The good side of Micro

Something I want to do: Define how I lay out my base structures. Even make a wall out of solar collectors if I think it would strategically help.

Something I don't want to do: Babysit every skirmisher to always engage at maximum age, or every fast unit to move in a zig-zag to reduce getting hit by slow projectiles, when simple scripts can do this.

There needs some advantage of micro control in game. In fact, there must be some element in game even if the Uber do not want to make it. Anybody remember that a mech marine could destroy few strikers alone?

I don't think splitting attention is quite the same as micro. Micro implies you are carrying out some action whereas attention can be just keeping track of what is going on.

I believe the intent is that it takes a lot of attention to perform micro, and the skill in being able to spend as little attention as necessary on any one activity is what they meant with the splitting attention.

By automating micro, it reduces the attention requirements for that specific activity, and that the unspent attention can be used elsewhere. It's a similar discussion to the difference between a rear (RWD) and front (FWD) wheel drive car.

A tire has a tangible limit of traction. With a FWD car, the front tires have to split duty between providing propulsion and also turning forces, so the amount of traction that can be dedicated to either is always less than the total value. With a RWD car, the front tires are only providing turning forces, and the rear tires are only providing propulsion forces, so each activity can use the full limits of the traction for those tires.

I think that's the same concept that folks are trying to talk about when it comes to attention and it's relationship to micro.

I never said, that micro adds strategy. I said, it has strategic value, which it gets through its tactical nature.

I am impressed. I have never seen anyone sitting so high on his horse, that he wrote multiple paragraphs about a motivation I apparently have, even though I never said anything about why I think that way. Basically, you assumed the crap out of me and you were ridicoulously wrong.
Now, that we both looked down on each other at the beginning of our posts, let's talk without subtle insults and putting words in each others mouth.

I don't care about "the old ways" or any other kind of tradition fallacy. By the way, you sound a lot like "It's new and therefore better.", which also is a fallacy.

I admit, that my opposition to scripts/advanced AI is ideologically influenced, but in a different way.
I want players to be responsible for their actions. A player who wins should have earned it and a player who loses should have deserved it, otherwise winning or losing doesn't hold any meaning and basically is worthless.
Something like a harass script can do a terrible amount of damage, without a player lifting more than the finger for pressing the activation hotkey. While you could argue, that the player receiving the damage deserved it, since he didn't defend properly, I don't think you can argue, that the player whose script did the damage deserves to do that well.
Giving each side scripts of equal strength also doesn't make it any better. That's not PvP anymore, it's more like Player directed AIvAI.
This may just be a bad example or driving the idea of scripts/AI too far, but it definitely shows the danger of going too far with it. Going too far devalues what a player is doing and gets dangerously close to completely detaching players not only from what is going on, but also from the fact, that they are facing each other. In other words, it makes the interesting part of PvP -that you are actually facing a human, not an AI- less and less relevant.

De-emphasizing micro instead of automating it avoids trying to strike the balance between too much AI and having to do annoying tasks to be good at the game.
Seeing how missing it in favor of too much AI has the dangerous potential to ruin the game (from the perspective of my view on PvP) avoiding to fiddle with it looks like the right choice to me.
Also, you don't have to code the scripts/AI, you simply have to design the game well.

You can't change the fact that harassing is just one decision being carried over many clicks. A more swift excecution of the same decisions the player wants would make everyone less reluctant to perform such moves. The onus then is on the defender to figure out how to prevent such attacks. The game is balanced less to mechanical tedium and more to stroke of genius.

Hold on, are you talking about something like click a button and the AI automatically harrasses the enemy base? Because that would be a little ridiculous. (Although I wouldn't say it's a bad thing) I think people are talking more about kiting units and moving back and forth to dodge bullets.

Harassing is many, many decisions in a short period of time. Basically, since you are using a small group of units, and your opponent has an advantage of units and commander nearby, you need to be really careful, smart, and attentive to make sure your units aren't getting into harms way for a second while they damage the opponent's infrastructure.

This is true even if the units automatically kite and dodge, sometimes especially true, since your opponent will be acting to exploit the flaws of the automatic kiting and dodging. Your opponent's decisions in this case require you to make active decisions to keep everything alive.

I speak about this from my experience with Zero-K, which has automatic kiting and dodging, and quite effective automatic kiting and dodging at that.

Also, I've never seen a player try to exploit the flaws of the automatic kiting and dodging in Zero-K. There are much more important things to be done.

I've found that nothing reduces the feeling of facing another player more than an excessive need for micro. But that's just a difference in view.

I have a feeling a lot of people who argue against replacing micro with AI are thinking that it'll allow players to sit back and relax and let the game play for them. Which is obviously not what's going to happen. Instead, players will need to be focussing on other stuff all the time, that's more important on the grand scale.

Moving from minor skirmishes to major battles is the basic idea behind getting rid of micro.

So excessive micro is bad and having an AI do everything is bad. That's what I was talking about in the part about striking the balance. And that you can avoid trying to hit something you can only miss, by ditching advanced AI or scripts and simply designing with putting no emphasis on micro at all and letting the big scale of PA work its magic.
Big scale stuff basically is the natural enemy of micro, so if the design focuses on that (which it afaik already does), bad micro will automatically go away.
Advanced AI isn't only what I see as the wrong way to go about it, it is also unneccessary work and therefore a waste of time and money.

The advanced AI will in large part be built by the community, who will hook these scripts into their UI.
The balance will automatically shift around, probably in favor of more big gestures by players and more AI scripts doing the little stuff.

Sorry, I am talking about the game made by Uber, not the AI mods people will make for it.
The base game is what is played on the ranked ladder, people don't simply get to hook in their AI scripts. Competitive PvP is about who is better at playing the game, not who has the better scripts.

Also, good job at simply ignoring everything I said about the impossible task of getting the right balance between too much and too little AI. The point isn't to have it shift around, the point is getting a good one to stick to. (And as for every balance you can't get it right because there will always be people who are not satisfied.) And, to reiterate it: It is pointless to care about something you can't get right, if you can simply make it unimportant. In PA this balance is unimportant because of the big scale and because nobody wants to make it important. Don't fix what isn't broken, simply keep it that way.

Actually from what I heard, the game will allow you to insert UI hooks into the base game, which would probably mean you can use them in the ladder as well. Although I'm not 100% certain on that.

That is questionable. In fact, by not using support AIs or automated controls for reducing count of routine operations and handling low-level tasks, developers shift gameplay away from strategy towards tactics. Reason is simple: human mind's ability to control things is limited to 7-9 objects at time. To overcome this limit, human can, for example, concentrate on subtask, thus keeping object count within range, but losing grip to other subtasks. Other option is to delegate subtasks to someone, giving them high-level instructions and letting them to handle details - that's exactly how a real military works: commander of each rank have strictly limited count of direct subordinates that he can effectively guide in a combat situation without losing the whole picture of task that is assessed to him. In somewhat simplified model, army is divided into small cells bound together by a hierarchical control system, where every formation have a task, an area to operate and underlying formations or units to execute subtasks of root task in some points of given area. Commander's job is to divide root task onto those subtasks with accordance to accessible forces, designate key locations for performing subtasks, assign subtasks to correspondiong lower-level commanders and oversee the execution process, ajusting or reassigning subtasks and key points in process. By overloading commander with strategy-irrevelant routine tasks, one hampers his ability to control the course of battle. In real life, such a situation will cause a defeat of distrated commander's forces. In game that will result in degeneration of strategy down to tactics - player who able to do more of those tasks in a fraction of time will win. So having automated UI and virtual subordinate AIs for lower level tasks is not only handy, but also strongly correlates with gameplay aspect of strategy as a high-level planning.

Thanks for continuing the theme.

It's only a fallacy if it's wrong, either way one slices it.

"I want" is an opinion, based on a feeling. "Should have" is a qualitative argument. "Meaning" is a qualitative statement.

"Deserves", "better", and "too far" are all qualitative.

"Value" and "interesting" are qualitative terms.

"My view" and "to me" are about the most easily identified opinions in this whole thing.

"Missing" is an opinion, not a fact, because it's based on your subjective scale of what types of activities a player should be doing.

Yay more opinions!

MOAR opinions! A fact is that if two players are going against each other, it's PvP. It's irrelevant whether one believes that there is too much or too little involvement to achieve a specific outcome, because that's an opinion.

You've proved my point, by identifying that what makes it "impossible" is people's satisfaction (their opinions), and not whether it's a quantitatively good idea or not.

I had to leave this full.

If it's pointless to care about something one can't get right, why do you care about the opinions of folks when an opinion cannot be right or wrong?

Why should your opinion hold any more or less weight than anyone else's?

If nobody wanted to make it important, nobody would be arguing to use AI/scripts to replace micro to achieve the same ends via different means. People are, so this assertion is factually incorrect.

So what really is your purpose in this thread? Are you upset with how people came to their opinions? Are you upset that people aren't giving your opinion its due reverence?

These are much more quantitative arguments.

AI/scripts are more efficient than humans at achieving the same task, and dmii, you don't seem to understand or be good at explaining why there is less micro in TA, yet insist on arguing your opinions as a fact. You also seem to miss the fact that there will still be strategic decisions that would require either micro or AI/scripts to accomplish so long as we can deal with units on an individual basis. These include:

1) Area repairs - should the player be required to spend time maintaining specific units or buildings in an army of thousands on multiple theaters of conflict?
2) Area assault - do things like bombers focus their attack one unit in an area, or spread it out? Should units focus on weaker units or stronger units?

Basic AI/scripts were there in Total Annihilation in the form of movement and firing orders. They were more advanced in DoW, where the movement scripts allowed "follow a short distance and then come back" in addition to "sit still" or "chase enemy till you die". In the same vein, movement micro in TA is significantly smaller than in SC because the ability for a ground unit to respond to an input was much slower than the speed at which a player could provide that input. Airplanes not so much, so the "balancing insulation" wasn't even consistent across the board.

As such there will certainly be some micro that never existed in TA, or PA, because of that type of insulation from the player, but insulation from player control is one of the reasons you are also arguing against AI/scripts in the first place. Is having no decision to make preferred to having an AI/script carry out that decision?

Quick question,

Has anyone here played a game called AI War: Fleet Command? It's a very grand-scale RTS, with multiple fields of battle at once, where the player has to be able to focus on large-scale strategy, but everything still happens in real time supcom/ta/starcraft/whatever-style. I highly recommend trying out the demo, mainly to have a look at a completely different approach to interface and reduction of micro.

This game has what I consider to be a good balance between micro and macro, mainly in its "Free-roaming defender" mode. Idle units can be ordered into FRD mode, which means that they automatically pursue enemy units within your territory. They even prioritize targets depending on what they're good against. This means the player can focus on his overall strategy instead of having to personally supervise his defenses all the time. However, this doesn't mean that the player can simply leave their defense to the computer, since enemy units might just run the blockade strike at a vital target before retreating. Maybe the enemy is making a diversion, drawing your defenses away from what they need to defend.

So, even though the interface takes a significant load off of the player's mind, the player is still needed to make important decisions (such as what actually needs defending).

And that's not even going into the whole "Control menu" with options for auto-free-roaming-defender mode, engineer auto-rebuilding, limitations on what engineers can assist, target prioritisation...

EDIT: Link

No it is not. In the context I said that, it meant the AI is doing everything with everything meaning everything. At that point I was simply saying, that both ends of the spectrum, AI doing everything and AI doing nothing are bad.

For a competitive PvP to be meaningful there has to be an even playing field. I can't really imagine, that the ladder would allow something that could provide an unfair advantage.

As for you chronoblip: I was trying to explain why I hold my opinion on advanced AI, which is based on my understanding of what PvP is about. Of course those are subjective points, but I thought trying to explain why I think that way would at least cause a little understanding for my position.
But I guess now this thread has derailed more towards people talking past each other. (Also I worded some things horribly, so I guess that's my fault.)

Also, I didn't even see area commands as scripts. To me those hold the same position as a simple move command. Same with setting unit behaviours.
This looks like a severe case of misunderstanding caused by using the same words for a different thing.
Scripts to me are more something like automated harrassing, which I see as a way more complex action, than telling an engi to repair everything in an area.

But where do you draw the line?
Correct me if I'm wrong:
In Starcraft 1 you couldn't use anything but the standard hotkeys and no macros.
In Starcraft 2 you can customize hotkeys but you can get banned for using macros.
In SupCom you can use UI mods but not much else.

In the Spring community advancement of unit control and automated unit behaviour is usually embraced by the game developers. If something is abusable by a script the game developers are more inclined to change the game rather than ban the script.

I have yet to see something like an automated harassing script in Spring. Harassment in Spring games can be very complex. You can hide behind enemy buildings to take cover from their defences, your own units block your Line of Fire, kiting sometimes, overrunning the enemy, spreading out your units to avoid splash damage or exploding buildings.
Making a script to take advantage of this could be very heavy to compute, be very complex to write or both.

I like what masterkane wrote:
Once the AI actually controls your units you can start playing like a general.
But I think we are far from there with such complex gamemechanics as PA will have.

So give everyone access to the same UI improvements. You could just make all scripts available and let players pick whichever ones they like.

In case any of the developers are reading this thread, can they give any sort of insight on whether UI scripts will be allowed in regular matches? It sounds like an interesting feature but I can't remember if it's been mentioned by someone in the team or just as a suggestion that I'd really like to see.