Why dont you make a modern polygon based 3D engine, like Quake & Doom 3's engines? Why do you stick to voxels these days, you dont have a chance of competing like this. Or is it that engines like Quake are too complex these days?
JonoF at
Why do you stick to voxels these days, you dont have a chance of competing like this.
Perhaps it's not the competition that's important. Every man and his pet Labrador these days are writing polygon engines to rival a Pixar film, so the fringe areas are more interesting and fun to play in.
Jonathon
cragtek at
Re: Engine
Anonymous said
Or is it that engines like Quake are too complex these days?
That's also the case. It takes a large team 2 years+ to code a proprietry engine these days and there are significant barriers to entry in the marketplace to those who haven't got a commercial track-record. There is a compromise, which involves licensing an existing engine, but the initial cost is too significant to be considered viable.
So, to summarise, Jonathon has a point: albeit what I imagine to be a not very financially rewarding one.
My personal belief is that the only successful low-budget games in today's market (successful being a relative term) are isometric or psudeo-3D. Games like Locomotion, or The Sims. This way, the quality of the visuals rely solely on the abilities of the artist you've hired and not how many polygons your engine can produce efficiently.
Plus, it gives you plenty of time to concentrate on other things, like coding the actual gameplay.
-C
Sebastian at
Engine
altough I'm not ken, I want to say that voxel engines are not so easy to program as you might think. voxel engines need superior software optimization because there is no hardware support for it. there are also lot of other issues like how you're going to organize the world, i.e. octree or rle like in voxlap or so.
Sebastian
Awesoken at
Why do you stick to voxels these days
I didn't know that Voxlap would not be competitive when I started writing it in 1999. In fact, things were going quite smoothly for the first 3 years as I was optimizing it. Then around September of 2002, I upgraded to a Pentium 4. The frame rate was certainly faster, but something was causing it to not run smoothly. I noticed this was happening on all P4's. Many months passed before I realized what was wrong. Unfortunately, by that time, I had already lost interest in the project. Since then, I've been continuing to support Voxlap, but only in a passive way.
(FYI: the P4 slowdown was related to a "64K aliasing" conflict in the cache ... somehow, my screen buffer and Z-buffer were exactly a multiple of 64K apart : /)
Or is it that engines like Quake are too complex these days?
If I could go back in time, I think I could easily write a Quake engine from scratch. In 1996 though, it was tough. My main stumbling block was coming up with my own 3D polygon format and an editor for it. In those days, I never considered supporting an external modelling program since all the existing standards were rather bloated. It's really a shame because if I could have figured that out, I probably would have been successful.
Today's state-of-the-art engines are in fact quite complex. It's not a matter of 1 feature being conceptually difficult to implement (like in the old days). Today, the problem is that these engines must support hundreds of little features. This takes a lot of time - too much time for a single person to reproduce - even for me. I find it more interesting to work on new things that haven't been done before.
the only successful low-budget games in today's market are isometric or psudeo-3D.
I still believe it's possible for a low-budget game to use true 3D and also be successful. Don't overlook the possibility of licensing an existing 3D engine such as Genesis3D. There are plenty of others too ... hopefully JFBuild will join the list someday : )
Anonymous at
Awesoken said
Why do you stick to voxels these days
I didn't know that Voxlap would not be competitive when I started writing it in 1999. In fact, things were going quite smoothly for the first 3 years as I was optimizing it. Then around September of 2002, I upgraded to a Pentium 4. The frame rate was certainly faster, but something was causing it to not run smoothly. I noticed this was happening on all P4's. Many months passed before I realized what was wrong. Unfortunately, by that time, I had already lost interest in the project. Since then, I've been continuing to support Voxlap, but only in a passive way.
(FYI: the P4 slowdown was related to a "64K aliasing" conflict in the cache ... somehow, my screen buffer and Z-buffer were exactly a multiple of 64K apart : /)
Or is it that engines like Quake are too complex these days?
If I could go back in time, I think I could easily write a Quake engine from scratch. In 1996 though, it was tough. My main stumbling block was coming up with my own 3D polygon format and an editor for it. In those days, I never considered supporting an external modelling program since all the existing standards were rather bloated. It's really a shame because if I could have figured that out, I probably would have been successful.
Today's state-of-the-art engines are in fact quite complex. It's not a matter of 1 feature being conceptually difficult to implement (like in the old days). Today, the problem is that these engines must support hundreds of little features. This takes a lot of time - too much time for a single person to reproduce - even for me. I find it more interesting to work on new things that haven't been done before.
the only successful low-budget games in today's market are isometric or psudeo-3D.
I still believe it's possible for a low-budget game to use true 3D and also be successful. Don't overlook the possibility of licensing an existing 3D engine such as Genesis3D. There are plenty of others too ... hopefully JFBuild will join the list someday : )
struct _polygontype {
int x,y,z;
} polygontype;
there ive solved your problem 8)
Anonymous at
Anonymous said
Awesoken said
Why do you stick to voxels these days
I didn't know that Voxlap would not be competitive when I started writing it in 1999. In fact, things were going quite smoothly for the first 3 years as I was optimizing it. Then around September of 2002, I upgraded to a Pentium 4. The frame rate was certainly faster, but something was causing it to not run smoothly. I noticed this was happening on all P4's. Many months passed before I realized what was wrong. Unfortunately, by that time, I had already lost interest in the project. Since then, I've been continuing to support Voxlap, but only in a passive way.
(FYI: the P4 slowdown was related to a "64K aliasing" conflict in the cache ... somehow, my screen buffer and Z-buffer were exactly a multiple of 64K apart : /)
Or is it that engines like Quake are too complex these days?
If I could go back in time, I think I could easily write a Quake engine from scratch. In 1996 though, it was tough. My main stumbling block was coming up with my own 3D polygon format and an editor for it. In those days, I never considered supporting an external modelling program since all the existing standards were rather bloated. It's really a shame because if I could have figured that out, I probably would have been successful.
Today's state-of-the-art engines are in fact quite complex. It's not a matter of 1 feature being conceptually difficult to implement (like in the old days). Today, the problem is that these engines must support hundreds of little features. This takes a lot of time - too much time for a single person to reproduce - even for me. I find it more interesting to work on new things that haven't been done before.
the only successful low-budget games in today's market are isometric or psudeo-3D.
I still believe it's possible for a low-budget game to use true 3D and also be successful. Don't overlook the possibility of licensing an existing 3D engine such as Genesis3D. There are plenty of others too ... hopefully JFBuild will join the list someday : )
Today, the problem is that these engines must support hundreds of little features. This takes a lot of time - too much time for a single person to reproduce - even for me. I find it more interesting to work on new things that haven't been done before.
Perhaps it would be tenable, albeit difficult to manage, to construct some sort of design consortium to spread this workload over many people via daily team updates and communication over the Internet.
Having once overseen the creation of a truly disasterous mod for Thief, I can see the problems inherent with this concept, though. Reliability was the key problem. After all, it's not like you can enforce deadlines to any great extent - the worst that can happen is you kick them out of the team, and if they weren't committed in the first place they haven't lost anything.
A profit-share system would provide motivation, and threatening deductions would certainly provide a mild incentive to stick around - but I suspect the people you'd need for a project like this would have to be already deemed trustworthy.
Then you've got issues regarding someone having to wait for someone else to finish up before they can continue with their own work, though, and you get a chain of missed deadlines.
One solution might be some sort of modular program design.
Anyway, I'm waffling. Just a few random thoughts.
Anonymous at
How long would it take to make a Half Life 1 style engine, then?
Awesoken at
If that question is for me, I can't answer it because I've never seen or played Half Life (1 or 2).
ShadowX at
Ragnarok Online is a prime example of a simple polygon engine used in a successful game. With MMPORPGs being the big thing right now, graphics really isn't a focus of the game, kinda just a nice feature.
Ken, if you worked together a decent 3d engine (or used one of the ones out) and integrated the voxels in there, the engine would be plenty good to create a visually stunning game.
A little of topic, but you should do things different from what everyone thinks of voxel engine. Make the base terrain out polys, and just focus on architecture and characters in voxels. Now you have buildings that can be destroyed and character that can be drawn as beautifully as old 2d sprites.
I was trying to do it, but I'm just not a good enough programmer and just don't have the time to commit. I'd pay for an engine (along the lines of Torque) that supported both voxels and polys in a heart beat. I have a feeling there would be other people too once they saw what could be done with such an engine.
Edited by JonoF at
Silhouette of a Can at
Awesoken said
If that question is for me, I can't answer it because I've never seen or played Half Life (1 or 2).
Wow! Absolutely stunning! Where have you been all this time? Is Ken's Labrynth real? Are you trapped in it where the only access to the world is through a computer?
j/k
I just found that a little odd. You haven't even seen it? I don't play games as much today as I used to, but it's still hard for me to not run into someone playing a popular game.
Awesoken at
You haven't even seen it? I don't play games as much today as I used to, but it's still hard for me to not run into someone playing a popular game.
Sure - it's easy when you're at "video game" college.
Before college, my brother would find and play all the popular games - that's how I knew about the latest stuff. When I worked for 3D Realms, there was always somebody showing off another game and pointing out the highlights. Unfortunately, I don't live in that kind of environment anymore. Downloading and playing games is very time consuming and I just don't have the time or desire to weed them out myself. You'll understand when you're older.
Anonymous at
Awesoken said
You haven't even seen it? I don't play games as much today as I used to, but it's still hard for me to not run into someone playing a popular game.
Sure - it's easy when you're at "video game" college.
Before college, my brother would find and play all the popular games - that's how I knew about the latest stuff. When I worked for 3D Realms, there was always somebody showing off another game and pointing out the highlights. Unfortunately, I don't live in that kind of environment anymore. Downloading and playing games is very time consuming and I just don't have the time or desire to weed them out myself. You'll understand when you're older.
I know what you mean! Ive given up playing computer games for a long time! There are much more interesting things to do than watch pixels on a screen...
Anyway, do you have dialup internet? lol (i do :oops:).
cragtek at
Oh crikey, Guest. If you're going to suck up to Ken at least make it a little more subtle than entirely changing your opinion based on his!
Personally, I find that the written media is a good way of keeping abreast of the latest developments. I tend to do most of that on the toilet.