Advertisement

RTS Passive Abilites versus Active abilites

Started by August 09, 2005 06:57 AM
85 comments, last by Sandman 19 years, 5 months ago
Quote:
Original post by Daniel Miller
Quote:
Original post by Telastyn
Quote:

Playing general would be very boring... don't you think?


Given the number of panzer/fantasy/* general games.... I don't think so.

Gameplay is making interesting choices. Manually selecting and executing 'proper abilities' isn't a choice, it's tedium.


You are in the minoritry by hating fast-paced RTS.


Well thank you for that wholy inaccurate commentary.

[edit: inaccurate because I don't hate RTSs that is. I just disagree that popularity or 'observer appeal' has anything to do with a good design. I also disagree with the sheer divinity that you seem to hold the game, they're fun, but not exactly a paragon of game design [imo]. To me, FPSs almost always provide a better arena for dexterous competition, where skill determines victory rather than environment exploits. Similarly, TBSs almost always provide a better arena for strategic competition.

IMO of course.]

[Edited by - Telastyn on August 19, 2005 7:54:46 AM]
Some people never get to like RTS because it's harder to watch yourself get better. In a PFS, you can watch as your kills per round increases, but in a RTS, you'll get raped every time until you get good.

BTW: "Exploiting the environment" is no different than using anything else to your advanage (physics, weapons, etc.). The problem is when it is impossible to stop, which isn't true in this case.
Advertisement
Except that no FPS programmer will 'remove' gravity. A RTS programmer would make the AI better if they could.
Quote:
Original post by Telastyn
Except that no FPS programmer will 'remove' gravity. A RTS programmer would make the AI better if they could.


But in this case it would change the game for the worse (you realize that decreasing player involvement does not mean increasing player fun?).
Quote:
Original post by Daniel Miller
Quote:
Original post by Telastyn
Except that no FPS programmer will 'remove' gravity. A RTS programmer would make the AI better if they could.


But in this case it would change the game for the worse (you realize that decreasing player involvement does not mean increasing player fun?).
I don't know about you, but I'd love it if game developers would remove my involvement in the tedium and instead let me focus on actually doing something. This is the primary reason I don't like MMOs right now - most of the time is spent grinding or walking and neither of those is very much fun.
"Walk not the trodden path, for it has borne it's burden." -John, Flying Monk
Just wanted to chime in Daniel. While I agree with some of your points ... I have to say that I have never met a single person (in person) who would agree with your idea that improving the AI would hurt the game ...

Now people will argue about what is an improvement. But the main thing I mean is the removal of AI reactions that consistently cause players to go "What the Fuck!?!@#$" and "Why can't this stupid AI do W@#$%@$ like it does in that other game ..."

For example, many games used to not have proper "agresive" "defensive" "hold posision" and "hold fire" choices for your troops ... and NONE currently do those activities perfectly. A perfect UI/AI would be one in which each player could correctly command their troops to do what they "want" them to do.

The reason we complain about RTS AI exploits is because they are 100% artifical frustration causers. The fact that no human, not even experts, can command their troops as well as if they we're a trained squad yelling verbal orders is 1 piece of evidence. The fact that the average, non expert players constantly find themselves frustrated because the experts can defeat them via means they do not think of as the "game" (not being able to make your unit stop acting stupid is NOT what MOST people enjoy about RTS games).

Nearly all of my college RTS playing friends love/loved Starcraft. Nearly every one of us dislikes BroodWars, because it opens up even more avenues that play on the AI / control problems of the game (medics for the humans act AUTOMATICALLY - while other abilities that would kill them and the troops they augment don't - which is very overpowering for the average player) (clocked units and lurkers are the newbie killer, and they don't enjoy how they HAVE to go down the EXACT right path of defense to survive for 6 minutes against a moderately good player).

Starcraft actually has more interesting gameplay when played by 2 bad players than good players .. because bad players are THINKING up things to try, and suceeding or failing and REACTING to them ...

Good players are executing known patterns and plans against likely enemy actions ...

Wasn't it true that your most enjoyable starcraft moments are likely the ones where something new / cool was tried against you? And better still, when you struggled your ass off to think up a response before it was too late? Notice I say "first time" and "think up" ... not, click click, drag, click click.
Advertisement
Quote:
Original post by Xai
Just wanted to chime in Daniel. While I agree with some of your points ... I have to say that I have never met a single person (in person) who would agree with your idea that improving the AI would hurt the game ...


I'm actually pretty normal when it comes to fixing AI. There are very few Broodwar players that I have met that want to change it (except for two exceptions, which will be mentioned shortly).

Quote:
Now people will argue about what is an improvement. But the main thing I mean is the removal of AI reactions that consistently cause players to go "What the Fuck!?!@#$" and "Why can't this stupid AI do W@#$%@$ like it does in that other game ..."


Chances are that the "other game" flopped. When you make the player play the game rather than the AI play it for you, it does get harder to make your troops do what they want, but it also makes it much more competitive. Imagine a FPS with autoaiming. Sure, it's doing what you want it to do (aim at your opponent), but now every player is just about as good as every other player, and it becomes very difficult to actually improve.

Now, about those two exceptions:

Scarab AI. Without going into too much detail, it's random. Sometimes, it will do exactly as you say, other times, it won't do anything at all. That's a big no-no. As long as it's predictable, its ok, but don't make something that can cause a player to randomly lose.

Dragoon AI. They sometimes freeze up for no reason, unable to shoot or move. This one isn't as bad because you have some control over it (you can spam Hold Position which gets them unstuck, but it's still something that randomly happens that can turn the tides in a battle.

Quote:
For example, many games used to not have proper "agresive" "defensive" "hold posision" and "hold fire" choices for your troops ... and NONE currently do those activities perfectly. A perfect UI/AI would be one in which each player could correctly command their troops to do what they "want" them to do.


No, that would be a nightmare! :P

Like I said, you want players to to play the game. Too much hand-holding and the game becomes bland and easy.

Quote:
The reason we complain about RTS AI exploits is because they are 100% artifical frustration causers. The fact that no human, not even experts, can command their troops as well as if they we're a trained squad yelling verbal orders is 1 piece of evidence. The fact that the average, non expert players constantly find themselves frustrated because the experts can defeat them via means they do not think of as the "game" (not being able to make your unit stop acting stupid is NOT what MOST people enjoy about RTS games).


The reason that so many people stick with Broodwar is because they see how well the proffesionals control their troops (it doesn't even make sense how good they are), and they constantly strive to improve. Broodwar is a game where a better player will beat a worse one 9 out of 10 games, so there is a reason to always improve (even at the professional level players are still getting better, which is a constant source of amazement).

Quote:
Nearly all of my college RTS playing friends love/loved Starcraft. Nearly every one of us dislikes BroodWars, because it opens up even more avenues that play on the AI / control problems of the game (medics for the humans act AUTOMATICALLY - while other abilities that would kill them and the troops they augment don't - which is very overpowering for the average player)


I'm not quite sure what you are trying to say. It's too hard to beat it? It is much easier to control Lurker/Zergling or Muta/Zergling than it is to control Marine/Medic. One mistake with "M&M" and you have lost your entire army, which can't be said about Lurkerling or Mutaling.

It's too hard to control? Yes, it's very difficult to control it effectively, though it's one of those things that -- once you get very good at it -- allows you to do some very creative things with them (dodging lurker spines, constantly moving your front line back so Zerglings can't get many hits in, etc.).

Quote:
(clocked units and lurkers are the newbie killer, and they don't enjoy how they HAVE to go down the EXACT right path of defense to survive for 6 minutes against a moderately good player).


The players that do quit Broodwar are almost always people who get frustrated early on because they lose every game. You have to realize that's the nature of a truely skill-based game. Once you play and improve (and you will improve), you realize that those things really aren't so hard to beat.

Quote:
Starcraft actually has more interesting gameplay when played by 2 bad players than good players .. because bad players are THINKING up things to try, and suceeding or failing and REACTING to them ...

Good players are executing known patterns and plans against likely enemy actions ...


When I watch to bad players, there are so many moments where I go, "No! You clearly shouldn't have done that!".

Games from good players are much more entertaining to watch, because they have better unit control, and can make much larger numbers of units (bigger, more skilled battles).

Good players come in two varieties:

First, there is the "memorize the most effective counters and use them" type of player. Their play is of course effective; they generally use the most proven strategies and counters, and there isn't a problem with making the smartest choice. Their games aren't always the same, however, because the outcome of every battle is so important, and one player might control his troops more effectively in a given battle, which will completely change the game.

The other type of player generally tries unusual and creative strategies in the hopes of flustering their opponent, who probably doesn't know how to beat it effectively. These players are generally hit-and-miss: if their opponent can figure out how to beat their strategy in time, they will probably lose unless they can safely transition to a more effective strategy. Otherwise, they win.

Both players are fun to watch in my opinion.

Quote:
Wasn't it true that your most enjoyable starcraft moments are likely the ones where something new / cool was tried against you? And better still, when you struggled your ass off to think up a response before it was too late? Notice I say "first time" and "think up" ... not, click click, drag, click click.


My favorite games are games that I eitther play a very solid game all the way through, or games where I try some insane strategy and end up winning.

Both are very fun (you can't tell me that demolishing a standard Terran slow-push with Zealot/Dragoon isn't fun! :P).

Earlier, somebody mentioned the 'strategy' of sending in a hard-to-kill unit so the units of the opponent would automatically attack it and then the opponent would have to do a lot of micromanagement to get his/her units to attack the weaker units that came charging in after the hard-to-kill unit.

To me, this sounds like exactly what you're arguing against, Daniel Miller: The AI is playing for the opponent and making very poor choices so the player has to do a lot of work to make up for it. A better AI would (IMO) either target the weaker units as they come in, or would have some options the player could configure that would let him/her set his units anywhere from 'do nothing unless I command it' to 'attack every enemy you see' and it would be up to the player to set his units to not attack the hard-to-kill unit first before the frenzy of combat starts.
Maybe to balance out the customizable AI, if the units were set to 'attack weak creatures first', they might hesitate before attacking an all-strong-unit army. Thus, if you have the wrong setting, they won't work as well.
"Walk not the trodden path, for it has borne it's burden." -John, Flying Monk
Quote:
Original post by Extrarius
Earlier, somebody mentioned the 'strategy' of sending in a hard-to-kill unit so the units of the opponent would automatically attack it and then the opponent would have to do a lot of micromanagement to get his/her units to attack the weaker units that came charging in after the hard-to-kill unit.

To me, this sounds like exactly what you're arguing against, Daniel Miller: The AI is playing for the opponent and making very poor choices so the player has to do a lot of work to make up for it. A better AI would (IMO) either target the weaker units as they come in, or would have some options the player could configure that would let him/her set his units anywhere from 'do nothing unless I command it' to 'attack every enemy you see' and it would be up to the player to set his units to not attack the hard-to-kill unit first before the frenzy of combat starts.
Maybe to balance out the customizable AI, if the units were set to 'attack weak creatures first', they might hesitate before attacking an all-strong-unit army. Thus, if you have the wrong setting, they won't work as well.


Yes, the AI is screwing the player over, but the player can manually tell it to stop, so I don't see how this contradicts my other views.

I don't like the idea of customizable AI (it's too big of an abstraction), but the idea of a "don't do anything" command sounds perfectly good to me, though it's uses are limited if you think about it (but there certainly are uses for it, such as hiding on a cliff until your opponents miners come by, and then attacking once they all arive).
Quote:
Original post by Daniel Miller
[...]Yes, the AI is screwing the player over, but the player can manually tell it to stop, so I don't see how this contradicts my other views.[...]
You can't tell the AI to stop, you have to replace the AI's order with your own and waste time fighting it. The AI is playing for you and you have to fight it while the other player defeats you. Surely you would agree such gameplay that revolves around fighting the AI that is supposed to help you isn't much fun?
"Walk not the trodden path, for it has borne it's burden." -John, Flying Monk

This topic is closed to new replies.

Advertisement