Thoughts on graphics...

[XxStratoMasterXx]

Ok guys, I've been mulling this for a while, but unsure if it is true, so I'm going to post it here and I want you guys to tell me your opinion;

In the real-time graphics world, many people see the pioneers in graphics as being the game engine architects such as id software, Epic Games or CryTek. But I feel that the true pioneers in the industry (Not to take away from John Carmack or Andrew Shiedecker) are the guys at ATi and nVidia who are constantly doing research on the algorithms to be used in real-time rendering fo their techdemos.

Yes, games like FarCry, and doom3 look great, but techdemo's really exploit the graphics cards features and are the things that pave the way for the game engine guys to use in their engine.

What do you think? The pioneers are at ATi and nVidia, or am I full of crap and it's really the game people?

 

[karlotta]

i would say SGi. Just follow the employment trail over the last eight years.

 

[Cryect]

I would have said SGI in the past but haven't been recently for graphics.

 

[karlotta]

I would have said SGI in the past but haven't been recently for graphics.

No not now, my point was the guys who left SGI and went to the IHVs. Follow the trail and you will see the big jumps in hardware and drivers. What we need is a Engineer Tree of GFXs.

 

[XxStratoMasterXx]

So...do you guys agree that they're at the IHV's and not the game devs?

 

[Inane_Dork]

Yes, games like FarCry, and doom3 look great, but techdemo's really exploit the graphics cards features and are the things that pave the way for the game engine guys to use in their engine.

What do you think? The pioneers are at ATi and nVidia, or am I full of crap and it's really the game people?

Well, part both.

Game programmers have the non-envious task of taking tech demo work and making it work in a game (A) without crashing, (B) with reasonable performance, (C) without gratuitous errors and (D) with much more variables (light types, model types, shaders, platforms, etc.). There's a good reason tech demos have cool stuff way before games do.

But on the other hand, you could claim that the "real pioneers" are the academic graphics guys. The guys who spend a year developing a new BRDF implementation. After all, they do it before it becomes feasible for a real-time tech demo.

It's just the natural progression of technology you're seeing. If you want to label a group "pioneers," I suggest you label the game devs "Gods of Patience" for handling all the crap that no one else wants to.

 

[ERP]

I'm going to agree with the previous poster to some extent.

Writing Demos is a totally different world to writing games.

If your looking for graphics pioneers, look at the academic world, most of what's happening in games is revisiting the work so you ccan make it run in real time dependably.

The hardware guys employ some smart people, but they certainly don't have a lock on inovation.

MS has a huge graphics research division, with some of the better known graphics people on staff. There are also various other groups in academia that turn out a lot of work year to year.

 

[nutball]

If you want to know who were the real pioneers you need to look back to pioneering times. Take a look at a copy of IEEE Transaction, or CG&I from the 60's, 70's or 80's, names like Whitted, Appel, Bresenham. It's really quite amazing what those guys achieved with really limited computing resources.

 

[Simon F]

If you want to know who were the real pioneers you need to look back to pioneering times. Take a look at a copy of IEEE Transaction, or CG&I from the 60's, 70's or 80's, names like Whitted, Appel, Bresenham. It's really quite amazing what those guys achieved with really limited computing resources.

Errrr... You forgot guys like

Blinn, Catmull, Blinn, Williams, Carpenter, and Blinn.

 

[nutball]

If you want to know who were the real pioneers you need to look back to pioneering times. Take a look at a copy of IEEE Transaction, or CG&I from the 60's, 70's or 80's, names like Whitted, Appel, Bresenham. It's really quite amazing what those guys achieved with really limited computing resources.

Errrr... You forgot guys like

Blinn, Catmull, Blinn, Williams, Carpenter, and Blinn.

Yep. Them too. (I didn't forget them, I was just too hungover this morning to remember their names )

 

[Reverend]

Errrr... You forgot guys like

Blinn, Catmull, Blinn, Williams, Carpenter, and Blinn.

You blinnking sure about that?

 

[darkblu]

If you want to know who were the real pioneers you need to look back to pioneering times. Take a look at a copy of IEEE Transaction, or CG&I from the 60's, 70's or 80's, names like Whitted, Appel, Bresenham. It's really quite amazing what those guys achieved with really limited computing resources.

Errrr... You forgot guys like

Blinn, Catmull, Blinn, Williams, Carpenter, and Blinn.

s/Blinn$/Evans, Sutherland and Newell

 

[Simon F]

s/Blinn$/Evans, Sutherland and Newell

I'll see your "Evans, Sutherland and Newell" and raise you a "Clark, Kajiya, Bui Tuong (Phong) and Perlin"

 

[darkblu]

s/Blinn$/Evans, Sutherland and Newell

I'll see your "Evans, Sutherland and Newell" and raise you a "Clark, Kajiya, Bui Tuong (Phong) and Perlin"

ermm, lambert ?*

* research pre-dating the 18th century won't be considered eligible 8)