A practical case of game design

Started by
4 comments, last by Yvanhoe 17 years, 4 months ago
Hello everybody. I often see here calls of "game designers" with lots of great ideas looking for a programmer to code them. I thought it would be a refreshing change to reverse the roles... So here I am, I coded this : Dwarf Warrior, a few weeks ago. I had the vague idea I wanted to do something of the form of a dungeon crawler, so this is where I ended after aprox. 20 hours of coding. Right now, the code needs some clean-ups, a lot of improvements can be made, but for now I find it funny... for a minute at least. So I think it may be possible to make an entertaining game based on this basic gameplay. But now I need ideas to make it evolve, to make it more interesting. * If you are a programmer, the code is in python, it is GPLed, do what you want (it is a bit messy however) * If you are a designer with ideas about how to make the game evolve, answer me publicly or PM me, I am willing to implement good ideas! What I don't need : * A storyline. Let's hook the player for more than 5 minutes before thinking about a storyline. * Irrealistic ideas. No, it won't become a MMORPG. * Graphical ideas. I really want to develop an intersting gameplay before wondering about the rest. Some leads : * I think it could become a fast-paced tactical game where positioning coud be important * AI is a subject that interest me, so if you have an idea that require complex AI behaviours, do not consider this as an inconvenient. One of the way to improve the game may be to have many different monsters with different behaviors. * Maybe the game would benefit from a limited multiplayer mode ? * The ability to dig, for monsters and/or for player may bring interesting strategies. * What kind of objects/effects would make the game more interesting ? PS : I tried to write this post in a correct english, but being not a native speaker, some obvious errors may have slipped. Please be indulgent !
Advertisement
Ultimately, I think that the game could be improved by adding in more of that "fast-paced" tactical gameplay you speak of, and then differentiating enemy types via AI to take advantage of the new tactical depth of the game. What would be interesting is, barring tactical changes to gameplay, developing and updating the game from the ground up with the idea that nothing is static, and, like you said, allowing players to dig.

More than just dig, allow players to find alternative ways around problems by dynamically modifying the world. Dig holes, build walls, fill in dirt gaps, trap enemies, sequester allies for your own use. Get rid of the static barries in gameplay and give most everything some degree of intuitive maleability. With a relatively small production like that, you have (ironically) far more leeway to experiment than people working with big, clumsy code bases.
::FDL::The world will never be the same
I think my longest game has been about 45 seconds. Difficulty might be a bit high.

Good idea for the forums, though.
all around the world today, a kilo is a measure;a kilo is 1000 grams, easy to remember
thundergunslinger :
It may seem difficult until you find the cheap strategy : flee for 10 seconds or something end then get into a narrow corridor, then you can use the pierce attack (shift) on ennemies at a slower rate than your action bar regeneration. The only tricky thing is to manage not to get surrounded.


Quote:Original post by Nytehauq
Ultimately, I think that the game could be improved by adding in more of that "fast-paced" tactical gameplay you speak of, and then differentiating enemy types via AI to take advantage of the new tactical depth of the game. What would be interesting is, barring tactical changes to gameplay, developing and updating the game from the ground up with the idea that nothing is static, and, like you said, allowing players to dig.

More than just dig, allow players to find alternative ways around problems by dynamically modifying the world. Dig holes, build walls, fill in dirt gaps, trap enemies, sequester allies for your own use. Get rid of the static barries in gameplay and give most everything some degree of intuitive maleability.


Ok, less things static. However I think that before this I need to have different ennemies. Right now the AI is finding the player with a A star algorithm, so it never loses it (I explain this by their sense of smell) but it is very stupid, the 100 monster often being killed in the same trap. Maybe I could make them afraid of their comrade corpses in an ant-like fashion...
What interesting behavior would you like to see for a dungeon monster ? I tinkered with a "goblin" idea where a goblin officer would use his grunts to elaborate some basic tactics.

Quote: With a relatively small production like that, you have (ironically) far more leeway to experiment than people working with big, clumsy code bases.


Yes, that is why I didn't want to spend to much time on graphics, story, etc... This is more a gameplay prototype than anything else.
Quote:* AI is a subject that interest me, so if you have an idea that require complex AI behaviours, do not consider this as an inconvenient. One of the way to improve the game may be to have many different monsters with different behaviors.

For the AI you could have the creatures have a fear level. The more creatures that die the more fearful they are. As the player kills the creatures their fear level increases (or for the area they die in traps). The creatures avoid fearful areas and objects. Over time their fear value will reduce.

The creatures also have a Hate value. Each creature contributes a constant value, so the more creatures in an area the higher this value is. When the Hate value is higher than their fear value (or at a threshold above their fear value), the creatures will move towards the player.

This should create mobbing behaviours that will ebb and flow as the player kills creatures, sets traps, etc.

Quote:* What kind of objects/effects would make the game more interesting ?

I would say a finer granularity to the game. That is the character moves in steps smaller than their bounding box size.

This will mainly give 2 benefits
1) It will seem more realistic

and

2) It will allow you to control the speeds of the monsters and player better.

This finer control will give a more tactical feel to the game.

For more ideas have a look at the "Rogue" type games (there are too many to list them. A simple Google should give you quite a few hits).
Quote:Original post by Edtharan
For the AI you could have the creatures have a fear level. The more creatures that die the more fearful they are. As the player kills the creatures their fear level increases (or for the area they die in traps). The creatures avoid fearful areas and objects. Over time their fear value will reduce.

The creatures also have a Hate value. Each creature contributes a constant value, so the more creatures in an area the higher this value is. When the Hate value is higher than their fear value (or at a threshold above their fear value), the creatures will move towards the player.

This should create mobbing behaviours that will ebb and flow as the player kills creatures, sets traps, etc.


I know that 'emotion values' are used a lot nowadays, but I am not sure it results in interesting behaviors : It attacks then while at range, suddenly flees then, when begining to be far away, comes back whatever its damage are. I mean, I agree, this may be realistic, but when looked from a gamer perspective, this is just bad tactics.

But thinking a bit over it... If the fear makes the enemy come and see a healer or if it makes enemies gather together or find a greater level creature to plan a more powerful attack, that may indeed be interesting.

That raises also the question : Is it better (more fun) to have multiple creature types each with a single behavior pattern or to have only one kind of creatures able to show a wide range of behaviors ?

Quote:
I would say a finer granularity to the game. That is the character moves in steps smaller than their bounding box size.

This will mainly give 2 benefits
1) It will seem more realistic

and

2) It will allow you to control the speeds of the monsters and player better.

This finer control will give a more tactical feel to the game.

For more ideas have a look at the "Rogue" type games (there are too many to list them. A simple Google should give you quite a few hits).


I think I will not implement a finer granularity immediatly, because
1) It adds a lot of complexity, therefore bugs, strange behaviors, a more complex AI
2) I think the tactical feeling of the game will be more present if the position of each element is seen as a discrete value.
3) The speed of the player right now is unlimited and the speed of the monsters is determined by a timer, so I have a good control over it. At the begining I thought that I should limit the speed of the player in some way, but I kinda like the feeling of control it gives. And being able to move faster won't give you such an advantage : if enemies block you, it won't help you pass through them.

About other Rogue games : IMHO they tend to focus a lot on the RPG stats-gaining side. I would like to make one that focus on quick tactical decisions. Maybe a character sheet with some stats would be a good addition, but should it become the heart of the gameplay ? I am not sure. Right now it is possible to make other weapons with different attack patterns. I have toyed with : a spear with a longer pierce attack but no slash attack, an axe with an overpowerful slash attack but only a one square pierce attack

This topic is closed to new replies.

Advertisement