Attack Move

[raymundo]

I think it would be a great feature if we had 2 types of attack move. One is the regular attack move which goes for the next available enemy unit or structure, the new type would be just against enemy units it wouldn't target buildings. This is already a defensive game, lets say each player has 20 shock troopers. The enemy attacks you. Your units are in the base, so you do attack move and they start firing at the shock troopers. The enemy shock troopers will always lose because you either can't click that fast, or they will just attack buildings as well as some of the shock troops. This kind of thing wouldn't just be for shock troops obviously but you get the point. A unit-only attack move would be a great addition to the game. This kind of situation also happens when a player has an oil derrick, whoever has the derrick can leave a couple of troops by it and be able to defend it against a great number of rifleman because of building targeting.

[newwe]

Great idea, the oil derrick thing really annoys me.

[BaronOfStuff]

To be quite honest I think that it's dumb when units auto-attack any structures that aren't defences. Why? Because what if you want to secure certain bits of an enemy base for capture? It becomes a load more work than it ever was in classic TD/RA since your stuff will just flatten everything in the vicinity, unless you explicitly tell it to not shoot everything in sight. Setting unit stance to "return fire" isn't exactly ideal either, as they'll just let Ore Trucks roll on by, along with any enemies that aren't directly attacking them (which is a real shitter when your units technically outrange their target, but don't/won't do anything until they're shot at first).

[af]

Rather than a new type of attack move, I'd prefer to see per-unit "priority targeting". This way when you attack move, each unit would target enemies in their vision radius according to their strengths (and de-emphasize things like oil derricks).

For example, a Hind is great against infantry, but not so much against vehicles or defensive structures. Yet when you attack move them into a crowd of enemies, often they will waste all their ammo on a light tank rather than picking off infantry like you'd want. So maybe the Hind could have a targeting priority list as follows:
1. Infantry
2. Vehicles
3. Buildings

Likewise, Flak trucks are effective against air units and artillery/V2, less so everything else. Their targeting priority could be:
1. Air units
2. Artillery/V2
3. Infantry
4. Tanks/vehicles
5. Buildings

In my opinion this could remove a lot of fiddly micro from the game. Maybe some people like that element, but I'd rather emphasize strategy than mouse gymnastics.

[raymundo]

I agree, priorities per individual unit is a great idea!

Edit: even if it was simply prioritize units before buildings i'd be happy.

[noobmapmaker]

If you program an AI with that priority targeting then it already has some microskills that surpasses most players!

[Murto the Ray]

I don't think an extra attack move would be a good idea as i feel it would be unnecessary and wouldn't effect play as it is at the moment - i.e. roughly 60% of players wouldn't use it. It has already been suggested that an automatic targeting system be implemented but i am also opposed to that as i think it would probably hinder performance and also make the game slightly easier - this may seem like an odd concern but making the game slightly easier with every feature added only makes the gap between veteran players and new players smaller, meaning winning is more random than skill based.

What i suggest is that the stance system is reworked like so:

AttackAnything - works like defend does at the moment
returnfire - no difference/removed
holdfire - removed/no difference
Defend - units only fire at units and defences - no buildings

Such a reworking would actually encourage use of the stance system (rather than just putting every unit on defend), deliver what you want and still require enough micro to retain the skill gap between this system and the old one.

[raymundo]

Murt those changes would be good, as long as they aren't going to target a building I'm okay with defenses being targeted

I don't think an extra attack move would be a good idea as i feel it would be unnecessary and wouldn't effect play as it is at the moment - i.e. roughly 60% of players wouldn't use it. It has already been suggested that an automatic targeting system be implemented but i am also opposed to that as i think it would probably hinder performance and also make the game slightly easier - this may seem like an odd concern but making the game slightly easier with every feature added only makes the gap between veteran players and new players smaller, meaning winning is more random than skill based.

What i suggest is that the stance system is reworked like so:

AttackAnything - works like defend does at the moment
returnfire - no difference/removed
holdfire - removed/no difference
Defend - units only fire at units and defences - no buildings

Such a reworking would actually encourage use of the stance system (rather than just putting every unit on defend), deliver what you want and still require enough micro to retain the skill gap between this system and the old one.

This is t the best idea. I'd be happy to see the units prioritizing units/defenses over the buildings, and you are right the only stance that gets used is defend or hold fire. Not sure what you mean by returnfire/holdfire removed, but I believe holdfire is very important to the game when making stealth attacks.

[Murto the Ray]

This is t the best idea. I'd be happy to see the units prioritizing units/defenses over the buildings, and you are right the only stance that gets used is defend or hold fire. Not sure what you mean by returnfire/holdfire removed, but I believe holdfire is very important to the game when making stealth attacks.

All i mean is that either holdfire or returnfire would be kept because not both are needed but sometimes you do want your units to fire back - deciding which should be kept is really a decision for everyone to makeand personally i'm split between them.

[af]

It has already been suggested that an automatic targeting system be implemented but i am also opposed to that as i think it would probably hinder performance and also make the game slightly easier - this may seem like an odd concern but making the game slightly easier with every feature added only makes the gap between veteran players and new players smaller, meaning winning is more random than skill based.

A smaller micromanagement gap between veteran/new players does not mean victory is more dependent on chance. Rather, other concerns like strategy/tactics/teamwork become more important. Those are the things I'd rather have emphasized in an RTS than micro, but YMMV of course.

The performance concern is pure conjecture at this point– sure it would require extra logic, but there's no proof that there would be a noticeable performance impact.

[Murto the Ray]

A smaller micromanagement gap between veteran/new players does not mean victory is more dependent on chance.

Well actually it does. For example:

A Hind is great against infantry, but not so much against vehicles or defensive structures. Yet when you attack move them into a crowd of enemies, often they will waste all their ammo on a light tank rather than picking off infantry like you'd want.

So, as things are a veteran player will be able to micro his/her hinds (perhaps using shift+ctrl) to take out the infantry without touching the tank - thus stopping the incoming assault. A new player, on the other hand, attack-moves his/her hinds into the group and damages the light tank, failing to stop the assault - thus ending up losing buildings and eventually the game because the group was not stopped in time.

With your mechanic the new player and the veteran player both have the same outcome. This means that the gap between the veteran player and new player has become thinner.

Other concerns like strategy/tactics/teamwork become more important. Those are the things I'd rather have emphasized in an RTS than micro, but YMMV of course.

I understand that you have your own opinion on this but micro is essential 50% of an RTS game. Due to the similarities in people's build orders and unit production simply microing better than your opponent whilst matching their macro will result in a win for you. Micro is where games can be won or lost and how the most exciting comebacks are made.

Maybe some people like that element, but I'd rather emphasize strategy than mouse gymnastics.

I'm sorry but i can't help thinking that the RTS genre just isn't for you because as i've said, about 50% of it is "mouse gymnastics".

The performance concern is pure conjecture at this point– sure it would require extra logic, but there's no proof that there would be a noticeable performance impact.

Well think about it like this: A large number ( lets say 100 ) infantry units run into the enemy's base and start shooting at a building. Then a ranger comes from no where and drives into the middle of them. Each infantry unit then need to do a comparison between the ranger and the building at which point they start shooting at the ranger. 100 simultaneous comparisons. And that gets worse as you start to add helis flying in from the top and tanks coming in, new defences being made etc.

[raymundo]

The shift control method of attacking is a valid tactic and I feel a player should be rewarded for this skill, so I actually change my mind on what I originally discussed. Like Murto said, possibly redesign the stances so in one stance the units will attack units or defenses before buildings. That is only change There shouldn't be a hierarchy of units they will prioritze, plus there can only be so many units within distance. As it stands I don't believe even a veteran player would be able to attack just 10 infantry with 20 if those ten are standing near a oil derrick. Maybe we can test?

[af]

So, as things are a veteran player will be able to micro his/her hinds (perhaps using shift+ctrl) to take out the infantry without touching the tank - thus stopping the incoming assault. A new player, on the other hand, attack-moves his/her hinds into the group and damages the light tank, failing to stop the assault - thus ending up losing buildings and eventually the game because the group was not stopped in time.

With your mechanic the new player and the veteran player both have the same outcome. This means that the gap between the veteran player and new player has become thinner.

Even with the proposed changes, a veteran player using Shift-click will get a better result than the person using attack move (eg. Hind attacking rocketmen vs riflemen). I'm just advocating that attack move shouldn't be worthless in this scenario (as it currently is). Being able to micro to attack specific units is still a big advantage.

Also, there's the extra benefit that the AI should be more capable with this added behaviour.

I understand that you have your own opinion on this but micro is essential 50% of an RTS game. Due to the similarities in people's build orders and unit production simply microing better than your opponent whilst matching their macro will result in a win for you. Micro is where games can be won or lost and how the most exciting comebacks are made.

Right, like I said, YMMV, and every RTS player has their own preferences for types of gameplay. Personally I prefer high-level strategy over seeing who can issue the fastest and most precise right clicks. Admittedly it's partially because I'm not great at micro either. Of course with all else being equal, the better microing player should win. This proposal would not change that.

Well think about it like this: A large number ( lets say 100 ) infantry units run into the enemy's base and start shooting at a building. Then a ranger comes from no where and drives into the middle of them. Each infantry unit then need to do a comparison between the ranger and the building at which point they start shooting at the ranger. 100 simultaneous comparisons. And that gets worse as you start to add helis flying in from the top and tanks coming in, new defences being made etc.

I think the devs can probably come up with a better implementation than that In cases like the above you'd want to cache the result of each unit type's targeting and re-use it for any neighbouring units of the same type.

[winftw]

I think the way of westwood is the best: units auto-target all enemy units and base defenses but ignore structures. When attacking an enemy base you generally want to either snipe a key structure (thus giving an explicit attack command to that structure) or kill enemy units first and then kill enemy buildings in order of importance (again giving explicit attack commands on the structures). There is absolutely no reason for units to ever auto-target structures because they will be explicitly targetted by the player regardless. And if you want to capture buildings with an engineer the engineer must be allowed to work in peace.

No target priorization is needed if units would just ignore buildings by default. Also I agee with Murto here. Micromanagement is good and should be equally useful as high level strategy.

"Hold fire" stance is useful and should be kept. I use it occasionally on riflemen that manage to get into enemy base so they can scout around quietly before they eventually get spotted. And when I use V2 rockets I always set them on hold-fire because otherwise they waste their rockets and then get killed for nothing because it takes 50 years to reload the damn things. Also useful for any stealth units and submarines.

And btw performance is irrelevant. If there is a real need for some feature it can surely be optimized into fast enough.

Here is my suggestion:

AttackAnything - removed
Returnfire - removed
Holdfire - works as before.
Defend - Units only fire at units and defences - no buildings. The default stance. Disallow any movement. Attack enemies in range but don't take a step in any direction and don't chase anyone.