Basic Vs Advanced Radar

Oooo You need to tell me how to do that exterminans. Pretty please?

I must have internalised that post, forgotten about it and then used it as a jumping off point for an idea later. Glad to see that technical limitations wouldn't really be a problem for the idea to come to fruition.

You basically have two data structures, one is quad tree which also stores the number of units, grouped by signature size, per node (recursive sum), and the second one is a set of radar cells, each has a certain threshold.

Using the radar cells to sample the quadtree is quite simple then. First filter out units which are big enough to show up as individual entities (remember, they are grouped, so you can tell at first glance). Afterwards just use the quad tree to test the sum of signatures against the threshold.

Most expensive part? Moving units, because when a unit moves, the quad tree has to be updated, but that's not THAT problematic either. It's just O(x log y) (x units, map of size y) for the update, and it's O((x log y) + z) (x sensor cells, map of size y, z units) for generating the output.

Each cell can be transmitted as a whole, as a geometric shape and a "heat" value.



But to be honest:
Personally, I would rather see an "awareness" based system than the heatmap. While the heatmap offers a better result for a static situation, it lacks the temporal component and active stealth gameplay.

Would a negative size value for a unit act as a 'dampener', basically a radar-jammer?
I assume that it would as you sum the value of all the units in a cell... I'm completely getting this from a mathematical standpoint, but I'm no programmer so I'm grasping the Math that you're using... but not the programming application as it were.

Edit: maybe I should be frequenting the Modding section more often, eh?

You could do that with the awareness system, where a stealth system could lower the awareness with each tick while enemy radar increases the awareness. So your stealth system could even be eventually outnumbered.

It's a bit difficult with the heatmap. You would need to treat friendly units in the range of the radar jammer different and it would make unit movement more expensive. You would need to alternate the affected unit itself when moving into / out of the stealth field. But then it would work.

I took that into account.

I decided that there are counter considerations:

1. It takes a longer time to get T2 than it does to travel between all build sites.
2. Travel time is accounted for in the production of a T2 fabber (to lay the initial foundation, and then contribute no more to the build)
   
3. If the T2 fabber is left to its own devices it will take 60 seconds to complete the build. 6 fabbers can each travel to their allocated radar position from the position of the t2 radar, and construct it, in that time. (the distance is 360, they move at 12)
   
4. It took me less than 5 seconds to figure out how to go from poorly placed to more efficiently placed, less than 10 seconds to implement it, and I'm not practised. And more efficiently placed still isn't optimally placed.
5. You can build the network over a period of time
6. You can build the network as part of your base defences anyway.

Of course, you do pay for it with player micro. But my comment was aimed at the "T1 radar isn't efficient enough". Which is correct, if you don't know how to place it.

And the thread is about making T1 radar obsolete. But it's much faster to put up than t2 radar (takes the same group of fabbers half the time), it costs half as much, so it isn't as much of a loss if you place it in an advanced position later game.

I see a place for it, and I'm happy with how it's currently in game.

You just made one big missassumption:
At some point you must upgrade to T2, because you need the additional coverage OUTSIDE your base. Even if you place the T1 radar at the outmost end of your base, it will never give you the same amount of intelligence as a T2 radar in the very same location.

And the moment you are building the T2 radar (thats inevitable if the game advances), it completely invalidates the existing T1 structures. All you can do, is demolishing them because they are now a waste of energy.

So yes, the T1 radar fulfills a role. But no more than the T1 units in SupCom did. While most cost efficient, they became an invalid choice if the game went past the T1 phase, and the existing units (except for the engineers) were just trash metal / canon fodder.

Look at the image I posted please. If you're going to place a T2 radar on the outskirts of your base, you may as well build a field of T1 radar outposts that extends out of your base and is cheaper to replace parts of. You can gain more coverage for the same amount of mass. If there's an incoming ant attack the T1 field has 24 seconds more warning than the lone T2 radar. If there's a breach in the field, the T2 radar complements the field and fills the gap until you replace the T1. If you lose the T2, you still have extensive coverage from the T1. And if you never replace the T1 field, the enemy has an extra 24 seconds to prepare each attack that you have given them for free.

If you want to play that game, that's on you.

Hi there,

My apologies if this question has been asked already, but I can't think of a better thread to ask.

At the current time, anytime I lose line of sight of enemy buildings, and I have no radar coverage, there are no markings left there, and it makes it quite hard sometimes to locate the enemy base under certain circumstances, such as night time, or when there are loads of terrain objects around.

Is it possible to leave the building icons in place for the strategic view, for buildings that were spotted already? Of course there would be no updates to them until there is radar coverage or line of sight established.

Are there any plans for this ATM?

Thanks!

I see, sounds solid then. I may have to reconsider my radar build strategy.

While the idea of a heat map radar is intriguing, it's still missing the most important thing: player feedback. The ability to be detected is both based on how "noisy" your army is, and how "strong" the opponent's radar layer is. Therefore, there is really no way to know if you have been detected or how much more leeway you do or don't have. Even then it's all on some kind of curve, and curve extrapolation is not something you should be expecting of players in live play.

I can get aboard stacking radar slowly having a more accurate effect, either heat mapping or triangulating.

That, or any other idea, is acceptable.

Just plain bigger is not acceptable.

People mentioned earlier about being able to shoot radar dots from certain things. What if radar worked off radar? Isn't that how it works in real life/ace combat games? You have large radar, then you have a branch radar. The large radar detects targets, the small radar sharpens accuracy so your units can engage it out of LOS. I wouldn't even mind if, if the t1 radar was in t2 radar range or if the unit was in both their range, and the t1 radar can in fact see them, then it acts as line of site instead of radar... and then making radar bleeps not shootable and line of sight shootable only.

I would very much enjoy seeing Exterminans' unified signature system implemented, especially if active and passive sensors systems were considered (A fairly straightforward and inexpensive addition - just increase the signature and resolution of a unit when it turns on its active sensors) to further differentiate stealth units.

Given that system existed, T2 RADAR could be the passive and thus much harder to detect version, or vice versa.