The Case for Removing Radar

I am about to make a rather unusual argument. However, after quite a bit of thought about TA and SupCom and how the intel game played out under both systems, I have concluded that radar giving away unit positions and allowing targeting at range just destroys the entire intelligence dimension of the game and replaces it with just "build radar." Imperfect information is an integral part of strategy games, and as a result acquiring information should be both expensive and difficult, and the information available to players should be highly asymmetric based on their decisions about where to move, how to scout, and how to deny scouting.

Radar (at least as we know it) should simply be removed from the game.


Radar in Total Annihilation

In TA, radar alone could not be used for targeting at range. The primary function of the radar tower was to show blips to the player on the minimap, giving the player information about where the enemy was moving, even though the units under that player's control would only fire on an enemy they could actually see.

In conjunction with the quite short sight ranges of most units, this made long-range units both extremely powerful and also quite difficult to use effectively. In order to utilize their range effectively, the player must deploy units further forward to get eyes on a target. Blind fire into the fog was possible, of course, but was highly likely to be ineffective, especially against mobile units.

However, TA also had a separate, and quite expensive, advanced structure which allowed long-range weapons to fire on radar signatures. But under most circumstances this structure was not in play, and units could not fire at radar blips. In the late game, with multiple fusions in play, dedicating a fusion to powering the radar targeting facility made sense just to save APM for executing long-range fires.

But, long story short, radar targeting was not a significant component of the primary TA gameplay. Units primarily relied on vision for shooting. As a direct result, there was a lot of variability in engagements due to vision and range differentials, and a lot of close combat.


Land Radar is Boring

Radar detection of land targets just gives players too much information too cheaply. The omni sensor in SupCom is a very compelling proof of the fact that too much information available to both sides is very detrimental to gameplay. All weapons can use their full range efficiently, no scouting or forward interaction is necessary. Long-range units dominate on such a battlefield, especially in groups, unless prohibitively priced.

Land radar also makes it very difficult to execute small maneuvers, since a relatively small amount of long-range firepower will use radar targeting to efficiently eliminate an individual unit or small group for free.

Worse still, if you are sure you can utilize superior range, you gain a huge edge by even a small range advantage. The result is binary one-upmanship in an arms race over who can construct the longest-ranged guns. The critical mechanic behind this range arms race is the expectation that you can always use your full range. A one-time investment in the best radar available increases the efficiency of all units whose ranges exceed their vision immensely.

However if you must use scouts to gain vision of the target area, your ability to use your range is no longer assured. Scouts can be killed, recon can be denied, but radar reaches everywhere. If the enemy has bigger guns than you, you should try to deny them vision. It doesn't matter if you're within their range if they don't know where you are. In fact you might even be able to sneak up on those guns and destroy them for free using raiders, or with shorter-range guns because your scout found their artillery.


Replacing Radar with Vision

However players still need tools to get intelligence. My point is simply that radar is just way too cheap, effective, and worst of all convenient. So instead of having a tower that gives a huge range of "blip detection," I propose to create a new category of units; recon units. Recon units have excellent vision, making them very important assets on a battlefield composed primarily of combat units with poor vision.

In addition, I propose that the radar tower itself actually provide a significantly large area of vision, as it did in TA. Nowhere near its current blip detection range, of course. But a considerable improvement over all but the most expensive recon units.


Using Vision

I expect some people may balk at the idea of relying entirely on vision instead of the highly convenient radar, purely because of the huge decrease in available information and accompanying security. That's a scary thought. However, remember that your opponent will be in the same position. This opens up the possibility of making sneak attacks, of hiding groups of units close to the enemy's base, and of mind games like pretending to defend an area where is really only a skeleton defense, or feinting an attack to provoke a reaction. Without radar, both sides have to make educated guesses using their own vision, which the other side can exploit to give a misleading impression.

Furthermore, given the number of units that will be in play in PA, relying on unit vision actually does not necessarily mean less information. However the lack of radar will drastically change the way your forces should be deployed. A defensive line provides not only a barrier to the enemy, but also an unbroken line of vision. An army of units arranged in a line, or a circle around an enemy position, can create a very different intel picture for one side than the other based on the position of both sides' units.

Instead of relying on a single structure to tell you where enemy units are, you are going to need to deploy your units in a shape that covers area, not just a blob that shoots at anything the radar detects. Spreading your units across the map gives you more intel, while consolidating them together gives you more combat strength. And the same is true for the enemy.


Recon Units

The removal of land radar means players need effective tools for gathering intel. To fill this need, recon units are necessary. While one scout has nowhere near the sensor-power of a current land radar tower, many scouts strategically placed around the map certainly do. The most important difference is that a scout must be in the area it is observing, and can be destroyed.

Historically, your basic 'scout' is an incredibly cheap unit, barely armed, with negligible armor. This unit role will obviously be very useful, and even a potentially integral part of any land war on both sides. These units would be invaluable to determine if a mex is occupied, if an enemy army is coming from a particular path, and to extend the range of your combat forces.

However not all recon units would be such minimalist scouts. A variety of different recon units would be quite useful, potentially even including some quite advanced and/or expensive recon units. Heavily armored recon would integrate well into an armored column. Fast recon with strong weapons can double as raiders or fire support. Stationary flying recon becomes extremely powerful, but also vulnerable. And tactical use of aerial recon by overflight with a plane can give a lot of information quickly. Such air scouts might range from expendable drone scout to a highly sophisticated spy plane.


Radar as Anti-Air & Orbital Targeting

For land warfare, vision has a lot more meat than radar does. Without radar tucked safely behind the lines both sides need to scout and will have to worry about what is hiding in the fog. Players will need to adjust how they use land units to try to obtain more comprehensive intelligence than their opponent.

But air units are different. They are so fast that they can fly in and out of vision in an instant. Long detection range makes sense in order to have sensible engagement ranges for such fast units.

Therefore, I propose that radar be assigned the role of detecting flying units at long ranges, and allow radar targeting. With this more specialized capability, a radar tower can be cheap and still have extensive range. But it will still be useful on the ground due to its low cost and large vision radius. But radar's primary purpose is to allow your long-range anti-air to shoot down enemy aircraft from far away.

To build upon this system, I propose that radar also detect orbital units above the planet's surface. Radar detection is necessary to use surface-to-orbital weapons, unless you have orbital units of your own with radar that are able to detect other orbital units instead.


Conclusion

Radar targeting has become a fairly standard-issue feature of most games that qualify as 'children of TA.' However TA itself actually relied much more on unit vision, with the minor exception of the radar targeting facility, only available in large-economy situations.

Many games have looked at the radar targeting facility and gone "Why is that a structure? Make it standard." However I think this was the incorrect approach because radar makes too much information available, and does it too cheaply and conveniently.

Recon units and the entire intel dimension of gameplay should be much more interesting and important than merely placing radar towers. The fog of war should be oppressive, encouraging players to spread out and get vision across the map instead of relying on long-range radar. Fighting for intel using units in the field has the potential to be much more interesting than static towers with huge range.

Radar could still serve to target anti-air weaponry, where detection at distance is both necessary and proper due to their speed. And it might also detect orbital units, because some feature must detect orbital units from the surface, and radar is the most logical candidate.

I don't really like this. I'd rather not have this game turn into Starcraft where if you don't scout every 30 seconds you get swamped by an enemy you never saw coming.

I'd rather see radar range reduced than have it eliminated completely, it's an integral component of creating a far-reaching strategy that takes the whole planet into account. You are simply physically incapable of keeping the entire planet scouted so you know what's going on, otherwise.

It'd introduce unnecessary micro.

This is a good point. It does create the potential for micromanagement of scouts.

However micromanagement of scouts is unlikely to be as pernicious or as dangerous in creating a one-sided battle as, say, micromanagement of a Colossus in Starcraft 2. More scout micro conveys an advantage, but is it that same kind of direct combat advantage that a well-micromanaged unit in Starcraft gets from killing far more enemy assets? Or is this more of a soft advantage of being more informed about what is going in more places on around the map?

Players with faster hands and better micro will be able to manage more scouts operating as individuals or in small recon elements. But a player might also just as easily choose to group their recon units with their armies and use minimal micromanagement. Does this give players with faster hands an advantage? Yes. But there is absolutely no way to avoid this to some extent, because being able to do more in-game actions is strictly an advantage.

I don't think scout micro is 'forced' or is likely to cause players to be forced to micro in response to an enemy that is heavily microing their scouts. But yes, scouts certainly can potentially consume more APM than radar towers.

It isn't a problem once you got a good balance on the costs of radar and radar jammming.

I'm willing to give it a try. I actually play TA without actually utilising radar until about 20 to 25 minutes in. I build it of course but TA taught me not to rely on radar for anything beyond an early warning system until the extreme late game.

I'd like to bring up the way TA: Spring did radar. Unit blips had a radar wobble, that would make long-range weapons extremely ineffective against them. It made for really interesting gameplay, where scouting for artillery units actually mattered.

A good alternative to this is making Artillery highly inaccurate when firing into FoW, which is another thing I like.

I was originally behind the radar-wobble mechanic as in Zero-K (I don't think all Spring mods do radar wobble). However, if you really think about the function of radar wobble, the underlying purpose is to turn radar into a minimap-awareness feature and not a long-range targeting feature. It is sort of an attempt to duplicate TA's use of radar as information to the player without having units fire at radar blips. The difference being that they will fire at radar blips, but may be firing at the wrong place unless you have a visual as well.

It seems like an attempt to encourage scouting by requiring line of sight in order to shoot accurately while still allowing long-range weapons to shoot at blips. But the result is quite similar to having no wobble, only it takes more shots to have the same effect.

Wobble is a definite improvement over perfectly accurate targeting information. But it seems like an attempt to recapture the gameplay of just not being able to target using radar at all, while still allowing radar targeting.

Wobble seems like it would work well for very short distances outside of vision, but as the detection range increases, the relative significance of the wobble drops drastically, to the point that you really don't care about the wobble since the radar lets you shoot at long range with impunity. The probabilistic component introduced by the wobble doesn't really change the value proposition of the radar and long range weaponry combination.

Perhaps it would be better just to do away with radar targeting completely and require scouting, full stop.

To be honest, I find that with basic radar the way it is, I usually use scouts or fighters in a patrol ring from factorys to cover my intel.

If my memory is correct, you can make units fire at radar blips by clicking them on minimap, and it is the most micro heavy part of TA.
Radar in TA-like games mostly function as a weaker type of vision that tell you the positions of units but not the types of units, cheap scout units spam with real visions could be a bigger hinderance to surprising attack than that, to deal with radar you only need a mobile radar jammer, to deal with scout units you have to kill them before they find out you are attacking.

Every single unit that you build in TA-like games needs to be destroyed since they all pose a threat. Having 'weaker' vision doesn't really matter... you were going to shoot it anyway.

And if the most micro intensive thing that TA did was make you work for your long range super-effective artillery to be as good as it was, I'd say that's no bad thing.

It is not that much a difference if you have enough apm.
How threatful/important a unit is could vary a lot, and they all look the same as radar blips.

This is correct. And it is for this reason that I did not propose to exactly copy TA's system of showing radar blips, but not having units attack them.

TA made the (understandable) mistake of assuming no player would be hardcore enough to manually aim every weapon to manually attack each enemy blip on the minimap, and allowed this. And it is for exactly this reason that you cannot allow a large land radar to even tell the player where the blips are; because a sufficiently dedicated player can then use that information to manually attack ground at that location.

The best solution is to remove such cheap long-range detection and targeting from the game entirely, and to intend players to use units that are more interesting than a static tower for recon. Units that cost resources, which can move around the map and perform various different tasks, interact with other units, fight enemy units, and which can be destroyed to deny scouting. Radar towers do basically none of these things; they just allow your long-range guns to engage blips with impunity.

Take the large land radar out, and put effective recon units in instead.

90% of the OP is about removing radar.
10% of it is about adding radar back in to counter air and orbital units.

Man Radar give the information about a Army, Fleet that is moving toward you...
Like in 2WW British used it to detec Nazi Bombers before it pass the English Channel, they aren't able to know if was Bombers or a bunch of fighters.
In Vietnam the USA launch some F-4 to kill the fighters that are killing bombers, and work well.

so if you want to know what is, one word Scout, not remove the radar.

you already have some scout units.

Radar do interactive with other units, they can be killed by attacks, and in some TA mods we have professional radar killing weapons.

All units with some kind of vision has its intel value in a RTS game, you might think radar tower is too cheap, effective and safe for its position detection value compare to other units, make the position detection part of the game lack diversity, or you just think position detection in game is just too easy overall, which makes keeping units' positions unknown to enemy too difficult.

To solve these problems there is no need to remove radar, you can just make the radar tower more expensive, more deny-able, more killable or having some cheap ways to produce fake radar blips, thus making other ways of position detection more useful in game and nerf the position detection overall.
Or you can remove the radar tower, reduce all units' los then give them all a low range radar.

There are RTS games that actually show everyone every units' positions without the need of radar tower, this does not cause any problem if any player can easily deny or deceive this default vision.

Can we just make arty not target radar'd units?? That would solve 90% of the problems you have with it.

I personally hate it when someone camps my base with units. I have to waste time going after them and circumventing them, and usually I just build a Holkins in the center of my base to eliminate the threat. Its really annoying, and its great for rushing/scaring new players, but more experienced ones will just shake their heads and attack your eco while you are busy trying to penetrate his defenses with a smaller production line.

That's why building proxy bases AROUND, not NEAR your opponent is important.

Its really the only way to counter fast moving air and orbital units. Radar. Wonder why

One thing to note is by making the first level of detection (unit position) easy, the second level (unit type) might actually become more difficult, because with the first level the player can get aware of enemy's second level detection earlier and intercept or avoid it.

I hope land based radar is eventually affected by terrain and positioning.

I like this more than simply removing radar. Radar adds lots of strategic elements to the game.

Making the radar wobble and/or make it inaccurate the greater the range would take a chunk out of the power of catapult spam – which annoys the hell out of me.

And making it so long range defensive structures are inaccurate when firing at radar designated targets would also be a huge benefit.

I like that route a lot more than removing radar. Removing radar gives too much of a benefit to micro.

And making it so the farther away from the radar tower the unit is, the less accurate the intel is, would also be awesome. Could even make it so the basic radar is the most accurate, advanced radar being less accurate, and the radar satellite being even less accurate would also be pretty cool.

Something like this would also be pretty awesome, though possibly a technical nightmare. Trees interfere with micro, giving them a strategic ambush benefit. Placing units and buildings behind mountains also making the radar more ineffective would also be awesome. Lots more strategy involved, and not much micro.

Either way, ledarsi does bring up a good point that catapult/holkins/pelter spam is very powerful and is difficult to win against. Whoever builds the most of them has a significant advantage.

We had another thread about radar. There needs to be diversity, limits, and counters to radar. Then radar can exist, and one can naturally learn not to rely on it heavily because of its limits.

I liked lowering the range of basic radar to just past basic artillery range, changing longrange radar to do something else like t1 radar range but directional (and thus twice as far from danger), making scouts radar immune as well as possibly other few antiradar, and lastly adding some form of radar assistance to allow radar targeting but not like TA instead make it like mildly expensive fragile structure that has to be inside of the radar range itself and anything in its almost touching short range is given radar targeting.

That last one is a bit of a stretch, but i find it less macro and more "establishing a radar infrastructure". Multiple links where it can break if sniped.

What all this does, is make radar have a cost associated with its coverage in the form of building so many radars, rendering them as early warning initially, providing a counter and effective competitor to radar in the form of scout since it both provides vision and is itself radar invisible, and establishes really two systems for artillery targeting, either using forward position scouts instead of radar, or needing both radar and targetting.

I sympathize with nano, i liked having to use scouts for artillery in alpha. It isn't radical, it is just putting scouts foreward and units/backup scouts behind and artillery in the back. Ranks and files.