The Case for Removing Radar

Taken into account. Just removing radar targeting while leaving radar as information to the user, as was present in TA, leads to micro of manually targeting radar blips.

One solution is to just make all units automatically attack all radar blips. This system was used in SupCom and various other games.

Removing radar entirely also solves that problem. And in my opinion does it in a way that leads to much better gameplay, involving scouting and strategic positioning of units and structures for intel purposes instead of just blanketing the map in radar.

If by backward you mean implemented in a game prior to the more recent SupCom, then technically yes it is a removal of a feature added later. TA didn't have automatic radar targeting; the Radar Targeting Facility excepted, units required line of sight to fire. And I think everyone agrees the Radar Targeting Facility was a dreadful idea. It ties a UI feature to an in-game asset that costs resources and can be destroyed, which is just icky. And regardless, the Targeting Facility was so expensive to construct and maintain it was not present most of the time.

By saying it is 'a step backward' you aren't really making an argument for why it is bad for gameplay to remove radar, however. I for one think the long-range radar and automatic targeting was one of the biggest weaknesses in the SupCom gameplay.

I think that disallowing artillery to target via radar is the best solution. Currently, I find that radar is very important; with it you can respond to threats without knowing what the threat actually is. If you throw levelers into an army of ants, unless I scout, I can only guess what the army is made up of by it's speed.

One of the problems with disallowing radar targeting is that it makes the range of artillery useless for defense. If it was disallowed, then there should be an alternative way to be able to use artillery defensively. I know there has been a great deal of discussion about giving radar special properties to make it more than just consecutive upgrades, perhaps that could be one of the properties, the ability to target with artillery? It always bothered me in TA that you have to build the automatic radar targeting facility-- it forced me to invest in something I didn't feel i needed because I just used my artillery at enemy bases.

Also, think about the amount of issues you'd get from trying to constantly keep LOS on enemy bases, constantly scout your borders. If the base just had a few laser turrets, you could never do any real damage with artillery, tactical missiles, nukes or laser satellites. It would give cause to turtling even more, and that just reduces the quality of game play.

The game was meant to be played on a macro scale, with the ability to micro to make units more effective. Radar helps to keep this game macro. You react to seeing an enemy blob by sending an army to counter, usually more than needed in case there is T2, and even then you can be thoroughly surprised to find that it was an army of levelers. Radar is a (in my opinion, obviously) necessary part of the game, and in warfare in general, however, how the artillery interacts with it should be re-balanced considerably.

There is no great reason to disallow pelters from radar targeting; their range and damage is too small for anything but creeps (defeatable by radar, ironically) and early game defense. T2 can easily destroy pelters if they are left unguarded and you use your force properly. Holkins guns are a little bit different: They are so powerful that you have to usually make a specific effort to destroy them, which is annoying, even considering that the planet size with make them slightly less useful for destroying bases. Holkins guns have a good reason to be denied radar targeting because even large T2 blobs can't reach them.

Catapults, being tactical missile launchers should require their own building for radar targeting. It should pull energy according to how many catapults it has to manage on the planet, for example: if you have 40+, it would cost 10 T2 power generators. This would help to force people to build them to be useful, rather than over powered spam. The ability of missiles is that they are accurate and long range, constantly scouting to use doesn't make sense-- we don't do that with AEGIS ships, if you know the location of something, you can hit it with a missile, but you have to have dedicated data centers to target.

Please realize that all of these are possibilities that I believe are solutions. Some of these are mixed and matched from other posts, including the original, try not to rip it apart, thanks!

I have made this exact case in the past for Supreme Commander. Ledarsi you do not stand alone. This would be more realistic; however, this is a PA and in PA Advanced Radar satellites see entire planets. Not only that, but such functionality of satellites and radar is necessary for the game to progress to an end in an orderly fashion. If someone is winning late game and someone else is planet hopping to avoid death, we need to be able to check a whole planet at a glance. If that's the case, we need to be able to build up to such intelligence gathering at least some what gradually.

If radar doesn't see land, you cannot put stealth tanks into the game and stealth tanks are cool.

If radar doesn't see land, Holkins and Catapults won't get used.

Also, because the planets are round, sneak attacks are at an all time premium. I don't think making sneak attacks even easier is necessarily the best course of action at this time. Other than better economy, Level 2 radar is one of the fine benefits of reaching Tech 2.

The pitch game is not intended to be played for four hours plus like originally envisioned by Nuetrino. Maybe a mod of PA that is more intrinsically based on TA/SupCom (doesn't have an orbital layer) will be. In such a mod, I think Radar on air is a great idea. It's really hard to say.

I think this idea could have worked so well for Supreme Commander, because between the gunships and the tech 3, air was the force for ending games. In PA air is a commander or asset snipe tool that isn't necessarily that good compared to your late game alternatives.

Also, I really liked the super computer in TA. It just felt awesome when the AI pitched in for more kick-@#$.

Bastilean, you are correct that certain functionality is necessary in order to end games that have gone to the "hide and seek" stage. But assuming that radar is necessary goes too far. What you need is some way to search a large area quickly, or have permanent eyes on a large area.

You don't necessarily need radar; vision will actually do the same job. A group of spy satellites could provide actual vision over an entire planet. Multiple smaller, cheaper units would also be much better than having a single immensely expensive satellite that just reveals the whole planet by itself.

I am confident that Uber will eventually get orbital to the point where it is interesting, but the real focus of the gameplay is always on the surface of the planet(s). Just making them extremely expensive and powerful is actually not the way to do that; that just means once you have a ton of economy you just start using orbital instead of surface units, especially since most anti-orbital assets are also expensive, and dedicated. I think I have a pretty good set of ideas for PA on this front, but that's a different thread.


Stealth tanks for their own sake are a gimmick. Without land radar, essentially every tank is a 'stealth' tank for radar purposes. And you could still have stealth tanks; by making them cloak against vision except when they get too close to an enemy unit.

Furthermore, if radar still works against air units, longer range air radar detection would still be a significant benefit of advanced radar. For a suitably higher price, of course.

I still believe this post is the one of the best ideas to come along in terms of intelligence gathering. Rather than removing radar, it becomes a question of how much information do you have on a particular target, and the farther away / more things blocking your view to the target, the less you know about it.

The problem with going pure LOS with a definitive cut off line is that it creates a perfectly binary setup that requires more fiddly work on the user's part - it's much harder to cover all areas of a circle bit by bit. There ends up being no 'smart' scout setup, you simply patrol scouts around your perimenter. Usually performed in TA by a air factory pumping out peepers.

So, I agree that Radar as a concept is dated and kind of gimmicky, but I would maintain that exterminans' ideas on having a radar 'footprint' is much more flexible and interesting.

I this disallowing radar targeting is the worst solution. It doesn't change anything because the manual attack ground command exists (if it didn't exist that would be another matter entirely). Making radar dots untargetable does not change the optimal behaviour of units, it simply hides the optimal behaviour behind a really bad UI. Mircomanagement to shoot at radar will be a requirement and the best case for me would be the creation of a widget to do this tedious work for me. When this happens we will be back where we started, not targeting radar dots is a degenerate mechanic.

The ideas in this thread are interesting. I'm fairly happy with what radar does in Zero-K because accurate artillery is significantly improved by scouting out the exact location of structures. Inaccurate artillery would want to force fire at the ground if you merely suspect there is an army present so wobbly radar takes away some of this micromanagement cost.

Anyway, as for ideas how about directional radar? People seem to be want early warning and completely accurate radar basically works as LOS so this would be a bit of a compromise. Directional radar would basically be a 1-dimentional heatmap which gives you information along the lines of "There are some units mostly East-ish of this radar tower". More radar towers covering a single point would give a bit of a better estimate of army position but nothing accurate enough to fire at and expect to hit.

I like directional radar because you can spoof really long ranged radar towers. Sit some units far away from your opponent such that he becomes used to their presence. Then you can attack along that line without causing alarm.

I agree that the radar is too convenient. They clearly have too much vision. But perhaps removing it is too big for a change...
I would consider a compromise:

• Radar only update every x seconds ( y for advanced radar )
• Radar must be fragile and costly to be a target of choice
• Radar are always visible within enemy radar range.
• ( optional ) units will target fire only when the radar is updated.

I am not fundamentally opposed to the idea of detecting blips. But the TA paradigm of blip detection that doesn't do much, available for cheap, is quite different from the kind of very comprehensive radar-targeting-capable options offered in SupCom and (currently) in PA. And it still has that issue of just manually firing on any location where your detection tells you there is an enemy.

Giving certain units short-range land radar could certainly work. Even if it gives accurate information for targeting- these units would just have good sensors. These short-range detection systems might even be motion-sensors only, or use a variety of other implementations proposed in this thread. But if such systems allow targeting, I will emphasize that their range must be kept quite short.

Actionable radar targeting is basically just like direct LOS for aiming purposes, and should be designed with that effect in mind. It would be ridiculous to design such a cheap unit which gave as much actual vision as a radar gives radar range. But for most intents and purposes, radar detection is just as good, since you are primarily interested in the ability to shoot at the enemy. Hiding the unit's type is an absolutely miniscule price to pay for such inexpensive long-range target acquisition.

Even a compromise solution that gives radar even limited long-range targeting capability should be heavily scrutinized. Fragility of the structure is not important because of its range. And cost is only a limitation if is rather extreme for what you are purchasing. A high-cost radar, which allows targeting at excellent range is actually a very powerful and efficient asset. Its cost would have to be truly exorbitant to actually discourage anyone from building it, and in general whenever the best solution to a unit or mechanic is just to marginalize it through massively exorbitant costs, there is a more fundamental problem that needs to be addressed.

I also find more complex long-range detection compromises like radar giving you a direction are rather interesting. Perhaps some system could be devised that would work well. However such compromise solutions will tend to run into the same problems that you do with targetable long-range solutions, and with non-targetable long-range solutions.

Either your long-range detection is useful for long-range targeting, although perhaps made less effective by any limitations it has. Or it is not useful for long-range targeting, in which case you can use the information given to you by the detection to issue manual fire orders.

At the end of the day. I really think that extreme long-range detection as a mechanic is just not a good idea.

Btw. Units targeting radar automatically isn't necessarily a ui feature. I mean you could manually click fire but unless it's stationary you are going to miss. If there was a smaller shorter unit needed to target confirm, then your units could be allowed to lead shots.

Also, the point is, the radar needs a nerf. Either ranges, quality of information, cost to build, or mechanics added to counter or supplement radar information.

Right now, the argument is right. Removing radar requires people to use scouts and using scouts is more hands on than one structure doing everything constantly. Right now the radar is a replacement for playing the game. It is the equivalent to building handfuls of scouts and positioning them everywhere and never losing them.

The reason i also think scouts should be radar resistant, is that radars can detect them but they can't detect radar, and radar is strong enough for that to work the other way around.

Long range detection is actually a necessary mechanic. At some range, the existence of any unit must be detectable, for bigger threats even at long range to allow for early scouting followed by countermeasures. Even more since the planets have a giant surface without fixed choke points and you NEED a hint where to scout. This is not true for scouts or stealth attacks, they actually must have a chance to close in unnoticed.

The issue is identifying the unit, where both SupCom and now PA made a wrong conclusion:
The precise location of any unit is far to valuable to give away just on detection. A consequence of this is also that detection the presence of an enemy army is not the same as distinguishing the individual units.

What is important to know though, is how big the threat in a certain region possibly is.

It is not sufficient to just decrease the update interval of the radar, as it would still provide perfect targets for any immobile unit and would enforce "preventive idle kiting" for any valuable target.

Neither is it sufficient to make distant radar targets just "wobble", as it is easy to eliminate the wobble by observing the moving blip of an stationary unit over time (even possible in form of a clientside mod). In addition, even with the wobble, formations and such are still given away on far too long range since individual units are distinguished at a far too high distance.

Actually a good point, although only a limited sight on a generic issue. Some units propose such an threat that it must be possible to identify them at a larger range. Even at such a range where other units would go by completely unnoticed.

NOTA gameplay what I think PA gameplay should try to emulate the most.

Positioning, terrain features and largescale manoeuvres are very important in NOTA and the cheap and long range radars are a vital part of allowing that type of gameplay.

Without you don't get the same kind of dynamics between the players as any push into enemy territory is more a blind guess rather than a calculated effort. You don't know the position of the enemy forces and you might just run straight into the bulk of the enemy army.
When you know the position of the enemy army you can take your fast units and skirt around the slower units and force them to split up.
Also when you hold an important hill in NOTA, the large radar cover allows you to go down from the hill to intercept enemy forces. Without radar you couldn't see incoming forces and you would be forced to spread your forces thin and you wouldn't be able to utilize the advantage of the hill because you cannot track enemy movements and the enemy can just crop up forces beneath the hill and bum-rush it without you being able to respond.
Now if you intend to have the same type of gameplay with only shorter ranged scout units you will put alot of strain on the player to keep the scouts alive.
If the difference between a dead scouts and a fully healthy scout is only a few seconds of movement between detection of enemy units and entering the range of said units then players will be forced to split their attention heavily if they want to keep track of enemy movements or stalk enemy units.
This might be acceptable if the scout have large vision radius and automatically stay out of range of enemy units like in R.U.S.E. for example.
However requiring the player to perform tedious scout micromanagement across several different planets to gain intelligence is something that I think is unacceptable.

In NOTA there is no radar wobble so everything accurate will hit their targets even if it is out of line sight. That said most long range weapons are inaccurate so radar wobble wouldn't really matter for them.

Like others have said and like done in NOTA you can have stealth units, radar-jammers and other types of radar-manipulating devices. Terrain blocks radar in NOTA like in other Spring games which is also notable. More on that here if you are interested.

This is actually the case in NOTA. I like it. Radars are really fragile in NOTA so you can kill them easily with air units.
It also give you a good indication of how good radar coverage the enemy have and if you strike the enemy radar before you perform an assault it gives you a few seconds where the enemy is effectively blinded before they can build up new radars.
Since terrain blocks radar, radar placement is also sensitive and when the enemy is up on ridge he wants to get some radar coverage up on the far side of that ridge, it require a radar on that ridge or some flanking radar which gives you ample of opportunity to snipe that radar.

LoS for radar isn't an option for PA - it's just not possible with the complex planets geometry. It works perfectly well (that means it is fast enough) if the map is a simple plane with a heightmap so the terrain is actually just 2D, but PA uses a much more complex, real 3D planet geometry.

Besides: The exaggerated curvature of the PA planets would kill the range of the radar completely.

There's a shipload of technical restrictions in place which make it almost impossible to simply copy the radar system of another game. The current implementation was choosen because it's a very simple one technical wise, even though it's not very appealing gameplay wise.

LoS for radar is viable. It is only showing it that is the problem.

Just place the radar on a hill or mountain. Problem solved.

Nope, problem not solved. Even if you would place the radar on top of a mountain, it's range would be only little more than the current T1 radar. It's really that bad...
Try doing the math for yourself, just calculate how many radar stations you would need for full coverage if you placed every single radar station on top of a mountain on a size 4 planet.

Actually, even the LoS checks themselves are non-trivial - considering the number of checks necessary and the complexity of the planets geometry. Possibly every radar equiped unit on the planet would need to be tested against each enemy unit. And each single test would be performed on the complex, concave/convex geometry of the planet. That's far more complex and expensive than just doing LoS in a world which only consists of primitives - or even better: Just a simple heightmap. Latter one only requires a number of trivial samples in an image.

LoS is hardly acceptable for nearby units as not too much geometry needs to be tested. But long range? Forget about it.

I wouldn't have thought it would be too hard to have map features produce something like a shadow. A small area behind the mountain that can't be seen by the radar (unless you put a second radar down on the other side).

I would also like to point out that radar can and does bend around the planet. Have a read about OTH radars.

Well if you are utilizing the full range of your radar it seems good enough. Also why not launch your radar into space?

Err... how high is a mountain? How big is a size 4 planet? Is the radar range in the same units as the planet radius?

There are more units with weapons than radars. Catapults and Holkins have about the same range as advanced radars. Click the arrow in the top of the quote window to go to the thread and see the post in its' whole context.

Can we please leave radar IN the game.
If you think they are over powered, then up the constant energy drain - that will make scouts cheaper and more energy efficient.

I prefer things with LESS micro when possible.

I prefer things that have the most strategically stimulating gameplay. Less micro does not automatically mean better. Gameplay is King as they say.

Automatic and precise radar targeting over very large distances is not conducive to the best gameplay (at least not right out of the box for all radar). Total Annihilation proves this.

The only radar I think is broken is advanced orbital. "Line of sight to basically the entire planet" is way too much power. It takes all of the mystery out.

As it stands, even with T2 radar being so good, I still have to scout and make sure you aren't building nukes. T2 radar doesn't tell me if you are. It's actually fairly unusual to get radar coverage into an enemy base (unless all you play are the tiniest of maps, I guess) so mostly what it reveals are efforts at making a forward firing base.

T2 radar also doesn't tell me if the blob of approaching units are Ants or Levelers, and there's a pretty big difference.

T2 radar doesn't even tell me if my automatically firing Catapult is hitting walls, the enemy Commander or the counter-Catapult he's rushing to build to destroy my Catapult with.

Basically I don't know where you guys are coming from in thinking that T2 radar eliminates scouting. It allows automatic blind firing at unknown blips. It doesn't do any prioritizing or tell you what those blips are. If you rely on radar to guide your defenses, I'm pretty much going to beat you every time. Radar keeps you from being defenseless but it doesn't make your defense intelligent.