Ashaman73

My game is based on generated levels too and I've spent some years now to get it right. As Krohm has said, use templates and addtionally different layers of "meaning".

As a layer of meaning I mean to build your level from top to bottom.

Level-Layer
At the first layer, let's call it the level-layer, you decide the general theme of a level (sci-lab, underground cave, space station).

Section-Layer
The next layer would be some kind of sections, a group of single rooms, a single area in your level. I.e. when choosing a space-station level you have a bio-lab, a mech-lab, a canteen, living section, a gym section, the command bridge etc. This layer is the base for a certain template set and monster set.

Template-Layer
Templates are finally your basic building block to create your actual level. The previous layers help you to choose consistent and meaningful sets of templates. A template can differ in size, all you need are defined links between different templates to "click" them togehter, then fill the level will some monsters.

Point-of-interest
You should add some points of interest to break up the monotony. I.e. in a bio-lab some kind of container with some alien in it. It must not have any function, but the player should stop for a short momentan to intake the scene. Addtional these points of interest help to navigate through the level by remembering these point-of-interest (ahh... there was a door near this alien)

Finally, you really can't answer the question if a level is interesting or not by just looking at the map. Try to get some alpha tester for your game to get a reasonable feedback. You, as developer, will loose quite fast the ability to measure your own game. You will find it quite boring, because you're able to beat it while sleeping and you already know each little secret.

I think that the problem of mages being hax starts earlier and can't be avoided by letting them get this power later.

There're two kinds of mages. The first one is the one who is able to do things normal people can't do (think of harry potter). The second one is a god in human form (think of wheel of time, dragonlance, etc.).

When taking a look at the god-mode mages most stories already have trouble of handling these kind of characters. The problem is, you need to continue to increase the power, but once you reach a hi-power level you need to get rid of them, else you loose your conflict (almost no-one is able to beat him) and therefor you loose your story. Often these characters are taken out of the active story or are crippled(aka nerfed) once they get too powerful.

An interesting alternative is to shift super powers, making the "common" hero equally powerful (Marvel & DC Universe). But these are not really powerful characters any longer, because the common human are just "cannon fooder" for the bad guys, they can't really beat any super hero without an other super power (item or character).

Therefor god-like mages in games are only feasable when you cut off the power once it gets too powerful. This happens when a single player game ends, but in a open game context (i.e. MMORPG) this is a real problem.

So, to be honest, the most interesting and feasable mage is the one who is able to do minor things other can't do, but not god-like things. I.e. by igniting some candles, tricking some opponents with a little illusion (only once), being invisilbe for 10 heartbeats etc. From a game perspective these mages are really interesting, because they let you experience a game from an other perspective without delivering some kind of hack by shifting game mechanism into a new direction.

Are there players, not RPG fans, but RTS fans, who think that RTS without campains is in bad situation?

It depends on your goals. For a multiplayer centric game, FPS or RTS, a campaign is more of an add on than a must have. When I want to play a multiplayer game I almost always skip any missions/campaign and jump directly into MP games.

Another cliche is the research tree and technology upgrades. Personally I think the development of a "race" could be possible even without research tree but maybe some people will be shocked when such are not presented.

A strategy game is not a tactical game. There you need to deploy strategically decisions and not only tacticals. With tactical decisions I mean direct countrol of units, combat , micromangement etc, with strategically decisions I mean resource mangement, production queue and research.

Research could be a powerful addition to a strategy game. A tech-tree makes long term planning really critical and you can out-rule an opponent through strategical planning of your path through the tech tree, even if your opponent is much better in a tactical sense (i.e. better micro-management).

I dont mean these things should be removed, but lets say a game is non-standard and is still RTS. What you can't miss in a RTS?

I think that all three categories - resource mangement, production queue(build order) and research - are important and delivers a minimum set of diversitiy. With only two your game play options are really limited and will get boring quite fast. When you want to get rid of one of them, I would sugguest to invest more in other parts of the game, i.e. a tactical component.

A very,really very, good book about 3d rendering is Real Time Rendering , but it is more an overview over current rendering techniques than a practical book about learning such things. My sugguest is, to get this book , read it and search the internet about techniques you want to know more about. It is really very complete, touching everything you need to know about realtime rendering.

When you really want to dig into NPR you should think about a few things:

1. It is like a magic triangle. Each corner represents a pure style: realism, abstraction, simplification
Examples:
- realism: a character looks like a real human
- abstraction: a character is just a blue square
- simplification: a character is stripped of all details, looks like a cartoon i.e. simpsons
Each (game) art style could be placed in the triangle, everything which is not really close to the realism corner is actually NPR.

2. From an artistic view, it is not easier to create a simple style instead of a realistic style, thought an abstract style could be easier. Some styles needed less work, but creating a good looking style is much harder when moving into the NPR realm. So, as coder, don't expect to find a good looking NPR style which will take the burden of finding an artist, this will not work as long as you don't choose a strict abstraction direction.

3. Moving away from the realistic corner will alienate more and more people. The amount of people who like or dislike your style depend heavily on culture. A comic book style or toon style is widely accepted in the western culture, thought it is associated with childhood. From a financial view, it is a logical choice to move to a comic style to minimise risk when you choose to utilize some kind of NPR.

4. NPR shading has its roots in improving image perception, these are not artistic roots but, more or less, technical one.
I.e. it is simpler for a child to recognise an image with less details, more contrast (color saturation), and outlines of the shape of an object. They don't use the toon shading to make it look cute, they use it, so that children can catch on what is happening. Going further, Gooch, the pioneer in NPR rendering, tries to improve image perception for technical reason (I.e. you want to improve the visualisation of X-ray images to accent certain medical features).

There're a lot more, i.e.
zelda windwalker(Wii)
team fortress 2(PC)
many MMORPGs (including WoW)
L.O.V.E.
many casual games
valkyria chronicles (PS3)
SWTOR
Borderlands
...

NPR games tend to polarize the gamer community and many have a bad attitude to NPR games. In the western culture everything which looks like a comic, is colorful or is exaggerated  is often viewed as childish. I think, that it needs some more time to be accepted by the majority of the gamer community, but the industry is already on a good way. Due to hardware restriction, NPR games were quite hard to be implemented a few years ago, but now hardware is more capable, the game community more open, and there're several NPR games with great success, so be patient

As far as I know, modern GPUs don't have any fixed-pipeline any longer (at least the parts which could be replaced by shader code). The driver use shaders to simulate them. So, no benefit from using the fixed pipeline over shaders.

And when adding a sample to the quee im playing it like this :

Vector2 sndPos(listener - sound); //sound : offset of sound in 2d space , listener : player's position in 2d space
sndPos.x /= 100.0f;
sndPos.y /= 100.0f;

alSource3f(source,AL_POSITION,sndPos.x,sndPos.y,0.0f);
alSourcePlay(source);

Hmmm. isn't sound the offset relative to the player ? Then you need to calculate your position like this:

Vector2 sndPos(listener); //sound : offset of sound in 2d space , listener : player's position in 2d space
sndPos.x += sound.x/100.0f;
sndPos.y += sound.y/100.0f;

Double post..

I have answered your question in your previous thread: http://www.gamedev.n...ost__p__4908635

Repost:

Are there any other systems to use besides the number inflation system?

Well, yes.
The basic,general 'game' concept is learning. That is, you need to have a challenge you need to overcome, once you learned to overcome it, you will be encountered with the next challenge. This game concept is often encountered in action games, or general speaking, in player skill based games. You need to learn your starting skills to overcome weak opponents, later on you encouter the next big challenge and need to master new skills etc. (i.e. bioshock, new plasmids/weapons are needed to overcome new challenges).

An other game concept is progression through collecting gear/equipement/runes etc. Many late-game MMORPGs like WoW or Guildwars work this way. Instead of gaining higher character levels, you need better skills/equipment to beat the next challenge (i.e. in Guildwars you reach the max level really quickly, but after that you need to find all those special skills).

Games without real character progression are often adventures (nowadays wrapped in a fps game). They are story driven, getting better weapons/skills etc. is not the game core any longer.

And eventually games with very low number inflation, but still with character progression, are often older CRPGs or PnP RPGs.


There're many ways to design (character) progression from a game design view. In fact you should think about your target audience first. Who do you want to play your game ? MMORPGs players, FPS player, casual gamers, adventure game players, yourself ? Each of these groups have certain expectations.

In short: you cannot simply just keep “adding stuff” to a game and have it retain its solid fidelity. There is a point with every game where adding one more monster, one more gun, one more playable class is too much. This may not entirely break the game, but it does degrade it. The quality and significance of each decision is slightly diminished, the identity of elements blurred.

Adding stuff to a game has multiple reasons, the majority of reasons have nothing to do with game design, more likely with business.

You can't compare games like chess with BF2/TF2, because chess is a game concept, a rule set, and not a product. To some degree you need to compare chess to a FPS or a RTS or a RPG and then you will see, that a FPS/RTS/RPG has not been greatly expanded since the first nethack,doom or warcraft game. Considering that games like chess evolved over 100 of years, a FPS/RTS/RPG game will be almost the same in 200 years, just the appearance changed, like a chess board from an other manufactory.

Take a look at half-life. This game was mostly a single-player game, but the real success was its multiplayer part due to great modding support. This game was one of the first one which demonstrated the power of solid game community. First, a multiplayer game is much more likely to generate a online community, secondly you need to feed your community to keep it at it. You know that gamers want to play, and that they will leave your game (more importantly your community) to join something else once the get bored. This is the main reason, publisher provided ongoing content:  to keep the people in their game until they release the according sequel

It is just standard business, like a payback card from your discounter. They don't want to directly influence the amount of products you consume (that is markting), they want to influence from where you get your products(when you buy it, buy it from us).

It isn't a secred, the industry want your money, not your happieness. So, when you want to make games which makes people happy, don't expect to get rich (play lottory, it is more likely to have success with it).

They sell 2d and 3d game engines on the net.

Gamedev marketplace
dexsoft
turbosquid
3drt

Have fun

I fear that you can't do much. In the real world people are trying for millennia to stop criminals, the current state of this undertaking should be known.

The problme in the real world is similar to the virtual world. Either you try to fight criminals by harsh laws and punishment, but this will always affect none or falsely accused people and make them unhappy, often other people abuse this laws to suppress others. Or you loosen the laws and punishment to make the people more happy, but this will increase criminal activities.

A virtual world is even more "critical", because unhappy people will just leave and play somehting else, so hard laws and punishment is not the way to go. So, you will not get rid of trolls, but you could at least soften the effect of them. As player I would always sugguest: ignore them, if this is not possible for you, choose something else to play.

chances are you don't want to have 30GB worth of game assets

In this case it could be an issue, but as already said, my requirements warning lights are full on.
That is, you try to build a mars vehicle before developing a rocket which could travel to the mars first(=> you don't have any real requirements yet).

Such a approach could be a nice exercise, but will become a burden quite fast when development a real project (=> most features without backup of any real requirement will get obsolete over time).

It is just my experiences and opinion...

you need at least two channels + 1bit for the sign of the 3rd channel if you stay in the same coordinate system

This is almost right. Some transformations on aras page(i.e. method 4&7) already consider the negative z axis. Atleast most of it, there're still some values which are 'unsafe' like -1 or almost near -1. But in general this resolution is enough to handle the normals in a gbuffer with only two channels. And, an other important advantage is, that you can use interpolation of normals without fearing many artifacts.

I'm hoping to have my game more about positioning, relying on your senses, paying close attention to your surroundings, etc and much less about picking up the Battleaxe of Unending Bonuses.

I think you have only the option to add some tactical combat simulation. This could give the player the chance to defeat an enemy or a group of enemies, otherwise too powerful for the character, by tactical finesse. It is like chess, roguelikes like tome4are going in this direction.

Otherwise the simple representation and turn base nature of a roguelike game are restricting the usage of certain human skills like reflex, fast movement, sound awareness etc. Adding sound to a game with a top down view will not help you to get a better sense of the scene, thought it could help to improve immersion. Whenever I play a fps game like BF3, sound helps a lot, but only because of the first person perspective which is mandatory to give you an immediate hint of events in your direct surrounding.

Awareness and fast reaction, which are important for many action games, not neccessarly only for FPS games, need atleast a realtime environment. You need to be aware of a certain situation, do a fast decision and need to execute this decision in a short period of time. This is not really useful in a turn based game.