivers horribly complicated to implement.

I still think GL 3.0 was the worst thing to do, they should have created a completely new API. However, given that we have GL 3.0 as a basis, 3.1 and 3.2 have been very good steps forward. Not yet comparable with D3D10 or 11, though.

Jan.

kai
Sunday, August 9, 2009

have you considered contributing to glew or glee instead of doing your own library in the future? i mean that's the reason why they have been released as open source.

Overlord
Sunday, August 9, 2009

@Sopyer: with everything that has been happening in glsl (especially 1.40) and the removal of immediate mode i think we are heading towards having a "unified object system", as the only big thing left to change is texturing.
And seeing as it was less than one year ago version 3.0 was released, i think they have a plan and sticking to it, so things are moving pretty fast now, so fast that i would even presume to guess that in about two years everything will end up in one beautiful graphics crescendo that is openGL 4.0

Sopyer
Sunday, August 9, 2009

Humus, I know that it is possible to create small class to immitate immediate mode, but
1) this should do 99% of all programmers
2) this affects learning curve - which I think is essential for API to succeed without heavy marketing support of some monopolic corporation.
3) well I agree that we do pay for simplicity by increased driver complexity, but if this functionality is implemented in user space and shared between different drivers version why not? Who said that every driver version should reimplement all the parts?
By the way don't you immitate display list in your game engine? For example to implement some sort of effect system? When you first record states, textures, samplers for some techniques? I think is quite common too.


General
	News 3D Pictures Textures Articles Cool stuff
Other
	FAQ About Humus
Other 3D sites
	OpenGL.org GameDev.net Beyond3D Rage3D
Programming info
	OpenGL extension registry NeHe