Defining micro

5sec might be too short for Go, I am not into it, as I said.
But introducing a challenging -not impossible, make it 30s for Go- timelimit to a game does not reduce its strategic depth. PA isnt turnbased in the first placed, so it is a bad comparision anyway, but a good PA-Player should be able to make a wise move within seconds. Thats just the essence of a RTs.
If you dont want time pressure an RTS is the wrong game for you. I repeated that often enough now :/

I think the 30s rules does reduce some strategic depth, it exist mainly because that the normal Go games takes too long to end, some players and spectators can't afford that.
Strategic rts games (NOTA, Kohan series, SupCom etc) usually tend to be a bit slow paced so the players have more time for thinking, they do tend to be a bit similar to the turn based games.
There is some depth in fast and accurate clickings too, but I don't think we need much of this kind of depth in PA.

Well, yes. The reason why we disagree is that "trivial" for me means anything that has a small set of defined behaviours that apply to all situations (minus the unlikely rare cases). The example of metal makers on/off is a clear case of a trivial decision; there are only 2 options and in each situation, at most 2 factors can be easily parsed to reveal which option is the "right" one. There is zero skill in deciding which decision is the best, therefore the player should not be required to make that decision.
Decisions with nearly zero impact are also trivial. It doesn't matter which side of the factory your assisting engineer stands, even if the decision might be "hard" to work out what is ideal. Just chuck it somewhere suitable and forget about it.

Workers should mine minerals in starcraft is a good example of what I consider trivial. Units should maintain maximum range is likewise trivial in nearly as many cases. (enemies behind you does not happen often)

Formations, flanking and army positioning. Now those are not trivial. They are still easy for a human, so I would consider them low level decisions.

And here is where we differ.

IMO, execution is second to planning. I would like to see emphasis on the Strategy part of the RTS. This means planning, managing information, attempting to guess enemy movements, etc.
A player who spends 100 hours a week practicing low level micro until he can eke out another bit of effectiveness from his artillery kiting should lose to a player who has spent 100 hours a week studying the overall metagame.
Even the *UI* should be as intuitive and as fluid as possible. I do not want to practice to use the UI. Ideally, I would pick up the game and spend a short period learning the ropes and be able to play on the macro scale immediately.

I do appreciate the point you're trying to get across about RTSs, Cola_Colin, and it's important to remember that an RTS is realtime for a reason. However, it doesn't necessarily mean that an RTS should be based around quick reactions, just because it's realtime. The realtime aspect has more to do with the fact that both players are making decisions and changing the game on a simultaneous, continuous basis, and that the time of an event occurring is done on a less deterministic scale than a turn based game. This has more the affect that the player has to deal with a greater level of uncertainty, making hypothetical situation planning the center of focus.

In short, the focus of an RTS should not be on quick reactions, but on the player's ability to deal with an ever changing battlefield. This is not something that low level unit AI would compromise.

But more importantly, you have to remember that PA will have battles spanning planets. No matter how many windows you have open, your attention will be focused in one area at any given time. As an RTS moves to a grander scale, player reaction time will decrease. With a proper notification system, this can be mitigated, but even still, there's a limit to how timely the player's response will be.

If a player sends a group of units to do some raiding to distract an enemy on one planet while they coordinate a larger attack on another planet, that raid will take a lot lower priority in how much they want to focus on it. But it doesn't mean those units should be considered cannon fodder. So you set things up to increase their survivability. Enable unit AI so they are smart enough not to run directly into the cannons. Place a retreat zone in an acceptable location and enable their retreat conditions. Then tell them to go do their thing.

For the most part, this group should now be able to carry out your orders and deal with most situations. But supposing an extreme case came up that your setup could not cover. The player would still be monitoring the group's progress to a certain degree, so seeing a batch of incoming red dots on the group radar should grab their attention pretty quickly. Similarly, a smart notification system, like someone already mentioned in a thread previously, would be enough to give the players the details they need about any issues that occur on this raid.

So now the player can deal with the problem in their own way, optionally disabling the AI and the retreat commands so the units can follow their exact orders.

It's also important to recall that the TA-style unit stances will be making a comeback, and it's not too much of a stretch to think of how these can be implemented to affect the unit AI. Set the unit stance to Roam and when the unit tries to kite the enemy, they'll be free to kite in any direction, at their own discretion. Set it to Hold, and kiting will only be done on a small scale.

It's also important to recall that using this method is entirely optional, much like the debate on Fixed Zoom vs Strategic Zoom. Don't like unit AI? Don't enable it. There are still players that will find a use for it.

yeah i've agreed with everything you've said in this thread. i only know about Go coz _Pink mentioned it on the GPG forums at once stage

You guys should check out the free online Go servers. Pick up a client and make a free account. Play against people of all skill levels from all over the world. I was a big chess buff when I was a kid, and then when I started playing Go it completely superseded chess as a far superior game.

It might be worth mentioning that in Go, low level decisions stop to exist after the first several steps, so with 30s rule, its need of quick reactions is quite different than kiting or jinking in a rts game.

Like I said: Comparing Go with PA is not really a good idea.
Go is a turn based board game.
PA is a RTS.

They have only few things in common

I want the unit AI to do this stuff when I explicitly tell it to. It's nice for example when in the new Starcraft you can queue workers to mine minerals directly, but I sure as hell don't want the game to assume that every worker that's idling near minerals should go do that.

Metal makers, well you're wrong that it's obvious in every decision should the be on/off, but also there I want to decide the exact conditions of managing them. Like turn off when I have <X energy, otherwise on. While that's simplistic and does not give me optimal MM use all the time, but it's close enough and more efficient handling would give quite diminishing returns.

Keeping max range is something you really don't always want. There's something behind you, you want to keep pushing forward, you want to maintain formation, you want to maintain facing... It's also easy to abuse - let's assume you've got a group of long ranged bots witha slow moving projectile. You're planning to attack with them against my line of turrets which they outrange. Then I send in a fast moving scout, who can easily evade the bullets when shot from max range. This one guy will push your artillery back and delay the attack when they assume it's obvious you want them to kite a lone scout with maximum efficiency.
Again, if you want to have units do this kind of thing by themselves, it should only ever be with a implicit command, that's distinct from the normal move, patrol, attack, whatever commands.

And what's to say that doesn't already happen in most RTS's? Overall strategy (NOT called metagame btw) already is the most important thing, having a good buildorder and getting more stuff mostly trumps slightly better micro.

And isn't the example - unit AI moving units moving out of position without your input - exactly a case of low level execution (by the AI) overriding the high level plan you had?
Compare to how most people had their stuff on Hold Pos in TA - because they rather respond to threats intelligently by hand than let the unreliable unit AI charge in.

I'd say that from the requirement 100hours micro training loses to 100 hours strategy/metagame training is true for Starcraft 2.
The micro part is just what looks most impressive in pro games, so it makes many new players try hard to micro, while their macro totally fails.

I think idling workers near minerals should start mining when:
1.they are not in hold position state.
2.they haven't received any commands for a while.
3.there is a command center near the minerals.
4.worker limit for the minerals haven't been reached.

The cases that you need to push forward with the ranged units or attack something behind them are rare.
You can still maintain the facing with kiting.
Maintaining formation in a TA like game is just to keep the ranged units behind the close combat ones, which isn't conflicting with kiting at all.

If you want to specifically kite some units and ignore others, then you can use a area attack command with auto-kiting instead.
And if you have any way to pay attention to multiple places, like minimap, zoomed out view or separate cameras, taking effort to a abuse that can be easily interrupted by your manual control won't be worthwhile for your opponent, he might waste his units because of it.

I doubt that.
If the most players hate auto-chasing, the most rts games won't have it as the default unit behavior anyway.

You can't find a top level starcraft 2 player with mediocre micro, I want that to be possible in PA.

Because at top level it gets very hard to improve macro people need to start using their units as effective as possible. That will be true for PA as well. It is just a fact for any RTS: If you are perfect at producing units you need to use them as effective as possible. Also while learning good macro you will have to deal with unitcontrol as well, so a player who is really good with macro wont be that bad with micro anyway.
Also: You dont need any kind of hardcore-micro at all to be really good in SupCom already. The move 1cm right-move 1cm left micro type is helpful, but not many players actually do a lot of it in SupCom at all. Sure you need to have an eye on your units, but thats not very hard micro. Thats just "oh I get attacked, move back" Sure you could instead put up a retreat-zone, so your units will move there as soon as they get shoot at, but always putting up such zones will take just as much time as giving one or two move commands more.

The essence of it is: In PA or any other RTS it is an important part of gameplay that the player constantly needs to watch at all his units and bases so he can quickly react to anything that comes up. Thats a part of what RTS stands for.

Not all unit controls are micro, sure improving micro would be more effective than improving macro after the player reach a certain skill level in the most rts games, I just want that level to be much higher in PA by making the micro much less effective, and I think game mechanisms that encourage macro is not enough for this.

Even with the same amount of clicks, be able to set a behavior to respond the events that not happened yet is still better than have to pay attention and respond events after they happened.

Wrong, what SupCom does is already enough.

Even with the same amount of clicks, be able to set a behavior to respond the events that not happened yet is still better than have to pay attention and respond events after they happened.[/quote]

Constantly paying attention to what happens is a BIG part of what RTS are. If you dont want to pay attention you really dont want to play an RTS.

@luckywaldo7:
???? You need to have an idea of what an RTS is to make on. Your view on what an RTS may differ and an RTS made by you might end up having turns so nobody needs to be quick about anything, but you would still think about what RTS means before making one.

EDIT: Did you delete the post that you just posted while I was writing my post? xD

Yeah, I had a lot more to write but it somehow got lost when I posted, so I deleted because it wasn't terribly productive on its own.

For the record, what I posted was "A bad way to make an RTS is to start out with lots of assumptions of what RTS needs to be."

Here is more of what I wanted to say:
RTS means real-time decision making, but there is a tremendous amount of flexability in that genre. You have some games like Starcraft which focus more being an ESport. On the other end of the spectrum you have, say, Sins of a Solar Empire, for more casual play but ridiculus strategy and scale. Somewhere in a comfortable middle you have TA and Supcom. If you try to trap yourself into being like any one game though you are creating a creative box that will be hard to escape from. Games like TA and SoSE weren't created by trying to be like something else.

Basically, outside of literally being real-time, and centered around strategy, there aren't fixed rules as to what an RTS needs to be.

(Although I admit I am here because I hope it will be very TA-like)

I'd hesitate to say a game like Sins or TA or Supcom is casual in any way. If anything, Starcraft is the more casual game, as it's a lot less involved in many aspects. You may have to learn to micro and memorize build orders, but the other games replace that with a wider array of potential strategies. But in any case, RTSs of any sort are not casual on any level, because they aren't easy to learn immediately, and getting good at them takes dedication.

Sins of a Solar Empire makes micro ineffective by making the units respond the orders very slowly, it is similar in SupCom.
So it seems relying on game mechanisms to reduce micro will limit the gameplay. If the game rely on automations more, at least the micro lovers might get some fast responding units.

Strategy games do not need micro to be deep or complex. Automation and lack of micro do not reduce complexity.
http://youtu.be/hfDJl7icJcs
Seen in this video:
Unit automation
No micro

But it's far from a simplified game. Here's an after-action report to summarize a battle:
http://www.wargamer.com/article/2694/af ... aar?page=2

Now it's nothing like TA/Supcom, as it's more of a wargame than an RTS, but the important thing to note is that there's no loss of complexity or depth by taking away unit micro. Instead there's more of a focus on higher level strategy and tactics.

Yes it is nothing like SupCom. I like the part of SupCom where you have to be quick to react and control units all over the place in parallel. Control units as in "give general move commands", not stutter-step-sc2-like. I just cant see any benefit in a better AI for the units. At least not as long as the AI is focused on reacting independently. However an AI that aims to perfectly executes the commands I give would be very appreciated. This mostly means: pathfinding, pathfinding and ... pathfinding. Units that dodge into each other and block each other are definitely just annoying. It also makes me think that tech-wise game-developers are mostly far away from implementing any kind of AI that is intelligent enough to control my units. They might try, but the result is very probable to interfere with my own tactical ideas.

One of the big reasons that SupCom has not much micro is that units react slow and shots are so fast that evading them is mostly impossible. Apart from few exceptions, mostly vs artillery. You are saying that relying on game mechanisms like this is reducing gameplay, so you are saying SupCom could have more gameplay if it had more micro?
And to add in the possibility to micro by an AI would help with this?
I disagree: Either the AI is so good that there is no need to micro at all, which means we just removed one aspect of the gameplay(which is paying attention at all times and making sure units move wisely) or the AI isnt very good and players will still constantly have to make sure that the AI is doing the right thing.
I dont want to baby-sitter a bad AI, I just rather do everything myself. And I DEFINITELY dont want an AI that is so good that units that have no attention of the player are just as effective as units that are directly controlled by the player.

@luckywaldo7:
Yeah, I am here cause I want a game that is SupCom/TA-like as well. It is just that directly controlling my units all over the map is a big part of that in my eyes
I also never said you are supposed to trap yourself. You just need to get an idea of what kind of RTS you want to create. So what RTS means to you.

No, it's just that if a player suddenly find out his units is running into a trap, he can't withdraw them as soon as giving them a move order, it's about the fast decisions, not micro, although it hinders micro too.
Fast projectiles means the game would have to lacks the units that rely on jinking for combat and the units that are bad at killing fast units.