So, here are my arguments for 2D:
* Easier to program
* Faster than 3D engines
* You can easily see what's going on - most 3D racers have an external view as default, so you can see around you (i.e. internal views are rubbish)
* 2D games usually have simpler controls, while 3D games (e.g. the SWAT series) have complex controls which are getting tricky to fit on just a keyboard and mouse.

And for 3D:
* More immersive/realistic environments
* Usually more control over the character (if a fps)
* Allows a lot of strategy for fps's, e.g. hiding behind a door, while in a top-down view you could see the enemy
* 2D games have a restricted view, while 3D's don't
* Depth/height is harder to perceive in a 2D world/view than in a 3D one
* Was there anything wrong with (Arc)Elite's graphics? Would we be bothered if we had another game with filled polygons instead of textured? The textures could just be an option - so new computers can take advantage of them, without any serious slowdowns

Have fun flaming me

I think this platform should get the ideas first and sort the realisation later

I don't actually believe that 3D necessarily leads to more immersive and realistic environments. Part of the problem is that there isn't a single game out there that reacts as you'd expect all the time. If I fire a missile at a wall in Quake does it blow a hole in it that I can walk through? No. There are more examples but 3D games, while giving an "impression" of a realistic environment rarely provide it. This is partly due to technological limitations but it's still a problem.

I don't believe you get more control over your character, what about Flashback? (Now there was an incredible game, and yet 2D).

The point about allowing for strategy, ie hiding behind a door while using a top down view to look around it destroys any sense of reality the game might have had....unless someone has invented the invisible floating camera and not told me . Besides you could achieve exactly the same effect in a top down 2D game - you only need to draw line of sight from the enemies to you, and because it's top down already you don't have to accept invisible floating cameras.

Would people be bothered if we had another Elite with flat shaded polygons? Well given some of the criticism I've read of Iron Dignity then...er yes.

Nathan also makes the point about there being no (actively supported and developed) 3D engines available to developers. This is a huge stumbling block. I've written a few basic ones in my time (nothing that could be used for a quality game) and people who haven't tried it just don't understand how complex it can be. I've made this point before but look at the PC market - very few people make engines, most people licence them. Those that make them don't tend to make very good games (sorry, I really think Quake is naff).

Anyway, not wanting to give up on my post, I think I'll pick a fight with Nathan... (Nothing personal, you understand)

SFE beat Quake because:
* It is a small download, compared to the shareware episode of Quake
* It runs at a decent speed, because it was programmed properly
* It has good multiplayer - up to 4 people on one computer, compared to Quake which had acceptaable net play but is now running out of players as people move on. Plus net games are harder to set up then getting a friend to sit down next to you to play against
* For single player, shareware Quake only has 1 episode and no bots, so will get boring after a while. However with SFE you could play for as long as you want against up to 3 computers.
* Plus there isn't much point comparing the two since they are completely different games

And Lee, that bit about the top down view - I was talking about 2D games there - In a 2D game you could see the person, while in 3D the door would be in the way.

Plus the bit about control - in Flashback you could look left, right, duck, run, roll, jump, and use items. In a modern 3D game you can look in all directions, duck, run, jump, strafe, lie down, climb ladders, and use items, as well as make up your own combinations - e.g. rocket jump scripts in Quake, although I'm not saying that's a good thing.

And just in case there as anyone else who was wondering, I'm talking about games where you are in direct control of the character - fps's, platformers, graphical RPG's such as Space Quest, Zelda on the gameboy, etc, and not really strategy games such as Exodus.

But you are probably right about this being irrelevant to RISC OS at the moment, although it would be a good idea to get the creative juices flowing so we know what we (and thus people who don't have an Acorn) want, so the games market can increase in size and RISC OS will rule the world.

About the creative juices, this is still irrelevant in RISC OS land. There are exceptionally few (I know none) coders who will write a game on what people want, they tend to think of something themselves and go down that route.

People want a new version of StuntRacer but you don't see people doing a version do you?

Also you could climb in flashback, and the only reason you couldn't climb ladders was because there weren't any 8P

Actually I do think SFE is better than Quake - I'll stop trashing Quake now (one dimensional, tedious game that it is...)

More seriously I would love to do an update of Stunt Racer however my skills tend to lie in other areas (ok, I know 3D, but probably not to the required level for a modern game). Also I have this darn annoying thing called a thesis to write. I suspect that 2 years down the line Andrew will be in the same boat. Of course if someone pays me 25k / yr after I've written up then perhaps I'll do it.

It all comes down to time really.

Anybody willing to go in on a joint venture? I doubt we'd be able to make a living off of a 4+ way cut, but we would all gain experience and put a bit more life in the community.

Either that or this is another of my half-baked ideas

So, here are my arguments for 2D
* Easier to program
GL is easier with a good programing library
* Faster than 3D engines
GL is hardware so is faster than software 2D
* You can easily see what's going on - most 3D
ok you got me on that one
* 2D games usually have simpler controls, while 3D games ¥e©g© the SWAT series¤ have complex controls which are getting tricky to fit on just a keyboard and mouse©
but complex controlles can often be used well I play Quake3 with 7 buttons which are always under my fingers and a load of extra buttons to do thoes extra thingys that arent really needed

With OpenGL

So, here are my arguments for 2D
* Easier to program
GL is easier with a good programing library

"With" a good programming library. Also I don't agree, it all depends on which APIs you're used to. I could argue quite convincingly that my incomplete Warp API is easier than OpenGL but then that's because I know it.

* Faster than 3D engines
GL is hardware so is faster than software 2D

Almost all 3D hardware has 2D acceleration as well. Just because something is done in hardware doesn't mean it's faster than the software equivalent. I remember an article on a hardware implementation of TCP/IP...it wasn't quicker than the software version.

* 2D games usually have simpler controls, while 3D games ¥e©g© the SWAT series¤ have complex controls which are getting tricky to fit on just a keyboard and mouse©
but complex controlles can often be used well I play Quake3 with 7 buttons which are always under my fingers and a load of extra buttons to do thoes extra thingys that arent really needed

I have nothing against complex controls - Starfighter has a few and I think it's great. I also spent considerable effort getting into Black Angel and didn't think the controls were that complex once you had familiarised yourself with them. However complex controls can put people off.

Finally regarding the comments on OpenGL. Well they seem slightly irrelevant given that we don't have a port that's suitable for games. Yes there is Mesa, and it's a fine effort, but it's huge and slow. We could spend ages optimising it but I figure porting or building a miniGL driver from scratch and working up from there would provide a better solution.

Finally regarding the comments on OpenGL. Well they seem slightly irrelevant given that we don't have a port that's suitable for games. Yes there is Mesa, and it's a fine effort, but it's huge and slow. We could spend ages optimising it but I figure porting or building a miniGL driver from scratch and working up from there would provide a better solution.

and a load of extra buttons to do thoes extra thingys that arent really needed

As for OpenGL, porting it to RISC OS could be a pretty good idea, and the Omega's 3D API is probably going to be compatable with it somehow. OpenGL would make making 3D games a lot easier (supposing it has a good software driver), and could attract PC programmers/multi-platform games. However it would make game conversion a lot easier, and us getting a game that PC's have had for years already doesn't help the market much (Although with the new computers coming out the time gap should shorten, making conversion a good idea again).

Even if we were only able to port miniGL it would be a positive step - the 3DFX Voodoo 2000/3000 only supports MiniGL and runs fine with it.

- Having the API would also make Half-Life that bit easier to port for Risc OS too

SFE beat Quake because:
* It is a small download, compared to the shareware episode of Quake
* It runs at a decent speed, because it was programmed properly
* It has good multiplayer - up to 4 people on one computer, compared to Quake which had acceptaable net play but is now running out of players as people move on. Plus net games are harder to set up then getting a friend to sit down next to you to play against
* For single player, shareware Quake only has 1 episode and no bots, so will get boring after a while. However with SFE you could play for as long as you want against up to 3 computers.
* Plus there isn't much point comparing the two since they are completely different games

How can you be so wrong about Quake!!!!!
-Although a little slow, It does run at a playable speed on a SA RiscPC,
-There are hundreds of Mods & Bots for Quake out there, just take a look on http://www.fileplanet.com
-And without Quake we would be without such materpieces (on Windoze) as Half-Life

All Hail Quake!
(and the almighty green penguin from Dorchester)

hey, why wait for these "wonderful" OpenGL games that require an Omega or better when you'll soon be able to get EMD & Overcast 2

<quote>Here, here. So Lee, when can we expect a miniGL driver and fully working, highly-optimised 3D engine so that we can have games that are "worth" buying on the RO market? </quote>

hey, why wait for these "wonderful" OpenGL games that require an Omega or better when you'll soon be able to get EMD & Overcast 2

Well people can expect a MiniGL driver just as soon as I've finished other stuff and I get a machine to develop it on.....You could be waiting a bit

Seriously I think it's better to wait and see what Microdigital have in store for Lightning and then perhaps trying to be compatible with that.

One nice thing about OpenGL is that it's a scalable state machine API. Basically any machine should be ablet o run a game written using it (assuming the software driver is fast enough) because you can easily turn off texture mapping, shading etc at will.

Anyway, Mark's right, EMD and Overcast 2 are going to be great (we all hope)

<quote>Here, here. So Lee, when can we expect a miniGL driver and fully working, highly-optimised 3D engine so that we can have games that are "worth" buying on the RO market? </quote>
hey, why wait for these "wonderful" OpenGL games that require an Omega or better when you'll soon be able to get EMD & Overcast 2

"Worth" buying?
Is that you Nathan?
Or are you quoting the 'must be 3D' quarter?

I was being sarky.

hey, why wait for these "wonderful" OpenGL games that require an Omega or better when you'll soon be able to get EMD & Overcast 2

:-)
They're more likely to be good now that VOTI have got additional support in addition to management and main programmers!

And that's decent management!

Andrew, I don't know if my emails are getting through to you!

Andrew