RTS innovations

[Stark]

The fact that 'vision radius' still exists - instead of proper modelled vision with terrain LOS, weather effects and sensors etc - is another sign of stagnation. The only way to see something far away is to a) go there or b) build the Super Satellite Control Centre. You sure can't just go somewhere high and LOOK AROUND. You can scout in a direction and suddenly 'uncover' huge fucking mountains that would be plainly visible from the other side of the map.

At least 'pre-explored' maps have become more common, but it's a ridiculous jump from 'nothing more than 20m from a unit' to 'you know everything'. Ironically, the ancient 'Lords of Midnight' did it in a workable way ... decades ago.

[D.Turtle]

I think his point about initiative making the game 'player free' is fucking dumb. Yeah, having units that can do their job without the player directing every single action will result in nothing for the player to do! It's not like the actual game is still exactly the same and battles are simply less onerous - having units that respond autonomously or throw their own grenades or follow preset battleplans will make the rest of the game automatic... somehow.

Frankly, it's an example of what I've been saying all along: these people don't want to play a 'real-time strategy game', they want to play 'regular' RTS's with all the baggage that brings along. The idea that units doing the sensible/obvious thing is some kind of noob-assistance reveals a great deal about the mentality of RTS players. Other statements - like cancelling a project in progress should always refund resources - shows that they're incapable of imagining different games: heavens, in SOME games it makes sense, in others it doesn't! No, it shouldn't be penalised because of RTS gameflow convention. :Smile:

The point I was trying to make is the following:
A game is not good because it has or does not have all this automation/autonomy.
A game is good, because USING this (lack of) autonomy/automation it is fun, challenging, gives you a noticeable advantage of being more (or less) skilled than the other player, has replayability, has variety, etc.

I am NOT bashing the idea of automation/autonomy. I am saying: You have to present ideas which would make such a game fun - ie what is the focus of the game, show what the player has to do, etc.

To use SPC Brungardt's example: You can set the 'initiative slider' to full - which makes the computer do many things for you. It might be more efficient/more effective to control them yourselves, but at least you won't die without any chance at all. As you get better you can lower the initiative slider for certain parts of the game - or use the initiative slider to do the base building for you, while you control the units (or vice versa) - in effect having several sliders for varrous parts of the game (base building, small unit fighting, special abilities). In effect using he AI as a second player that concentrates on those things you can'T/don'T want to do.
This would make a game more accessible to new players, while still allowing 'pro' gamers their advantage.
Yes, this is applying the autonomy stuff to a 'standard' RTS, but face it - thats what most people want and are comfortable with.

I'll reply to some of the other stuff later (I do have to study a bit, exams are coming )

[RogueIce]

The fact that 'vision radius' still exists - instead of proper modelled vision with terrain LOS, weather effects and sensors etc - is another sign of stagnation. The only way to see something far away is to a) go there or b) build the Super Satellite Control Centre. You sure can't just go somewhere high and LOOK AROUND. You can scout in a direction and suddenly 'uncover' huge fucking mountains that would be plainly visible from the other side of the map.

At least 'pre-explored' maps have become more common, but it's a ridiculous jump from 'nothing more than 20m from a unit' to 'you know everything'. Ironically, the ancient 'Lords of Midnight' did it in a workable way ... decades ago.

The funniest thing I've seen in EaW is that when in Cinematic mode, you can see things ahead of you that your units can not see in their vision radius! Take, for instance, when a fleet of mine hyperspaced in to an occupied system. When they jumped in (Cinematic) I could see the enemy space station. But when it went back to the overhead view, I couldn't see it anymore because it was outside the sight range!

[Medic]

I think his point about initiative making the game 'player free' is fucking dumb. Yeah, having units that can do their job without the player directing every single action will result in nothing for the player to do! It's not like the actual game is still exactly the same and battles are simply less onerous - having units that respond autonomously or throw their own grenades or follow preset battleplans will make the rest of the game automatic... somehow.

Frankly, it's an example of what I've been saying all along: these people don't want to play a 'real-time strategy game', they want to play 'regular' RTS's with all the baggage that brings along. The idea that units doing the sensible/obvious thing is some kind of noob-assistance reveals a great deal about the mentality of RTS players. Other statements - like cancelling a project in progress should always refund resources - shows that they're incapable of imagining different games: heavens, in SOME games it makes sense, in others it doesn't! No, it shouldn't be penalised because of RTS gameflow convention.

I don't MIND RTS gameflow convention, I even pointed out that the only thing I really care to do is take out some of the insane APM out of a typical Starcraft game, for example. I'm not going all-out on the automation angle. I limited my own idea of how far it would apply to macro tasks, I never cared to adopt the automatic grouping shemes put forward by others. I LIKE conventional RTS's, I only really care to chip into the overhead.

At my BEST I think I might have cracked the 150 APM threshold on occasion, but veteran Korean Starcraft players are idling at about 150 and break 200 IIRC. (I've not played the game in a while nor followed it as religious as I once did, so my facts may be shaky) WHEN everyone has equally, stupidly high APM, you know what's amazing though? STRATEGY wins. Strategy and just striking that winning balance between macro tasks and micro tasks that are different on a by-game basis. (smaller maps, micro will tend to be more important, longer games though macro inevitably is a very accurate predictor of who will win, since in a long game, you can micro and win 2/3 of every battle but slowly fall behind macro tasks such as expanding and powering up)

[MKSheppard]

I'd like to see more smarter units. I can't micro every god damn thing, look at Company of Heroes, I'd like my units to at least be smart enough to run like hell if they don't have any AT weapons and a tiger comes rumbling by

[Companion Cube]

I'd like to see a degree of synergy between grouped complementary units; for example, when grouping a scout helicopter with a pair of gunships and issuing an "attack move" command, it'd be nice to have the scout actually peek over terrain obstructions with a mast-mounted sensor pack before letting the gunships go on ahead if the coast is clear, or lighting up whatever target is available if it's something the gunships would be interested in destroying.

Same thing could apply to tanks or APCs and dismounted infantry, or with different types of ship in a naval context.

[Dendrobius]

The thing is, there are some RTS games out there which are fun BECAUSE of micromanagement. Dawn of War comes to mind immediately. All the automation and smarter AI you guys are suggesting would make this kind of game suck. The best analogy I can think of is asking for autoaim in a tactical shooter because "Oh the guy I'm playing is an elite soldier and can of course aim!", with the net effect of lowering skill levels all around. Considering early to middle game for Dawn of War, you're only controlling at best 10 discrete entities be they squads, heros or vehicles, that level of automation is going to make the game boil down to "Attack + Move = thank you come again".

It doesn't reward players who pay attention and/or are sneaky and use the right counter against the units he's facing if those units are going to run away on their own initiative without the opposing player even noticing it. Why the hell should AI save my opponent's Fire Dragons when I smash some close combat units into them?

Am I playing against my opponent or am I playing against the game's AI?

[Stark]

I am NOT bashing the idea of automation/autonomy. I am saying: You have to present ideas which would make such a game fun - ie what is the focus of the game, show what the player has to do, etc.

Are you going to provide any evidence at all for these claims? You're saying 'more autonomy = game easy', even though it allows more time to concentrate on complex maneuvers and tactics, making the game HARDER particularly for RTS fans who aren't used to such thinking. You then claim that I have to invent a new way for the games to be fun, because somehow they won't be fun IN EXACTLY THE SAME WAY. Can you show me how units that automatically use grenades totally ruin an RTS and force a redesign?

This would make a game more accessible to new players, while still allowing 'pro' gamers their advantage.
Yes, this is applying the autonomy stuff to a 'standard' RTS, but face it - thats what most people want and are comfortable with.

This is such bullshit. Autonomy is not for making it easy or noob-friendly. It's for moving the focus from 'clicking like a speed monkey' to 'being able to orchestrate large complex battleplans'. You're just falling into the RTS-convention of thinking 'less insane clicking = easy for n00bs lololololol', and you're going to actually have to prove that. Games like this won't be EASY RTS'S they'll be ACTUAL 'real time' 'strategy' 'games'. As Shep says, in games like CoH it's pretty lame that units don't have more human-like autonomy (or standing orders for retreat etc), but you think that would make the game easy! At least you're not as retarded as Dendrobius, who thinks units running away when defeated is some superintelligent AI he must face in combat. Seriously - comparing a strategy game with unit AI to a shooter where you don't have to shoot is so utterly broken - RTS players need to realise that RTS-convention is NOT the only way to play 'real time' 'strategy' 'games'.

However, at least you're all providing even more examples for my previous statements about RTS players being resistant to change and conventional. Unit morale = 'am i fighting an ai lol'. Automatically throwing nades = 'lowering skill level'. RTS not based on paper-scissors-stone counters = bad... somehow.

[Edward Yee]

Personally, I'd love it if there were a standing order to the effect of "drive/ride/fly-by shooting," at least so that I can Attack-Move without having to micro it. (If the unit would be able to shoot on the move, I'd rather not have him come to a stop as soon as an enemy enters his range, keep firing until he can't, possibly follow said enemy, and then go towards his original destination.)

Then again, I'd also rather not have conventional (or "stereotypical") RTS resouce gathering...

[Stark]

I was just discussing this elsewhere, acutally: in most RTSs there is no penalty for moving and firing... so why don't the units constantly move? Why do they stop and shoot, like in Dune 2? Many RTSs have laughably slow projectiles, and attack-move micro is very effective. Why don't they do it themselves, like Cantabrian circles in Rome?

Wait, that'd make the game hell easy and you'd be fighting the AI way more than your opponent.

[Uraniun235]

I was just discussing this elsewhere, acutally: in most RTSs there is no penalty for moving and firing... so why don't the units constantly move? Why do they stop and shoot, like in Dune 2? Many RTSs have laughably slow projectiles, and attack-move micro is very effective. Why don't they do it themselves, like Cantabrian circles in Rome?

Wait, that'd make the game hell easy and you'd be fighting the AI way more than your opponent.

So if you ordered your dudes to a specific point, they'd just keep running about in circles? I don't even want to think about the traffic jams that could cause.

Unit pathfinding is in desperate, desperate need of improvement.

[Dendrobius]

I think almost everybody who's commented in this thread in support of more autonomy for AI units in existing RTS games either doesn't play competitively against other live humans in ranked games (1v1 Automatch in DoW for example), or really should be playing turn-based strategy games more than RTSes.

The entire idea of RTSes designed so far is to be able to think fast on your feet IN COMBINATION WITH ludicriously high APM. I'm sorry if your APM sucks and therefore you can't compete in ranked games against other humans or take on the AI, but asking for that level of autonomy without changing major aspects of RTSs in general NEGATES AN ENTIRE ASPECT OF THE GAME. This would be the same as asking for an aimbot in CS so that you can "concentrate on tactical movement", or asking for auto gearbox in GTR2 so that you can "concentrate on driving the line without distractions".

I was just discussing this elsewhere, acutally: in most RTSs there is no penalty for moving and firing... so why don't the units constantly move? Why do they stop and shoot, like in Dune 2?

Because most RTSs actually don't let units fire on the move, and forces ranged units to stop and shoot. Dawn of War and I think CoH improved greatly on this by enabling ranged units to fire on the move, WITH an accuracy penalty. This in combination with ranged vs close combat units actually makes micromanagement far harder and far more rewarding.

[Stark]

So if you ordered your dudes to a specific point, they'd just keep running about in circles? I don't even want to think about the traffic jams that could cause.

Unit pathfinding is in desperate, desperate need of improvement.

Actually, the Rome example was to try to head off the 'zomg too EEEEASY where is the fun' comments I expected. Such movement isn't necessary in a 'modern' game and wouldn't work anyway (the shots would just hit the other guys in the circle lol). But agility is useful militarily, and yet the units just sit there forcing you to nanny them through avoiding hellishly slow projectiles - basically producing the 'micro best = win' thing many people don't like. Further, this is a behaviour not a 'mode' - units wouldn't do it unless they were actually in combat, so stealth attacks would be more effective than usual.

Pathfinding is one thing: dynamic, goal-oriented pathfinding is going to be even worse/harder. However, simple 'group moves this way' evasion would work passably, still allow the micromonkeys to do their thing, and is only going to cause problems when the area is packed... in which case evasion is useless anyway. In a more realistic game with sensible ranges and unit density these problems wouldn't exist either.

[Stark]

I think almost everybody who's commented in this thread in support of more autonomy for AI units in existing RTS games either doesn't play competitively against other live humans in ranked games (1v1 Automatch in DoW for example), or really should be playing turn-based strategy games more than RTSes.

Wah you don't play like micromonkeys you are bad people, bad gamers, and should leave my precious games alone wah! How about this - you don't want RTS innovation, great. Get the fuck out of the RTS innovation thread and go back to your clickfest. See how that logic works?

The entire idea of RTSes designed so far is to be able to think fast on your feet IN COMBINATION WITH ludicriously high APM. I'm sorry if your APM sucks and therefore you can't compete in ranked games against other humans or take on the AI, but asking for that level of autonomy without changing major aspects of RTSs in general NEGATES AN ENTIRE ASPECT OF THE GAME. This would be the same as asking for an aimbot in CS so that you can "concentrate on tactical movement", or asking for auto gearbox in GTR2 so that you can "concentrate on driving the line without distractions".

Wrong. The idea of RTSs is to have 'real time' 'strategy', not make another paper-scissors-stone StarCraft clone. It's not OUR fault you can't see beyond convention and your precious ladders. Remember everyone: either you have high 'APM' and like micro or you have low 'APM' and would like to see the genre actually go someone and maybe even become more realistic!

Your examples are just getting stupider. Games like Metroid are in fact shooters with autoaim, and they're hardly super easy OR unpopular. They're just - get this - DIFFERENT. Something that clearly terrifies you. Hilariously, you actually seem wholly ignorant of the concept of a strategy game in which you are a high-level commander in realtime... rather than playing glorified chequers with brainless proxies. These games even already exist.

In short, I will repeat AGAIN for the reading retarded: 'real time strategy' is a term, not just a group of genre conventions. There is scope for a 'real time strategy game' that doesn't follow the Dune2/Starcraft approach.

Because most RTSs actually don't let units fire on the move, and forces ranged units to stop and shoot. Dawn of War and I think CoH improved greatly on this by enabling ranged units to fire on the move, WITH an accuracy penalty. This in combination with ranged vs close combat units actually makes micromanagement far harder and far more rewarding.

Er, most RTSs have 'attack move' where units move WHILE SHOOTING AT ANYTHING IN RANGE. So... you're lying now?

The best DoW feature in this respect was 'weapon setup' where many support weapons couldn't be fired on the move and required a brief 'setup time' to begin firing.

Micromanagement = rewarding. That's really all anyone has to know about you, and I have to wonder why you're even posting in this thread. Afraid your precious genre standard might grow up a little bit?

[SWPIGWANG]

I'll just whine about the random stuff I disagree with:

Remember, your opponent in a RTS game is not the enemy force, but the opposing human on the other side. One beats the opponent by out playing them on whatever level the game is suppose to be at, not by automation. Control is the most personal, intimate and variable resource in the game that can not be described by a build order or a strat. It is the one personal piece of the strategy puzzle that can't be dumped into an AI.

1. There is always too much micro.
It sounds like if one made the game where the player simply have to do less things, it would be better.

As a player playing the game, it is the player's job to take command of things. Control is what the player DO. As the complexity and number of events in a RTS game is staggering, there is always more thing to manage and the player that can manage better and faster has the advantage.

AI is not the solution, as an "game feature" where the AI can manage is a "game feature" that is irrelevent to the player. The game can simply be simplified to remove the feature all together. Instead of complex bases and complex base AI, one can simply have an handful of buildings doing everything. Similiarly, instead of auto-casting spell casters, one can simply remove magic and integrate abilities into base attack.

An game feature only has meaning if the player has to spend time managing it. A games micro requirements basically is a function of game complexity and time.

Incidentally, RTS game that don't have "build-in micro" often are the worst. If there is little need to micro battles for example, the player would simply push an attack move on the enemy and spend time in "boring" things like concentrate on squeezing tiny ounces of resources by systematic peon/building management. It is almost impossible to build a game where micro is completely irrelevent without simplifying it to dullness.

If there is nothing that the player needs to do for most of the game, the game probably would be speed up to the point where there is something to do other than stare at the screen. (very few RTS are played online at the slowest speed, even for the most micro-intensive ones)

I think the problem with micro isn't that there is too much of it, but it is often very boring and has clumsy interface, and that micro demands come in share spikes (short battles) as oppose to smoothly. The micro troops to outflank the enemy is one thing, to spend 30 seconds rebuilding the farm complex due to natural exhaustion is another.

Finally, micro is the only true random variable involved in a RTS game not designed for chance to be important. What seperates good players and average ones is the ability to apply the resource of player concentration at the right place in the right time.

Simply, micro is an integral parts of RTS games and can not be removed without altering what an RTS is. If there is enough time for the player to sit there and think of an strategy, it is really a TBS with timed turns. It is probably better to implement a "We-Go" TBS system if that is the wish.

2. There is no tactics: There is only Rock Paper Scissors
Well, one should not consider tactics outside player input. The only tactics relevent to the player is the tactics implemented by himself or his opponent.

Curiously, the people that complaint of this often complain of micro too. Guess what, most combat micro is about implementing tactics by input. People complain of games that require little micro or tactics as "tank rush games" like Heavy Tank rush in RA, while they also complain of games where tactics and control has massive effects on the game, like Starcraft where a professional player can kill a force twice its size controlled by a casual player due to knowledge and precise 300+ action per minute controls.

People want to eat their cake and have it too. They want tactics but they want tactics to happen by itself while they press a butten say "please don't let units be stupid." Like I said above, any feature that does not require player input is one that is irrelevent. When your opponent's units are under the same limitations, the effect simply cancels out.

Either you make player inputted tactics important, or you don't. If you want the latter, it isn't too hard, simply by having penalities for concentrate fire(as in RoN), small units relative to map (less traffic problems), no micro-required abilities, few map modifiers, no area-effect weapons(more micro to scatter), flat ranges, slow damages and all that. Now that would truely be rock paper scissor or simple plain swarming.

3. There is no strategy:
In a true "rush, swarm and attack move game" devoid of tactics, what is the decisive element in battle? Yes, this lameness called having and sending more and better units to the enemy when your enemy lacks them. That is pretty much what strategy is about: The logistics, the production and the grand scale macro-management of the economy.

Things like placing an artillery on a hill or flanking troops is tactics, not strategy.

Now one might argue that such "strategy" really don't take that much thinking, but that is untrue. Things like optimal build orders is anything but self evident. The only reason why it is not considered "intelligent" is because RTS generally revolves around a few very, very well studied maps with millions of games played on them and all possible decisions tried.

Now, to fix that problem is extremely difficult given physical limits and the player base. Lets look at the solutions:
a. More maps: However given that all fractions have to be somewhat balanced on all maps, and that all fractions have to be unique, the set of possible maps is tiny. Unbalanced maps are not accepted by the RTS community. (unlike wargaming/tabletop gaming) In anycase, most players would stick to only a few known maps anyways.

b. More complex interactions that can not be studied to completion in millions of games: This would be far to non-linear and abstract for most of the RTS community. Games like Chess or Go have no real life counter part in terms of interaction after all. In RTS games with fog of war, information is imperfect and a complex interaction game would not consistant ensure the better player wins as randomness added up with uncountered non-linearity means chance plays an important role.

But really, the lack of "strategy" is also reflective of the generally uncreative player base. For all the whining about "lack of strategy", most players never put in the effort of actually learning about all aspect of the game to form a effective original strategy anyways. Most just learns some "power strat" and complain of lack of strategy. Some simply complain that their APM is too low and don't even bother.

4. The (opponent) AI is dumb
The AI is not a human, however the AI is not allowed to use its greatest assets while has many artifical weaknesses. In the same way that the AI is not allowed to aim-bot in a FPS game, the AI in a RTS game have many limitations. The biggest limitation is that the AI is build before the game is released in most cases. They will have no benefit from the strategies that is developed after millions of games played by the entire gaming community. The other limitation is that AI can not do what a human can not, for example one can not program an AI in starcraft to win by insane SCV micro in small maps. The AI can do probably do it, but no human would want to play against it.

So we are removing all strength of AI and adding many weaknesses over a human. It is no wonder that they are of little challenge. We need an AI that can read forums and evaluate good strats, then we are talking.

---------

[Dendrobius]

Your examples are just getting stupider. Games like Metroid are in fact shooters with autoaim

Red herring. I ask again, how is the degree of automation you're asking for with existing RTSs different from asking for autoaim in CS or autogearbox in GTR2? They all take away some fundamental skill required for the game. High APM is a requirement of playing competitively in contemporary mainstream RTSs.

Hilariously, you actually seem wholly ignorant of the concept of a strategy game in which you are a high-level commander in realtime... rather than playing glorified chequers with brainless proxies

Afraid your precious genre standard might grow up a little bit?

When I want to play an RTS with micro, I pick up DoW. When I want to play a more realistic RTS, I go and play something like Close Combat. If I want to feel like a high level commander, I go play something like Steel Panthers. Different horses for different courses. You of course wouldn't understand that because most likely you would never have played such games anyway.

To transpose this entire argument to a different genre, what you're asking for is closer to a tactical shooter (FPS), when I'm saying that you should not be modifying the equivalent of Quake 3 (also a FPS) into a faux tac shooter because you're better off with a blank sheet of paper, and because the appeal of Quake 3 is completely different from a tac shooter.

[SWPIGWANG]

Are you going to provide any evidence at all for these claims? You're saying 'more autonomy = game easy', even though it allows more time to concentrate on complex maneuvers and tactics, making the game HARDER particularly for RTS fans who aren't used to such thinking.

I disagree with improved interface making the game easier (because your opponent is the human, not the interface), but I disagree that it makes it harder for well balanced RTS games.

Even today, good players try to execute complex maneuvers all the time. People try to manage muti-directional attacks, flanking, carefully timed distractions and everything in between.

You know, those insane 300+ APM of professional players don't spend only on dancing those zergings, but doing things like launching a three direction attack by having one central push with mutiple raids at the same time to overwhelm the opponent's response capacity.

Of course, there is limits on how well people can control such complex maneuvers easily with a clumsy interface. That is where strategy comes in. Things like scouting, tactics, economics are easy to learn and not so complex compared to the strategy of managing concentration.

The idea of RTSs is to have 'real time' 'strategy', not make another paper-scissors-stone StarCraft clone. It's not OUR fault you can't see beyond convention and your precious ladders.

You have no idea what is the meaning of the term of strategy in a game like starcraft. I just played a game on lost temple today, and let me tell you what strategy in that game is about.

The main, Critical piece of strategy in that games and millions like them is where to focus my attention and my micro-control. With my control, everything from combat to mining to base structure can be optimalized. However, I can only be in one place at a time. I have also very limited APM (around 90).

However I know that my opponent is under the same limitation. Given that the number of viable strats on the map is limited, and the out come of most unit interaction is well know (except when micro is involved), what is the decisive factor between players?

Yes, it is the application of control and the strategy behind them. Knowing when to control your troops for that critical battle or knowing when to managed your economy.

My opponent is limited by the same rules. It is the thing I am fighting against. I am trying to make him send his APM and his attention on irrelevent and unimportant things. I deceive him of my axis of attack. I distract him of important tasks of his economy. I surprise him with intense micro at a right point of the battle when he is not expecting. I overwhelm him with a flood of units with my spare control in the economy. That is what the game is about.

When blizzard made the game, the limitation in control is build into the balance consideration. For example, it is known that most players can only manage one reaver drop or less than half a dozen templars and the game is balanced accordingly. Control is a managed resource carefully made by the game design.

There is no "perfect" game of starcraft. There is always holes in people's control, and it is due to that, there is different styles of play, impressive replays and dramatic turnarounds.

Remember everyone: either you have high 'APM' and like micro or you have low 'APM' and

APM is like a talent not unlike game experience or hand eye cooridnation. There is nothing wrong with it. Strategy is build on the APM limitation, not against it.

would like to see the genre actually go someone and maybe even become more realistic!

Its easy to have a game without micro. It is stupid having a realistic RTS game. (which would take 2 real month to fight an average campiagn scale battle)

In my experiences with various custom maps where just about every unit combination has been tried, reducing unit micro is simply a question of reducing the discrete to the continuous, and reducing the random variables involved. It can simply be done in the following manner.
1. No hard discrete-ness. No one shotting expensive units. Long engagement times for a battle as a whole. Masses of units reducing importance of individual unit micro.
2. Less force mutiplyers. Less special abilities, less terran effects (on traffic and otherwise), less non-return fire combat. Homogenous combat results following strictly N^2 rule.

The best imaginable micro-less RTS game can be build easily by using powerful mod tools now avalible. Using starcraft as an example, a example would be muta-islands/wraith-zone control. By tweeking the targeting/firing behaviour AI, any remaining amount of micro useful in maps like that can be removed as well. The easy method would probably to mod those air units to have "appear on target" attack and count as instaneous damage, and add in a raise of nations "muti-attack penaty" and everything gets reduces to pure numbers.

Or more drastically, one can simply remove fine grained control to those units. I have seen custom maps where combat is controlled and resolved by chaotic and uncontrolled map scripting languages that prevents any real control.

While I know this is not the "goal" of automation, from an player management and control perspective, it is the same. All the fancy AI would simply produce the same gameplay as simply greatly reducing tactical combat complexity.

=======================================
I don't really mind a game with excellent AI and everything, with the player managing one "absolute and defined" part of the battle.. However I'd perfer it be single player, or a TBS (with real time elements like the Combat Mission series).

When it is online mutiplayer, management is something that should be considered as THE gameplay. What the player ought to be doing should be strategically consider by the designer. There should always tradeoffs for player time (aka thus always more things to micro than the player can do) consumption. It is the truely interesting part about RTS game. If the player can do everything or only has one thing to do, than the dimesion of strategy unique to RTS have been lost.

If that dimension is to be tossed out, an TBS is perfectly fine. Strategically they are the same as a RTS game where the player is not constrained by time, but they are easier to make and can be filled with far more bells and whistles.

[D.Turtle]

The harder part is not (or shouldn't be) implementing the automation, but having a game DESPITE all this automation. In effect: what does the player do when his base-building, army formation, group behaviour, individual unit reactions are automated?

Strategy.

I think this kind of game would be far better, as well, with hard unit limits, because then strategy is more relevant than Tanks+1 and charge.

A little bit more detail would be nice.
Do you just want to have your army magically appear, how do you choose what it consists of, what are the objectives, how does the player control the 'strategy' etc.

Your post is exactly what I meant when I said you can not simply throw in a thousand ideas and expect to get a good game, but that you have to show how those ideas fit together to create a whole that is a good game.

I am NOT bashing the idea of automation/autonomy. I am saying: You have to present ideas which would make such a game fun - ie what is the focus of the game, show what the player has to do, etc.

Are you going to provide any evidence at all for these claims?

Sorry, I do not have a state funded study to support what I think, but I'll try my best - if its not enough I'll try again until I concede

You're saying 'more autonomy = game easy', even though it allows more time to concentrate on complex maneuvers and tactics, making the game HARDER particularly for RTS fans who aren't used to such thinking.

I would propose that alone from a control viewpoint having to control every unit is more difficult than controlling only groups consisting of units, with each unit automatically doing what best helps the group of units achieve its goal.

You then claim that I have to invent a new way for the games to be fun, because somehow they won't be fun IN EXACTLY THE SAME WAY. Can you show me how units that automatically use grenades totally ruin an RTS and force a redesign?

Most of the ideas presented here (especially taken in their totality) go a bit beyond making units automatically use their grenades - something like that would obviously not fundamentally change the gameplay of an RTS.
Take away the entire base building, unit training etc, and yes, I would propose that is a fundamental change to the game.

This would make a game more accessible to new players, while still allowing 'pro' gamers their advantage.
Yes, this is applying the autonomy stuff to a 'standard' RTS, but face it - thats what most people want and are comfortable with.

This is such bullshit. Autonomy is not for making it easy or noob-friendly. It's for moving the focus from 'clicking like a speed monkey' to 'being able to orchestrate large complex battleplans'.

In the context of what I wrote this (implementing increased autonomy within a standard RTS) I would certainly hope there is a difference between controlling your own units and letting the AI handle it. Otherwise multiplayer and especially tournament play would suffer a bit (why control your own units when the computer can do it better?).
Implementing autonomy in an entirely new game created from ground up with this in mind would let you have grand strategy on a truly strategic scale.

You're just falling into the RTS-convention of thinking 'less insane clicking = easy for n00bs lololololol', and you're going to actually have to prove that. Games like this won't be EASY RTS'S they'll be ACTUAL 'real time' 'strategy' 'games'.

See above.

As Shep says, in games like CoH it's pretty lame that units don't have more human-like autonomy (or standing orders for retreat etc), but you think that would make the game easy!

DoW had morale and retreating/broken units - that would be an example of this implemented in a standard RTS. In addition such decisions are most often situationally dependent: Maybe that units is supposed to hold the enemy there for just a few seconds more so your artillery strike hits him.

At least you're not as retarded as Dendrobius, who thinks units running away when defeated is some superintelligent AI he must face in combat. Seriously - comparing a strategy game with unit AI to a shooter where you don't have to shoot is so utterly broken - RTS players need to realise that RTS-convention is NOT the only way to play 'real time' 'strategy' 'games'.

I do not think RTS-convention is the only way to play real time strategy games. In fact I haven't played actively for several years already (German army got in the way, and I didn't like the Warcraft 3 style of play - which was what my friends were/are playing). At the moment I am enjoying Medieval Total War 2 (and have been for some time) - and I love being able to pause and give commands.

However, at least you're all providing even more examples for my previous statements about RTS players being resistant to change and conventional. Unit morale = 'am i fighting an ai lol'. Automatically throwing nades = 'lowering skill level'. RTS not based on paper-scissors-stone counters = bad... somehow. Rolling Eyes Rolling Eyes

Those are idiotic statements when taken alone (and none are from me, i hope).
However, you always have to take design decisions in the context of their respective games.
To use the grenades as an example: How powerful is the grenade: will it kill an entire squad with one shot, or only injure a few individuals. Does the unit have infinite grenades, or maybe only one?
A unit that has infinite grenades that only injure a few enemy individuals would rather use them automatically, while a unit with a single use grenade that kills of most of an enemy squad would rather use them only on express command of the player.

That is what I am trying to say: Choosing and picking singular ideas and throwing them together somehow does not necessarily lead to a good game.
You have to look at the big picture - how do the different ideas work together, etc.

[HSRTG]

Control is the most personal, intimate and variable resource in the game that can not be described by a build order or a strat. It is the one personal piece of the strategy puzzle that can't be dumped into an AI.

But you can limit how much control is needed. I shouldn't have to tell my machine guns to aim at infantry, not bounce .50 cal ammunition uselessly off of 6 inch armor plated steel (AKA A tank).

As a player playing the game, it is the player's job to take command of things. Control is what the player DO. As the complexity and number of events in a RTS game is staggering, there is always more thing to manage and the player that can manage better and faster has the advantage.

And instituting some common-sense shit like a machine gun firing on infantry is wrong? I'm sorry, I like having my units to have some semblance of brains.

AI is not the solution, as an "game feature" where the AI can manage is a "game feature" that is irrelevent to the player. The game can simply be simplified to remove the feature all together. Instead of complex bases and complex base AI, one can simply have an handful of buildings doing everything. Similiarly, instead of auto-casting spell casters, one can simply remove magic and integrate abilities into base attack.

We don't need to dumb down games like that, I just want to make my units a bit less stupid. I shouldn't have to tell my machine gun squad to shoot at infantry, and leave the tank until the guys without 6 inch armor plating are dead.

An game feature only has meaning if the player has to spend time managing it. A games micro requirements basically is a function of game complexity and time.

Bullshit. A good game feature requires zero player management. Or perhaps you think we should make resource harvesting a manual activity? I know! We should make SCVs only bring resources in when their told to, even if they're full. Brilliant, sir! Additionally, they only go out when told to! EVEN MORE MICRO! HAH! I COULD GO ON ALL DAY WITH THIS!

If there is little need to micro battles for example, the player would simply push an attack move on the enemy and spend time in "boring" things like concentrate on squeezing tiny ounces of resources by systematic peon/building management.

Or they could concentrate on creating diversions, traps, you know, the things that only 300 APM guys can do right now. No wait, that's boring.

If there is nothing that the player needs to do for most of the game, the game probably would be speed up to the point where there is something to do other than stare at the screen. (very few RTS are played online at the slowest speed, even for the most micro-intensive ones)

So?

I think the problem with micro isn't that there is too much of it, but it is often very boring and has clumsy interface, and that micro demands come in share spikes (short battles) as oppose to smoothly. The micro troops to outflank the enemy is one thing, to spend 30 seconds rebuilding the farm complex due to natural exhaustion is another.

So let's at least contain the unnecessary micro, then work on interface. RTS does have to equal "Micro 24/7".

Finally, micro is the only true random variable involved in a RTS game not designed for chance to be important. What seperates good players and average ones is the ability to apply the resource of player concentration at the right place in the right time.

Let's make things harder for no good reason! Because we CAN!

If there is enough time for the player to sit there and think of an strategy, it is really a TBS with timed turns.

No fucker. It means there's something called "strategy". No, wait, you like Real Time Tactics. Go play Hungry Hungry Hippos. That's micro intensive.

2. There is no tactics: There is only Rock Paper Scissors
Well, one should not consider tactics outside player input. The only tactics relevent to the player is the tactics implemented by himself or his opponent.

Increasing the range of possible strategies via minimizing unnecessary micro is a bad thing?

People complain of games that require little micro or tactics as "tank rush games" like Heavy Tank rush in RA, while they also complain of games where tactics and control has massive effects on the game, like Starcraft where a professional player can kill a force twice its size controlled by a casual player due to knowledge and precise 300+ action per minute controls.

So what you're saying is we have a choice between broken macro, and a shit-ton of unnecessary micro? Or am I missing the point?

People want to eat their cake and have it too.

Not part of the debate, but I hate this fucking expression with a passion. I have to ask, what the fuck is the point of cake if not to eat it? Do you just let it sit there and rot? What's wrong with you?

They want tactics but they want tactics to happen by itself while they press a butten say "please don't let units be stupid." Like I said above, any feature that does not require player input is one that is irrelevent.

Like what you said above, this is stupid.

Either you make player inputted tactics important, or you don't.

Black/White fallacy right here. Very nice. See what I described above about less-stupid units. Or would you prefer to have manual resourcing too?

In a true "rush, swarm and attack move game" devoid of tactics, what is the decisive element in battle? Yes, this lameness called having and sending more and better units to the enemy when your enemy lacks them. That is pretty much what strategy is about: The logistics, the production and the grand scale macro-management of the economy.

So you don't like having to have a logistics tree? Go play Ground Control, that has no resourcing or logistics whatsoever.

Things like placing an artillery on a hill or flanking troops is tactics, not strategy.

Yup. But I never said I didn't want tactics in my strategy, especially things like that. I just wanted my guys to be less blindingly stupid. I wouldn't mind my guys moving some to get in cover, either. That would be nice to have.

Now one might argue that such "strategy" really don't take that much thinking, but that is untrue. Things like optimal build orders is anything but self evident. The only reason why it is not considered "intelligent" is because RTS generally revolves around a few very, very well studied maps with millions of games played on them and all possible decisions tried.

Your kind of strategy is a mind-numbingly boring clickfest that sends me to sleep. I play Starcraft when I want to go to bed.

Now, to fix that problem is extremely difficult given physical limits and the player base.

No, it's not, you're just stupid.

Lets look at the solutions:

This should be hilarious.

a. More maps: However given that all fractions have to be somewhat balanced on all maps, and that all fractions have to be unique, the set of possible maps is tiny. Unbalanced maps are not accepted by the RTS community. (unlike wargaming/tabletop gaming) In anycase, most players would stick to only a few known maps anyways.

At last count I had 4000+ maps for Starcraft before I quit. More maps do not work.

b. More complex interactions that can not be studied to completion in millions of games: This would be far to non-linear and abstract for most of the RTS community. Games like Chess or Go have no real life counter part in terms of interaction after all. In RTS games with fog of war, information is imperfect and a complex interaction game would not consistant ensure the better player wins as randomness added up with uncountered non-linearity means chance plays an important role.

The more skill you have, the less chance plays a part. You've been an advocate of "skill" all along, and now you're saying...well...I'm not sure what you're saying here. Lay out how a game could be too complex for skill to be important.

But really, the lack of "strategy" is also reflective of the generally uncreative player base. For all the whining about "lack of strategy", most players never put in the effort of actually learning about all aspect of the game to form a effective original strategy anyways.

So creativity from developers should also be quashed? 'Cause I think if the developers gave a kick to the player base, there'd be a shitload of creativity very quickly.

Most just learns some "power strat" and complain of lack of strategy. Some simply complain that their APM is too low and don't even bother.

APM shouldn't be as important as it is in Starcraft. I'm sorry, no. And I don't learn "power strats", I just play the game for kicks. But I'm just pointlessly shouting into the wind at this point.

The AI is not a human, however the AI is not allowed to use its greatest assets while has many artifical weaknesses. In the same way that the AI is not allowed to aim-bot in a FPS game, the AI in a RTS game have many limitations. The biggest limitation is that the AI is build before the game is released in most cases. They will have no benefit from the strategies that is developed after millions of games played by the entire gaming community.

One word, chuckles: Patches. Galactic Civilizations 2's AI has been remarkably improved since release. I kicked its ass at Tough (the highest level sans AI cheating) when GC2 came out. Now? I have to fight for my fucking life at that level. Get my drift?

The other limitation is that AI can not do what a human can not, for example one can not program an AI in starcraft to win by insane SCV micro in small maps. The AI can do probably do it, but no human would want to play against it.

Why not? It'd certainly hone your skills. I wouldn't complain if an AI was actually "skilled" and honestly kicked my ass.

So we are removing all strength of AI and adding many weaknesses over a human. It is no wonder that they are of little challenge. We need an AI that can read forums and evaluate good strats, then we are talking.

I'm not sure what your last two sentences were supposed to mean. I don't speak gibberish. I think what you meant with the first one is that the AI is supposed to be able to micro better than humans? I don't know. Run this by me again, with your English code turned to "On".

[Edward Yee]

I was just discussing this elsewhere, acutally: in most RTSs there is no penalty for moving and firing... so why don't the units constantly move? Why do they stop and shoot, like in Dune 2? Many RTSs have laughably slow projectiles, and attack-move micro is very effective.

I must note that I don't think my example goes anywhere in this direction; the unit's reaction could toggle between it (movement is unaffected, unit only attacks until out of range) or the standard reaction (sustaining the attack).

HSRTG, would that example work with your examples? (i.e. machine guns versus heavy armor.)

To add to what innovations I'd do, I'd have one Move command and use a standing orders toggle for its ROE...

[HSRTG]

HSRTG, would that example work with your examples? (i.e. machine guns versus heavy armor.)

It would.

To add to what innovations I'd do, I'd have one Move command and use a standing orders toggle for its ROE...

Kind of like what Total Annihilation does. I would still like that, and something like attack-move. I want to tell my guys to burn that city to the ground, while moving through it.

[Edward Yee]

Incidentally, if we apply my idea (however half-assed it might be?) to the standard Move, it could do all three. Conventional "stay quiet" move, prioritizing the move (but allowing the unit fire initiative), and conventional attack-move (which would actually give the AI more initiative than conventional move).

Would you change though the AI reactions when using the standard "attack-move"? (Sticking to the chosen target until it's dead and moving the next one.)

[Edward Yee]

Or would you rather leave standard attack-move and manually choose targets/movement in an urban demo scenario?

[HSRTG]

Would you change though the AI reactions when using the standard "attack-move"? (Sticking to the chosen target until it's dead and moving the next one.)

There are four possible move choices I can think of.

Move somewhere, but don't engage under any circumstances: Used for setting up hidden bases, or ambush zones, or something that requires quietness.

Attack things, but don't pursue: General harassment of patrols/resource spots.

Attack things and pursue for a limited distance: I'd use this for interceptors and the like.

Attack things and destroy them no matter what: Base assault. Or similar.

Or would you rather leave standard attack-move and manually choose targets/movement in an urban demo scenario?

I'm in favor of having more options to pursue my goals.

[SWPIGWANG]

But you can limit how much control is needed. I shouldn't have to tell my machine guns to aim at infantry, not bounce .50 cal ammunition uselessly off of 6 inch armor plated steel (AKA A tank).

And instituting some common-sense shit like a machine gun firing on infantry is wrong? I'm sorry, I like having my units to have some semblance of brains.

It is not wrong, it is not that important.

A game's strategy is defined by a PLAYER's input. If you are willing to accept abstractions (which every RTS player is suppose to, just look at the mine and build times), one can simply remove the game play part involved.

Every unit, every event and every gameplay "feature" is really a black box with an "player input" and a "result output." What is inside the box and how things are resolved is simply irrelevent to the player.

In this particular case, you can simply replace rifle infantry and rocket infantry with an "infantry squad" that does similar damage to both tanks and infantry. There, you've removed micro without requiring AI. The game play remains.

Bullshit. A good game feature requires zero player management. Or perhaps you think we should make resource harvesting a manual activity? I know! We should make SCVs only bring resources in when their told to, even if they're full.

SCV mining is not a "gameplay feature" as mining is a boring event which few would want to pay attention to even if it takes strategy. Resource harvesting can be reduce to slapping a "mining building" on top of the resource instead of silly things like SCV running everywhere, which is mostly just eye candy.

I'm not saying that everything have to be done in a clumsy manner, but that down the line at some point, the player has to be spending its time choosing between an set of activities which he can not simulatously fulfill.

A game's gameplay is defined by where the player do have to manage things. The other things can just be toss out in abstraction boxes as they are just eye or ai-candy that do not define gameplay.

Or they could concentrate on creating diversions, traps, you know, the things that only 300 APM guys can do right now. No wait, that's boring.

Diversions and traps has most of its meaning when the person on the other side is overloaded with his own micromanagement problems. Try playing starcraft at 1/20 its base speed for example (giving the average player 1000APM post scaling) and things like distractions no longer mean anything. Your opponent can easily deal with any distraction and everything else at that speed and it ceases being a gameplay factor.

Yes, it is difficult for the average player to execute complex battle plans, however it is difficult for them to counter them as well. The goal of the game is not to have "perfect planning" but to have "plans that beats your opponent."

Because controls are so hard in Starcraft, things like muti-direction drops are effective as your opponent would have to struggle to move units out of traffic jams and manage not to have its units arrive in chaotic piecemeal.

Remember, the gameplay factor that defines RTS is player input. The ulitmate factor on whether something will be done in a game is on the cost-effectiveness of the action. If one designs the game so that complex battle plans pays off heavily relative to the amount of control required, it will be the focus of the game. This can be done in a infinite numbers of ways, and AI is merely another factor like health points, weapon range, unit speed, unit size, but it is one that is most expensive to develop and do not necessarily do anything more than passive factors.

Increasing the range of possible strategies via minimizing unnecessary micro is a bad thing?

You do NOT increase the range of strategy via reducing micro. Like I said, you can reduce micro VERY EASILY. It is so easy it is trivial.

If you've played an "RTS" with only one unit, one building, in numbers with mechanics where micro is infeasible, than you'd see that there is not more inherent strategy in that sistuation. In fact, there is no strategy beyond endless waves of units.

Strategy is NOT what your units or your forces do. Strategy is what YOU, as the player, do. Strategy comes from having a large set of choices to make at a give time. Strategy IS the choices one makes.

When the AI takes over a factor of gameplay, it removes that from strategy. There is a choice in "whether or not I micro my machinegunners to shoot enemy infantry" while there is no choice in "my machinegunners engage the enemy. The AI is good, I don't have to look at the troops."

The question of where, and how to micromanagement units is an dimension of RTS strategy. It is not just about clicking fast. It is about clicking at the right things.

Black/White fallacy right here. Very nice. See what I described above about less-stupid units. Or would you prefer to have manual resourcing too?

Incidentally, there are games with manual resourcing. (AoK comes to mind, as are some european offerings)

The particular part about Age of Kings is hunting animals to boost the economy. (one gathers food faster by hunting than killing sheep, picking berries or farming) It is poorly automated, requiring the player to babysit the peasents to not killed by wild animals or other forms of stupidity.

When it comes down to it, that part about the gameplay does not detract from the AoK in general. The reason is that it is that the player have a choice as to whether or not to micro for food or to use a less micro intensive method like berry gathering. This is a choice, and that is a good thing.

Your kind of strategy is a mind-numbingly boring clickfest that sends me to sleep.

Just because it is boring to you does not mean it has no strategy or is easy to master. I found standard Starcraft boring too, and spend my time playing use map setting. However it does not mean I know or could easily learn about all the details of strategy involved in a standard game.

At last count I had 4000+ maps for Starcraft before I quit. More maps do not work.

Given that you do not play at a high level where minor map differences make the difference, I don't think you are in a good position to judge. I doubt you even really understood what different start locations on lost temple really means for combat given the different races.

Also, because of all the assymetries of races, only maps of a very limited design could be balanced for every race. There is not enough leeway build into the system to allow for drastic changes in strategy.

The more skill you have, the less chance plays a part. You've been an advocate of "skill" all along, and now you're saying...well...I'm not sure what you're saying here. Lay out how a game could be too complex for skill to be important.

Because in a non-total information game, much of what is going on is gambling.

In a "simple" game, there is usually a set of "safe builds" that is reliable to be competitive against any other start the opponent could get. Because those builds, aka "strategy", is so consistant and reliable, it can be reused quite a bit and is not really "creative" after the few experimenting souls developed them. The starting strategy involve usually is for choosing a set of competive "safe builds" and use micro-skill and game knowledge to make up for any minor disadvantages between those builds.

However, to build a "complex" game, there can not be a set of safe builds that the competitive against any other build, or else it would be used exclusively. As an result, the game has to be between a large set of unsafe builds and the lucky could win by guessing right.

But really, the lack of "strategy" is also reflective of the generally uncreative player base. For all the whining about "lack of strategy", most players never put in the effort of actually learning about all aspect of the game to form a effective original strategy anyways.

So creativity from developers should also be quashed? 'Cause I think if the developers gave a kick to the player base, there'd be a shitload of creativity very quickly.

The players have spoken, and they've decided to stick to counterstrike until the end of time?

Anyways, few RTS players actually "want" to think very hard about all dimensions of strategy. Even the AI fanclub have little coherent idea of what strategy actually is.

Strategy is about the choices the player makes. The objective a game designer is to give the player a set of carefully crafted, non-obvious but still important, choices at every moment in the game. The rest is just eyecandy.

AI or no AI, that is the truely critical part of the game that is the engine behind the game. AI is just a tool, it is not a wonder drug. It can be used to remove, alter or create new choices, but that does not good without guidence. Gameplay is build by having a plan. IF one has a plan, than all else is just tools. The important part is having one.

Gameplay is not build be reducing unit stupidity. Stupid units are like badly render lego units and fugly terran. Annoying, but it is not by itself an driving factor.

The question isn't that units are stupid. The question is what is it that the player ought to be doing.

Once you have a complete answer on what sort of management tasks you want the player to follow, it is not to hard to trivialize all other parts of game dynamics to make it unimportant to the player in terms of management.

-----------------

Why not? It'd certainly hone your skills. I wouldn't complain if an AI was actually "skilled" and honestly kicked my ass.

Well, depending on the game, it is possible to seriously develop killer AI by focusing on the AI's greatest strength. Infinite APM and perfect execution. In long, complex and ambigous games, AI development becomes quite difficult and expensive.

In other words, the best AI opponent would be one that tank rushes in a micro-ed way that no human can match.

But thats cheesy like a aimbot FPS enemy with a sniper rifle.....so thats the end of it.