Archived

This topic is now archived and is closed to further replies.

modern graphics engines...

This topic is 4940 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

Hiya Everyone! Man, those screens of the unreal 3 engine look incredible. I agree with nehe, it's hard to believe that is how the gameplay graphics will look. Anyway, the levels of detail and realism achieved in modern video games are all thanks to programmable gpu's and complex shaders, and hell, shader technologies are becomming more intuitive and easy to use every six months or so. Techniques such as ray traced lighting are really becomming useless, lets face it, there will be a point where graphics can not get any better and I feel it's safe to say that day is comming soon. Now back to my point. It seems that nowdays, a rendering engine (not including animation, visibiltiy testing etc) is basically a collection of shaders. Am I wrong? Granted I am right, this means then that to pump out the next gen engine is simply a matter of creating a bunch of complex shaders. of course I realize there is a lot of work envolved still with it, but what im saying really is that, rendering engines are no longer as "complex" as they used to be. A rather vague question, well more of a question of opinion... When kids start programming, they think that because they can make a program say hello, soon they will be creating the doom 4 engine. While thise is obviously not possible, I wonder if it will be possible for your average hobbyist to be able to produce something that looks as good as the next gen graphics engines. ~Looking forward to your feedback... ~Jason [edited by - no one on May 28, 2004 4:22:12 AM]

Share this post


Link to post
Share on other sites
You forget the fact that Unreal 3 will become available in 2006, so the CPUs, GPUs and RAM will be maybe 2-3 higher than now.
As for your question, of course that a hobbyist team will never be able to compete with the big companies that have much more resources than the independent developers.

Share this post


Link to post
Share on other sites
I personally think the point that graphics can''t get any better will take at least (and i think it will take even a lot longer) another 100 years. Right now you (and i) are amazed by Unreal 3 Screenshots, in 5 years you will think of it as ugly and old. There are a lot of things that could be improved, physics, trees (i haven''t seen any high-detail realtime trees yet), rain, fog, dust and water (they are really expensive to do when you want everything to be perfect) and of course lighting.
Everyone knows Finding Nemo, rendering a frame for that film cost pixar 6 to 12 hours. I don''t think (along with high quality physics) that will be possible to do in realtime the next few decades, when that is possible the next step is (i think) adding realtime environments including chemestry (up to the atoms or even further), real life and real worlds, i think only when that is successfully done in realtime the graphics are so good humans can no longer see the difference between real life and computer graphics.

However i also question the realtime graphics in the Unreal 3 engine (i wonder how it runs on today''s pcs at full detail, or at the next generation pcs).

Share this post


Link to post
Share on other sites
well the thing about shaders and other stuff that make things that where hard easy is that they alow engines to evolve more and more.
They basicly let people concentrate on other things.

However writing a engine is a lot more complicated than just slapping together a rendering algorithm with stencil shadows, normal maps and some shaders ontop of that.
So it is possible to create stuff that are at the same class that of the profesional developers, it is actuarly all up to how good they are, allthough on a slightly smaller scale.

What comes in the future is hard to say, they first have to develop the next generation of engines(that includes Doom 3 and unreal 3) to their full potential.
but hey it took them 10 years to develop the current generation of game engines from the old quake to farcry and HL2.
However in about 10 or so years i suspect that the fourth generation of fps engines, this engine will be better bigger and more butiful than anything that exists today.
But still it will not be until around generation 7 that we will aproach a point where it''s hard to improve uppon it.
Hobyists will not be far behind, hey, im a hobyist, and im developing a third gen engine even though the closest release of a third gen engine is not for another 1-2 months.


---------------------------------
For an overdose of l33tness, flashbang.nu

Share this post


Link to post
Share on other sites
Real-time photorealism? I doubt we''ll see it in next 20-30 years. There was a long post about simmilar things on opengl.org where one guy stated that "we will be there in two years" (Bases on doom3 screenshots he saw)

I think hobyists developers can and will produce same (or even better) quality engines or at least parts of them. Just look at the demo scene. New techniques get implemented in those engines years before we see them in games. The limiting part is not the main engine, its the art resources (textures, models, levels, sounds) and the tools to create them.

You should never let your fears become the boundaries of your dreams.

Share this post


Link to post
Share on other sites
Those engines are the result of many man-years worth of work. So hobbyist will have a hard time keeping up with that. But as a hobbyist it''s easier to take shortcuts, just implement what you need to get a good result.
As for engine complexity, a few years ago you had to do all your rasterizing yourself, now it''s easy to render textured and lit polygons and it can be done in a "hello world"-sized program. But the state-of-the-art engine are not less complex than they were. The complexity has just moved elsewhere.

Share this post


Link to post
Share on other sites
I don''t see graphical levels settling any time soon.

While the stuff you see looks nice its a long way from looking real. I cant wait to see games with near unlimited drawing distances, cook torrance lighting, fluids that look and react like real water would act, true refraction/reflection, etc...

I wouldn’t worry about it getting tot he point where it cant get any better.

As for hobbyists, they wont be able to produce a id quality engine, but they will be able to produce a decent looking game. Just because some companies have teams of 20-40 people, it doesn’t mean there is no way a 5 man team cant get something done in the same amount of time. I myself have been working on a large scale game with a few other people for 10 months. It''s unbelievable the progress you can get done with a few determined programmers.

You can check out some screenshots of it at www.endgameinteractive.com

Share this post


Link to post
Share on other sites
You guys are right. Skow, nice work man! I just get depressed sometimes because I can''t stand just watching from the side lines, when i get depressed I start to worry about things: "how will I be able to ever make something as awsome!???" "what if it progresses to fast now that i will never be able to catch up?!!!" etc.

Anyway, what I was trying to say about the complexities of today''s graphics engines was not really that the ''average joe'' can make one, what I ment was that today''s programmer has a huge array of tools they can use to get the job done, where as in 2nnd gen engine programmers had really only what they made. heck, a shader can be made by an artist in 3ds max (if im not mistaken).

One thing that I feel programmers should worry about is coding them selves out of a job. If they make everything so that it can be done by an artist, what the hell is the programmer going to do when ai has advanced to a level where computers will be able to create game engines by them selvs??!!!

Well, I guess that is rather a science fiction idea, however, 20 years ago it was stated that computer graphics would never reach a point where they would look real, and well, they have...

Alrighty, Im exhausted. good night guys.
~Jason

Share this post


Link to post
Share on other sites
"One thing that I feel programmers should worry about is coding them selves out of a job. what the hell is the programmer going to do when ai has advanced to a level where computers will be able to create game engines by them selvs??!!!"

Heh, I think you have watched the matrix one too many times. This wont be happening in our life time.

Share this post


Link to post
Share on other sites
I think hobbiest can create something better then large corporations. Im a linux user and what is linux but an OS created by hobbiests. Its the best OS on the market right now and more and more people are going to it. If someone can organize enough people sure its possible.

Share this post


Link to post
Share on other sites
Large corporations(and small ones to) have only a couple of advantages over hobysts, otherwise it''s the same.

1.more dedicated time(latley i have been working on a two month long project, it really makes a difference when your working on it full time).

2. more recources(what can i say, money helps).

3. better organisation (documentation is key if a project is to succeed).

4. finaly, talent thends to gather in these places(it''s simple realy, who would hire untalanted people).

---------------------------------
For an overdose of l33tness, flashbang.nu

Share this post


Link to post
Share on other sites
quote:
Original post by Nukem
I think hobbiest can create something better then large corporations. Im a linux user and what is linux but an OS created by hobbiests. Its the best OS on the market right now and more and more people are going to it. If someone can organize enough people sure its possible.

Using Linux as an example of being able to compete with Unreal 3 is an unrealistic comparison. OS''s are built on well established princiles that have been tried, tested and refined. Graphics engines have a tendancy to require different structures and approaches every new hardware generation as the requirements and focus changes.

Thats not to say an open source project can''t do rather well, and projects like OGRE are pretty advanced. But you won''t see bleeding edge tech appearing in it.

Share this post


Link to post
Share on other sites
hehe, yeah the matrix has me... lol. I just really hate living where I am, people around here seem like there baffled by a light bulb let in though knowing anything about programming. I want to get involved with a project of some sort and actually start doing doing something with my skills instead of just watching what others are doing and have done. It just feels as though that time has stopped for me but the rest of the world is in hyper drive. Thus, I have a strong feeling of distance between my dreams and where Im at now. School will take another 4 years by which time, the 4th gen engines will comming out.

I realised something the other day, I started coding around the same time carmack started work on the doom 3 engine, about 4 years ago. Before that, he had already made like what, 8 engines.

Share this post


Link to post
Share on other sites
What''s the main reason you started programming? Games? Animations? Anything else? Just go and make that one thing (or at least try), don''t just sit there waiting for a project, start one, alone or with other people.

Share this post


Link to post
Share on other sites
quote:

Now back to my point. It seems that nowdays, a rendering engine (not including animation, visibiltiy testing etc) is basically a collection of shaders. Am I wrong?



Yes. Well, more specifically, the animation, visibility testing, precalculation, data management, and art path are where all the hard work happens. The actual shaders are mostly cake (or, if you prefer, Easy Cheese).

Share this post


Link to post
Share on other sites
I think the unreal 3 screen shots are really starting to show a trend in graphics that has been a long time comming:

How good a game looks depends entirly on the artists abilities to exploit the code, not the codes abilities to exploit the artist.

If you get me..?

I basically mean we are simulating now not emulating.

| - My project website - | - email me - |

Share this post


Link to post
Share on other sites
The only reason Unreal 3 looks so good is not as much because of shaders but bacause it uses HDR lightining. Which is a very effective way to produce photorealistic images. However this technique has its own limitations.

Check out http://www.debevec.org/ for more info.

Share this post


Link to post
Share on other sites
I still think 'hobby programmers' are in with a shot - perhaps more so nowadays, as what I think as the 'base level' of gfx quality & speed is now achievable a lot easier than ever with the gfx processors and faster CPUs. (OT: They also have I think more opportunity to tweak and include more gameplay elements than companies with deadlines.)
My pet project is the work of a single programmer and about 1 week's work from a modeller (and help from mailing lists - hplus stand up )
http://homepage.ntlworld.com/martingbell/projects/


[edited by - aph3x on June 3, 2004 8:13:35 AM]

Share this post


Link to post
Share on other sites