I thought that was what I was doing earlier on.

It very much was. Notice the last two words before my (abbreviated) moniker. Every time you see them, I mean them.

I am the kind of person who likes to argue things through. Presenting a counterpoint to a suggestion doesn''t mean I''ve dismissed it completely.

Let''s go back:
I said:

quote: Origninally posted by liquiddark
It''s simpler to start with. But it''s a mess to try to fit a system build bottom-up into this kind of top-down idea. To some extent, this replaces the component system I dread so readily with an attribute system instead.

Don''t get me wrong; your +/- curves idea made it into the datafile regarding "attribute sliders". I''m not throwing the idea away, just trying to make sure that the grounds we''re basing our conversation on are the same for both of us

The beauty of your system is, within a particular "epoch" (I imagine they''ll arise naturally as players build towards excellency in a particular area), players can screw around with their units without spending much time in the lab - most of the time, they can be out exploring and encountering. But at a slightly deeper level, one of them (ideally, both of them) is going I need to build a gun this bugger just can''t handle . So we go back into the deep lab, trying to twist the knife of technology ten ways ''til Sunday.

Same type of thing goes for Impossible Creatures and Empire Earth: once I''ve gotten my hands on them, I won''t stop until every joyful secret is wrung out of their cold, dead plastic......*hem*...yes, well, you get the picture, I hope

It''s not an either/or thing with me. It''s a "these people are smart and haven''t been at this same d*mn thing for a decade - maybe they''ve got some light to shed..." thing.

In future, I shall attempt to make my stance clearer

thank you .
ld

quote: Original post by liquiddark
Kylotan:
I *think* you''re suggesting an AC clone with specialized sides. Two basic objections:
1) Static parts aren''t directed-design goodies. They make for some novel gameplay, but they don''t empower the player, which is my intention.

2) The civilizations I envision change as a result of the player''s tech focus - a player building fast units automatically becomes a Nomad civilization, for example. These civilizations have a few associated dongles, but mainly influence the player''s tech aptitudes, creating a feedback loop wherein the better the player is at a particular thing, the less flexible they are overall.

I was just pointing out that there is a game out there that doesn''t use a fixed set of static combat units. By looking at things on a component level rather than a unit level, you can still keep some semblance of balance, without restricting the player. How you implemented the creation of such components would be up to you, but it would be difficult for the players to discover something totally ''new''... how would you represent it in-game, for example?


[ MSVC Fixes | STL | SDL | Game AI | Sockets | C++ Faq Lite | Boost ]

quote: Original post by Kylotan
I was just pointing out that there is a game out there that doesn''t use a fixed set of static combat units.

Ok, cool. I wasn''t sure where you were headed with the idea originally. I''ve played AC, and it is very much a burr in my side, along with War 2100 (a 3d component-based RTS) - I''m not yet sure of the technical aspects of going it one better, but from a player''s POV the component system only satisfies me *only to a degree*. I''m interested in a much more organic flow from epoch to epoch, and I''m very much not interested in "sequel parts" as a part of gameplay (and I''m sure you can name them: hovertanks I thru X or whatever) - in liquiddark, this is abstracted as part of the ongoing cost & aptitude flows.

quote: How you implemented the creation of such components would be up to you, but it would be difficult for the players to discover something totally ''new''... how would you represent it in-game, for example?

This is a question that I wrestle with every time I confront the design. It seems that the only way to know that this could be done would be to build it. Impossible Creatures (thanks again, Tacit!) gives me some hope that it''s a workable issue, and the Extreme Programming methodology seems ideally suited to creating an art-generation engine, but I really don''t have a good answer to this question. Yet.

One more elaboration, for good measure:
I''m not talking about a scripting system where you get a set of attributes like "firepower" and "armor". I''m talking about a physics VM running the whole world, including every unit at the component level. Those components are coded in a language describing the physical effects, and those physical effects then act upon other objects within the world according to the rules of the VM simulation. That is the level at which my mind operates when I''m talking about the player creating something the opponent didn''t even expect. "Firepower" doesn''t represent a single characteristic - it embraces any destructive power, from slugthrowers and knives to lasers and antimatter explosives; likewise, "Mobility" isn''t a single thing - it embraces speed & maneuverability in all their myriad manifestations; etcetera. Essentially, the granularity of the simulation is intended to be such that it is clearly on the other side of the component/systems line: a component-based engine uses large, discrete, premade blocks to affect the balance in-game, whereas a systems engine relies upon the actual operation of a "built system" to render the effects of unit mods on game balance.

If I haven''t explained myself clearly, I''ll gladly try to clear up any confusion.

thank you.
ld

I think the "technical aspects of going one better" will be the problem here. In fact, I think moving to a ''system'' approach from a component approach would be a very big and difficult stride to take.

In particular, I find that making this system work at all would take a hell of a lot of work: making it work within a game, and making that game playable and interesting will be near to impossible. Good luck though

And how does Extreme Programming lend itself to art-generation engines more than any other type of programming?


[ MSVC Fixes | STL | SDL | Game AI | Sockets | C++ Faq Lite | Boost ]

quote: Original post by Kylotan
In fact, I think moving to a ''system'' approach from a component approach would be a very big and difficult stride to take.

You''ll get no argument from me. If I knew exactly how to make it good and fast you''d already have seen it on store shelves. I''ve never shied away from the fact that this isn''t necessarily something possible right now, but I don''t think it''s so far-fetched to think technology will reach a point where it becomes possible, with a smart set of programmers and artists working together.

quote:
In particular, I find that making this system work at all would take a hell of a lot of work: making it work within a game, and making that game playable and interesting will be near to impossible.

From which viewpoint? Programmer (in which case, again, I agree in principle if not quite in severity) or designer (in which case, I don''t)?

quote:
And how does Extreme Programming lend itself to art-generation engines more than any other type of programming?

I consider XP to be a project-design philosophy rather than a method of programming persay. XP means, essentially, tying your ongoing progress directly to an increasingly clear design, rather than pulling a 600 page doc together and then fixing it up a little at a time. To my knowledge, nobody has a good solid idea of how to generate an arbitrary set of art assets based on a parameterization of the problem; again, Impossible Creatures gives me hope.

Having said that, using an XP methodology you start with a best-guess design, and then you and the artists work together throughout the entire process, feeding into one another, trying to write the ''stories'' involved with art creation as you go. Better, using XP means that both sides have an increasingly clear picture of the emerging software, so that at a certain critical point your art department can start pruning and streamlining the project''s art asset database to reflect much more exactly the core asset requirements of such an engine.

thank you.
ld

Ah, game balancing, one of my more favored topics

I''m probably the only person I know of that doesn''t necessarily believe in the concept of balancing your forces. In some ways, the whole crux of fighting wars is to see if your forces are stronger than your opponents. Wars are almost never fought between even sides. Think of the great wars fought throughout the ages, and most of its memorable ones have been lopsided (Greeks vs. Persians, America vs. England, Vietnam vs. US, Afghanistan vs. Russia, etc etc).

Games are so entrenched in the idea of balancing out unit forces and countries, that I think that the opportunity for great gameplay is being missed. Now, here''s the really prickly question to be asked. Why do we want games to be balanced? So that we can win and tell ourselves, "the best man won?".

I think games have to elevate themselves beyond this...as to me, that''s sport, not a game. Perhaps the goal should not be so focused on winning as it is about education. To enlighten the player on certain qualities. I mentioned in another post that I was thinking of setting up a scenario where the only way to ultimately "win", was to lose. And it was up to the player to determine how to "lose".

The scenario went something like this: The player was a commander of a division, and his commanding officer told him that he would have to defend a certain area. As the campaign progressed, it becomes apparent that the enemy is vastly superior, and is on the virge of encircling the players entire army. The player could be faced with two primary strategems. On one hand, he could try to break through and punch a hole in the enemy line...the other option would be to hold a pass (or some other vitally important area). However, to hold the pass would require the player''s unit to be abandoned...and essentially be left to death or capture...but their defense would buy the rest of the army time enough to escape. But the choice is to be made by the player.

Now, how many games even tempt you with the idea that you are supposed to lose a battle to win the war? Why does everything have to be balanced to give a player a fair shot at "winning"? Game balancing is wrought with subjective comparisons and even if a game comes out balanced....players will quickly learn the "formula" that the developers came up with and use them to their advantage. So my answer is....don''t come up with a formula. Don''t try to force balance with units. Come up with scenario ideas that stress the importance of more than winning...which in itself is a vague term. Let the players come up with their own formulas for creating units and duke it out themselves. The problem lies at the core of the player...the feeling that he needs to win to feel good about himself. I think this very idea needs to be challenged, and substituted with something else that will also make the player feel good about himself. In life, we learn about virtues like humility, sacrifice and duty, and yet there are so few outlets to actually practice them. Maybe it''s possible that games can actually help teach these in a manner that makes it fun and rewarding in its own way.

Unfortunately, I don''t think you know what the term "game balancing" means. I am too tired to explain it without smacking you upside the head. Maybe another time.

Dave Mark
Intrinsic Algorithm Development

Dauntless,

I''ll dive in so InnocuousFox can get his rest. Game balance is for the purpose of multiplayer play. In single player, scripting and enforced conditions can lead to the kind of circumstances you describe, and while these are perfectly good and rather interesting plot ideas, they do not apply to multiplayer. The idea behind balance in multiplayer is that although each side may have strengths and weaknesses, a skilled player of one army will have neither an overall advantage over nor a disadvantage compared to a skilled player of another army, no matter which army the two sides are playing. An unbalanced game is one in which the choice of army plays a major factor in the outcome of a match. And if this is what you call a "sport," then so be it... it is what I am looking for in any game, not just RTS. I don''t want the game to be decided by who gets the best gun, or who gets the best army, or who gets the best side of the map, or any other factor than skill and cleverness. But that''s just me.

Is my point now well taken? Or must the Fox throw his hat into the ring, tired though he is?

---------------------------------------------------
-SpittingTrashcan

You can''t have "civilization" without "civil".

quote: Original post by liquiddark
From which viewpoint? Programmer (in which case, again, I agree in principle if not quite in severity) or designer (in which case, I don''t)?

It seems like a system that would lend itself to massive imbalances in favour of whoever found something really useful first. This would obviously be imbalanced (in the way that SpittingTrashcan mentions above - it wouldn''t necessarily be an issue for single player games). The benefit of discrete numbers of units, whether explicit (most games) or implicit by the possible combinations (Alpha Centauri) is that all possibilities are predictable and can be allowed for in the game balance. I couldn''t see that your system would lend itself to this analysis... perhaps you can think of something that I cannot however, as you know your design better than I do.


[ MSVC Fixes | STL | SDL | Game AI | Sockets | C++ Faq Lite | Boost ]