Alright, so there's a thing I noticed in a game last night that kind of bugs me about radar and I think it always has. Maybe this is a chance to fix it. How does the radar know what color the signatures it is getting are? Without observing them visually it doesn't even know what unit it is, or if it's just a building. How does it know what team the results belong to? What I'm basically proposing is that all radar blips are some neutral non-team-color until observed visually.
Maybe each faction or color gives off a unique signature. IE: like submarines have a unique sound signature.
Except all the factions are using identical hardware. Even when a (real life) radar site identifies a plane, it doesn't know who owns it, just what model of plane it is.
Perhaps the radar towers are less like conventional systems that we think of where you throw out some radio waves to see what bounces back, and more like really sensitive antennas listening for each faction's IFF signal? I mean, I've yet to see a unit that can tell when it's being painted by radar already, so maybe the radar towers are just receivers.
It's probably a quantum radar or something like that. That's always the best excuse. TA had colored radar signatures and SupCom didn't. I liked the way SupCom did it more, but the reason they're colored is probably because that's the way TA did it.
I have another complain about radar and scouting, seeing any building should leave "ghost building", telling us what is there. And if we scout, and radar is on, icon should change from non specific to specific.
The changing from non-specific to specific is a good one. Once I know what that blip is I should always know what that blip is unless I lose the radar paint on it. I guess I just wish that the colors would remain hidden unless you got a visual on that blip at some point. It would make teamplay a little more strategic as you wouldn't know that yellow had gone over to blue's base to build up or assist with some building until you actually scouted it and saw (after which the radar would keep track of the ones it had seen).
You mean just the way it worked in SupCom? I can't agree unconditionally with that. While it would be an improvement over the current state, it would be still rather limiting since you still had the very same issues as with the SupCom version. Let me just give you a list of issues with the SupCom version: No differentiation between unit types whatsoever, radar and vision worked just the same for the biggest experimental building and the smallest scout. Well, differentiation was "possible" by stealth and cloak, but they only ever worked in absolutes too, enforcing the strange concept of an omni-radar. Radar was absolutely mandatory. It's impossible to play without radar, and if thats only to make use of the weapon range of your defensive structures. You don't really "gain" intelligence on an enemy. Observing him for a long time gets you NO benefits over sending an scout on an suicide mission, either you have his base fully scouted or not at all. You can't really prevent intelligence on your base. You can neither prevent the long range radar, nor the single scout which is all which is necessary to get perfect information on your whole base. It would be necessary to use an intelligence system, which at least took the size of units in consideration when determining visibility, and NOT just a static vision radius. But after intense consideration, I would even go a step further, take the time-component into consideration too. You basically have an "observation score" for every player on every unit which tells you, how long and intense the unit has been observed. Having intelligence (that can be direct vision, distant vision or just long range radar) on the unit increases the score up to a certain ceiling, at the same time the score also degrades at a fixed rate to ensure that units vanish again. The rate at which the score degrades and the thresholds for visibility can vary per unit to make a certain unit more or less "visible". Certain units can only contribute to the score up to a certain threshold in return for having long range, e.g. a radar will not be able to make the unit visible, but will only contribute until you have the location locked down. This would lead to a more slow paced intelligence gathering, where long term observation actually has a use. Also has using multiple scouts instead of just one, to increase the rate at which intelligence is gathered. Using a single plane to scout the enemy base would only give you a rough idea where the enemy base is (you can locate, maybe even identify the largest buildings), but it won't uncover the base for you yet, units inside the base and smaller structures remain hidden unless you invest more time. If your scout is blown up while gathering intelligence, you are stuck with what you already have. Radar also just gives the location of large structures right away, it takes much longer until small structures are tracked at all. Some may even be undetectable with radar on long range at all (since score degrades faster than radar can build it up), but radar WILL help to keep them uncovered for a bit longer once they have been spotted by other terms (because it slows down the degradation). I wouldn't worry so much about visualization, you only need to visualize hard borders if there are any because hard borders are usually counterintuitive. Why do you need it right now? Because the fixed vision radius causes counterintuitive blind spots in plain sight. Same for the radar, it only acts so weird because it has a counterintuitive hard range limit and you run into the same problem as with vision, blind spots in completely unexpected locations.
This is always debatable. What was true about radar is that it provided a HUGE amount of information for an absurdly low price. It's little surprise that radar became very important as a result. You can make anything mandatory by making it superior to everything else.
YOU WILL BE NECESSARY This is true. I liked the way TA Spring handled radar, radar icons had a slight "wander" to them, that made weapons that relied on radar slightly less effective. For those that played TA, the Targeting Center is arbitrarily cheap in TA Spring, costing about as much as a T1 Bot Fac. It didn't matter to terribly when you had a Buzzsaw or two firing at enemy bases, but it allowed for some mind-play with artillery wars.
Modern radar operates in passive and active modes. Active is a burst of radiation, which reflect off of an object and returns to detect range and bearing information Passive intercepts detects enemy EM radiations and matches the signature to known platforms to identify what is being detected. Additionally any units would be equiped with an IFF code. (identify friend or Foe) Therefore an Ally would be identified and everyone else should be shot at. Those three things would be combined for the commander to give him a total picture of the battlespace. I.E. they colors are totally cool
I think the most important thing to note about modern radar, is that it isn't just radar. It's a computer generated, 3 dimensional map of objects which is capable of combining many data sources into a single model. It also causes the mentioned effect that entities without IFF (which by the way ONLY works for friendly units, a code needs to be encoded in the original radar pulse in order for the target to answer with a corresponding radio signal upon detection) and smaller entities take quite a while to show up in the model. They start as simple noise in the model and it takes several iterations before they can be distinguished from the background noise. It takes even more time (and data!) before the radar station has collected sufficient informations, to create a 3D model of the target, allowing real time tracking of not only targets rough position, but also almost providing vision... Same goes for robotic vision, it actually works quite similar, except for the range. Computers are exceptionally bad at analyzing static pictures. They work a lot better if either them self or the target is moving, since they have to rely on some type of feature detection and tracking. Observation over time can also enhance the quality of the model since it help to eliminate noise caused by environmental effects like distortion, fog, dust, vibrations etc. Feature detection is also a lot easier for nearby targets since the increased resolution allows for the detection and tracking of more features per object. And again, only detection of friendly units is (almost) for free and again it only works because you know what you are supposed to be looking for. Identifying a target by feature detection is magnitudes harder, than to find and identify some type of visual tag. That is if you know how the tag is designed and encoded, which is only true for your own and allied units. (Your own units might use some type of checkboard code like QR, while enemy units might use a certain combination of spectral lines, it is impossible to search for "anomalies" unless you already know what type of marker you are looking for.) But again, it's not only about technology, it's also about gameplay. Like I mentioned before, both the TA and SupCom implementations of "intelligence" were far from perfect due to their perfect information concept. Enforcing artificial noise in the output of the model is pointless, while it may improve the immersion, it does not help at all bypassing the problems. Noise can be eliminated by simply calculating an average over time, no matter how fancy the noise is. If you are not supposed to have high quality information, you may not be given information at all! This requirement already enforces that the amount of information degrades with distance and visibility since you are not supposed to ever get high quality information over long distance for free. Anything short of completely hiding untracked units from the client already means giving PERFECT information already since permanent tracking from radar combined with one-time identification by a scout has the same result as having a permanent lock. This also results in a second requirement: The quality and therefor amount of information must vary over time. To prevent permanent perfect information, targets which the user is not supposed to have a permanent vision on, need to be hidden completely a short time after the active data gathering has ceased. It is not sufficient to just remove them from "vision" as radar also allows for full tracking of an identified unit which means that you keep getting very detailed information on the target, in an scenario where you have an lock on the commander, this would be fatal. This means that targets NEED to be completely purged from radar unless you are supposed to be able to track them. But it also works the other way around, the amount and quality of information does not only need to decrease over time, it also need to build up over time. It must not be possible to gather perfect (although only temporary!) information on the whole gamefield without significant effort. That means that a single scout may not be able to provide perfect information at first glance, improving the quality of intelligence requires an investment of resources, whereby time is a valid resource in the scenario. There is also a third, although optional prerequisite: The amount of resources invested into intelligence should scale with the amount of intelligence gathered. This requirement is automatically fulfilled if you make your scouts swam out. However it is not true yet for the scenario where you choose to focus on a single goal. The only logical solution to this problem was, to scale the amount and therefor quality of information with the number of observers. When combining this with the first and second prerequisite, this leads to a scenario where multiple scouts are capable of building up a high quality model faster than a single scout. There are two possible implementations now, both would be valid. Either only the speed of gathering intelligence can be increased, or the quality of informations can also scale. First one would mean only to get the best possible quality faster, but you would not be able to surpass the quality you could have achieved with a single source given infinite time. This is the implementation I recommend for huge range data sources, like radar, to prevent scenarios where players get perfect information on a large area by stockpiling radars with little risk. The second one means, that you become able to track and identify targets over a longer range by combining multiple sources, although range also reduces the speed at which information is gathered again. This is the implementation I suggest for mobile scouts, to encourage the investment of additional resources in active intelligence. Non-scout and non-radar type units would work quite well with either system, although I would advise to go with the first option or at least a limited hybrid, since the second option could cause performance issues due to the complexity class of the required algorithm ( O(n²) ). It also rewards the use of scouts as part of a regular strike force which is a good thing in my opinion because this means more diversity.
