The good side of Micro

That argument just jumps the gun to full automisation of everything.

But that is not what we are saying, perhaps you should give Zero-K a go? It is compleatly free.

I guess chess isn't very interesting to you because computers can play it pretty much perfectly?
The availability of an AI says nothing about how interesting or fun a task is.

Also, I would like to hear, why PAs big scale is not doing enough to make micro unimportant and why all this AI is needed.
Arguing against micro and for AI are two different things even though there is an obvious connetion.

Some tasks you need to deligate inorder to focus more attention to larger tasks.

There is a reason that company's hire managers.

To be honest, I don't consider chess to be interesting, mostly because of that exact reason. Chess is a great example of a game where most of the time spent is reading about others playing the game and copying their tactics, instead of actually playing yourself and coming up with your own twist on the game. At least, if you want to be any good at it.

Chess, like Starcraft, is a game where in order to win, you spend a lot of time copying plans other players came up with and just following their checklist. It gets boring pretty fast.

I really hope that PA will avoid that obvious trap.

I think what plusjen is trying to get at is that chess is not complicated enough.

It is actually somewhat possible for a human brain to brute-force all possible moves in chess, and memorize optimal openings and sequences pretty much right through to the endgame. At the grandmaster level, games of chess really are not that fundamentally interesting in terms of game playing. At that level, the game is so well understood by both sides that it is more about game solving which is also interesting, but it is quite a different kind of thought process.

Most likely PA will be complicated enough, and with a tense decision set and a large enough variety of compatibly effective styles that no human will ever "solve" the game. The best we can do is play it.

Mostly, yes. Not entirely. The difficulty in chess results from having more possibilities than a human can easily comprehend, but that a computer can easily calculate through to find the best option.

What I'm looking forward to is a game where a computer cannot calculate the number of possibilities, but a human can easily see which paths are worth exploring and which ones are not. Designing a game that is challenging the thing that sets us apart from an AI; strategic thinking, predicting moves of our opponent, and calculating through the unquantifiable.

Leave the AI to solve the easy number crunching tasks, like how to move a unit so that it doesn't get shot, or to keep issuing orders to a factory to continue building stuff and sending it to a specific location with a specific task to complete.

Yeah, but micro isn't neccessary to win a game, except maybe for the earlygame, where there are no larger tasks to focus on. Once the numbers get bigger it becomes unimportant, since picking off an extra unit or two isn't significant at all if there are one hundred in the battle.

Companies don't hire managers if they are not worth it. And in this case, they are not.


As for chess:
You can brute force your way through pretty much every turnbased game. They key however is, that you don't have the billion years to do that. Which is also a reason, why RTS AIs usually suck.
And what I wanted to say with that is: Having an AI for it doesn't make stuff less interesting for everyone. After all chess still has an enourmous amount of players, despite computers being able to wipe the floor with everyone who isn't a Grandmaster. Also I doubt, that anyone would abandon an RTS simply because someone managed to create a powerful AI which could defeat a pro player.

Say's you, and considering that these managers will be free I say why not?

Micro of unit movment and behavior would be a great addition, I don't see why there would be a problem with including it?

To be fair, any game that could be beaten by an AI like that would probably not attract a huge following in the first place.

Although I wonder if something could make an abusive AI in Starcraft that would completely rofl-stomp pro players, but I think even that would be too hard.

What exactly do you mean with microing unit movement and behaviour? To me that pretty much sounds like something that you can already do in almost any RTS.

Because hiring them costs time which could be spent on something more useful to the company. Also why should you even hire someone for fixing something that isn't broken?
Aka: The time spend on creating the AI could be spent on improving an other aspect of the game, which is more useful to the player. Why try to fix the micro nobody is going to do?

You remember chess? Also as computers get more and more powerful, this really is more a question of time than anything else.

Most yes, but not sup com.

Even TA has limited unit AI but in sup com they just sit there like bricks.

What I am suggesting is an extenction to what TA did, in line with what Zero-K has (I seriously suggest you give it a go, as it is absolutly free).

You cannot quantify sofware engineering, so that as a point makes no scence.

Also it has nothing to do with fixing somthing that is broken, but rather improving somthing that was already there.

Becuse people do, micro is needed to do any spesific action of the player, essentually anything you do could be considerd as micro, macro is the extention of it to a larger and brouder scale.

Moving 1 tank compared to moving a collomn of tanks.

If you think that micro is not needed then how is it that you actually play at all?

Who is queing up factorys? Who is selecting the units to build? Who is moving engineers away from artillery?

Its you, and considering that the actual AI will also be using this as well, we really won't be adding much at all so it has nothing to do with spending more money or resources the usual.

All it is, is an idea to somthing that is already going to be in the game.

Somehow I get the feeling, that we all want the same and everyone sucks at expressing or explaining it ...

What I mean with micro not being required is, that you are able to do good without being super detailed in terms of ordering your military units around.
Basically, a small number of well placed commands is enough to tell your army what to do.
Which is within the broadest definition of the word micro, but usually not what people think of as micro.

I guess how to go about the AI question really comes down to preference. In the end I don't see a really decisive argument on either side and after all following both sides can result in a great game if it is done well.

Probably discussing it should wait until we see how PAs gameplay turns out to be and go from there.

That happens suprisingly often. :lol:

I'll use this an an example to reiterate a clarification on this point which was made earlier.

There are two aspects to everything the player does in a game; the choice to do something and actually doing the actions to implement it.

The actions involved in dodging shells does not require much thinking. If a unit is moving towards something with slow projectiles it should just alternate between moving slightly to the left and slightly to the right. There is little thought involved but it takes a lot of actions.

But deciding to dodge shells does require thinking. It is not something you'll always want to do. Jinking left and right decreases the speed at which your unit moves towards the enemy, it could be important to kill the enemy unit as soon as possible. Maybe your unit will never catch a retreating enemy if your unit jinks. It is riskier but faster to approach a unit without dodging. Players will have to weigh up these factors and decide whether to dodge shells.

Players may be required to make quick choices about whether to dodge but they are unlikely to constantly change their choice. "Reducing micro" is giving the player controls to make choices which do not require constant attention to maintain. Once you decide that some units are going to dodge shells in an engagement any further management is busywork.

The above sounds about right. A setting "should dodge bullets?" (with probably a more meaningful name, but hey), a default setting for each unit type based on what it the more likely to be desired setting, done.

Anything beyond that is pointless busywork.

TA had the option to change unit movment so units would automaticly move when they se an enemy, move and follow enemys they saw or to not move at all.

All I ask is an extntion to the way that ond the other unit stances work so that it is a little smarter and more intuitive.

no I disagree how do you land your transport and where, you get more intel as you get near you select it and make sure the units get on firm ground safe and sound, you don't watch it it gets shot down, no drop.

Macro and micro are yin and yang to me, I won't have one without the other, doing both is the epitome of multitasking.

No this is fundamentally incorrect and arises from the corruption of the term micro from people opposed to it. I've posted this here before and will do so again.

Macro - Any action designed to increase the raw quantity or quality of what you have. Be it more resources, more troops, more technology.

Micro - Any action designed to increase the efficiency of your army. Getting a good concave is micro. Picking a choke point for an engagement is micro. Putting your tanky units to the front of an engagement is micro.

Mechanics - The physical actions required to understand battle field information and give out commands to execute your desired strategy.

The terms are essentially about the scale of the optimisations you are making. Micro does not mean "micro management" and never has. What people really want to remove from the game is SC2 mechanics. Such as tapping to keep an eye on production or fiddly control in battles.

Because people don't know what they are talking about they are complaining about the wrong things. We want micro in this game. Micro is precisely flanking, concaves and selection of battle ground. Macro is "I have more stompy stuff than you". Whenever people criticise micro they are criticising tactical optimisation. People should either use the terms correctly or stop talking about it because it seriously undermines their argument.

What you want out of PA is more automated micro mechanics. Formations for instance are an automated micro mechanic. If you could make your troops easily fall into a concave that is a mechanic designed to make micro less fiddly. However a neat concave painting tool is still micro. The fact the tool decides to move your individual troops does not make it less micro.

The only way to get rid of micro is to turn the entire fighting part over to the AI. For instance take the Total War series. The entire real time aspect of TW is micro. Nothing is being done except actions to optimise the efficiency of what you already have.

Essentially a macro only PA would be like TW while automating every battle.

Ironically I could create a game like this and still turn it into a click fest. If I give you factories that can't queue and mexs that turn off every 5 seconds. The units when built will be automated. Suddenly you are clicking like crazy and not doing any micro at all. All of that is macro. Building stuff but the person who can click fastest and most accurately will win. Even though we've eliminated micro.