Hi
Some views from me on a commercial game
I do not want to have another flame war about it just some constructive ideas of how it might work and how it can be a accomplished to some extent.
So to point out that this discussion is not about the game it self but rather to point out how certain things can be done in various forms in FPSC to have a quality game that I am sure well have a good following.
So lets get started.I would like to start off with level design.
Level Design
Straight off every one would say that this is nearly an impossible task to have a good looking level as any commercial game of any type.Well truth be told it is not all that hard to have some good looking functional levels as any other commercial game it does require prior understanding of level designing and a good knowledge of modeling in General.It is important to know that FPSC has limitations even a modded engine would still have moderate limitations.It is how one works around those limitations that gets you from point A to point B via point C.
A good example would be to design your levels in a 3D model editor I have shown that this can be done if you follow some simple rules.Time is however an issue here it would require a serious amount of time doing this allow as much as a 2 to 3 weeks per level depending on detail used even longer perhaps.Again to point out we are not planning to make another standard FPSC game here we want to push boundaries and limits.A single level might contain as much as 500 to a 1000 meshes 70000 plus polygons.But is not a major problem if you make use of culling when ever possible.
The rule of thumb with level design in general is to keep object at low polygons the further away it is form the player.So in other words if you have a house behind a wall where the player will only be able to see the front and is unreachable there is no need to have a solid mesh there are numerous parts of the mesh that can have inverted normals.Simple stuff like what I just mentioned helps a lot when building a level to reduce performance degradation.Level design plays the most important part in making or breaking performance.It also contains the most wastage of polygons.So really if you can help it to render a wall only partially or make use of a few less polygons even 300 less polygons means one more dynamic entity for example.
For one FPSC tries to manage this by culling it is still limited and at default FPSC will always have the other side of the wall there it might not be rendered but it is likely those polygons will still be counted in the end.
With proper managed static meshes and dynamic meshes one can truly create a good looking level.I think what puts most people off from designing a 1000 meshes is having to make the same wall perhaps 6 times over with different parts of the mesh inverted depending where it is in the level.This would go for each level you are doing so it is a task that can get repetitive.As I have said before before but I can not stress enough that the level design makes or breaks the over all look of a level from a polygon perspective.Some people will argue that other parts make or break again I do not want any misunderstandings and want to make it clear that I am not referring to level design but rather the way you design a level and with what you do.
FPSC has allowed numerous things that are used in other games for example level geometry CSG for one that makes it extremely easy to add doors windows fences any thing you can imagine.There is no excuse why you can not create a commercial looking level.A while back I was thinking about how I would make sunshine look good in FPSC and terrains.I found that using a terrain mesh made collision nearly impossible and by simple following the floor segments rule of mesh may not have inverted normals and must comply with the 100 x 100 x100 dimension law I was quickly able to create a terrain with in my model editor and export/import a terrain made of 100 x100 dimension mesh blocks that would make up my my terrain.
Your head needs to be screwed on correctly to make use of a 300 mesh terrain but the benefits was far reaching it allowed for proper entity and player collision by following some simple design laws.The sun light left me with another problem of how to get it working correctly I have never tried it but I have suggested it to a fellow forum member that to use a large 500 unit low polygon cone mesh angled at the direction I wanted the sun to shine with apply a simple shader to the cone it was rendered invisible and did not have a shadow.I did however try this with getting directional light mapping into my level and the results are promising it however lacks no collision so a player will be able to make contact with it in a game so it will have to be out of player and entity reach.So there the idea further evolved and quickly discovered I could make directional square circle triangle pretty much any shaped directional light mapping you could think of with no engine mod.
So the point I want to get across on the level design front it can be done will a little effort and understanding and time.If you say it can't be done prove me wrong....
Entities
Entities are fun to work with and a lot can be done but as for level design a lot more effort needs to be put into it.That needs having several entities doing different stuff in your game.If you understand how the animation slots work and scripting you can have a character skydive off a building for all we care by using a specific animation slot allocated for strafe for example.
Or a entity talking to you before dying there certain animation slots that can be used for that.It means getting your hands dirty.Want a Zombie to break a door just before you reach it with a little work you can do it with no problems.Want a ninja on a rope why not?
In all my time here I have not seen any one taking the time to make specific animations for specific actions in their games?If they have, I have missed it.I would love to see that.There is no reason why we can not have same type of animations with a little compromise.Have a World war 2 Nazi Gunner on a street corner with some sort of a machine gun.
You could even go as far as have entities hide behind boxes and throwing nades at you or blindly shoot at u from over the top.If you can have smart AI in FPSC, perhaps we can make them look smart via the help of animating them smart.I am sure some like Jon or bond or EAI will vouch for that, it can be done but it will likely be limited to some extent.
There is likely much more that can be said on the subject but I think what it all really comes down to is how much are you prepared to put into it.I think the main problem is people are thinking about the FPS genre too board and want all the features all at once.But we would have to be realistic on what we have now and what might come in the future and narrow our search down to what we have right now.In the future we might be able to drive cars into breakable bricks walls that makes dents in the car we drive.We do not have that feature but it is not stopping you from creating a game with just as much success as any other game.
I see no reason why one can not make a game of commercial success in FPSC.Point period.Rolfy prove that point very well with his game I am sure if he wanted to and had more time on his hands we would have seen the first FPSC game that would have pulled in a few hunderd thousands dollars.
Kindly point me to which FPSC game rule says we can not do the same.
"In my weird politically incorrect hypothetically incoherent contradicting obscured world definitively maybe"
