Okay so I've noticed that the topic of commander sniping using T1 bombers has been a pretty hot topic, not just lately, but for quite a while. I finally decided to go in-game with the dev mods enabled and actually examine some matchups of AA vs bomber spam. When you get to the beginning of late game is usually when these bomber snipes happen (from what I read on the forums and what I see in-game), so I tested a bunch of mid-late game situations. I tried a lot of different configurations of T1 AA towers, T2 flak towers. I also tried different compositions of T1 bombers and T1 AA fighters for the attacking side. Ultimately, I found that bomber snipes are extremely easy to counter. Example: late game situation. Here's 12 flak towers, costing 19200 metal, vs a bomber snipe consisting of around 250 bombers and 80 hummingbirds, which costs over 74000 metal. The flak towers were able to successfully protect the commander, so such a situation would leave you with a healthy 50k metal lead over your opponent. note: the walls are there just so the enemy AI doesn't walk away. 16+ flak towers can protect from over 500 bombers/hummingbirds, and once you get to around 20+ flak towers, you are basically invulnerable to any reasonable number under 1000 bombers. What about the early game?? I found that it took 20 bombers to solo kill a commander (no other units or structures to assist), and if a commander has 3-4 T1 turrets nearby, it takes between 26 and 28 (i tried various combinations of bombers assisted by hummingbirds as cannon fodder as well). This is actually a lot of bombers to be able to make in the early game, especially without being detected. Other findings: basically every configuration of 9 or so flak towers can withstand an attack from any reasonably sized snipe attempt. Essentially, to counter an opponent's bomber rush, it really doesn't take much effort. A couple of flak towers will do, and also note that in my tests, the only defense were the flak towers. It doesn't account for the fact that in most snipe attempts you a significant portion of your bombers as they try to make their way to the middle of the opponent's base. Yes I realize that the commander is sitting in the middle of the flak towers and in a real game that's unlikely, but these are just rough approximations. tl;dr You should actually be happy if your opponent goes for a bomber rush. It means, with minimal effort and scouting, you basically get a nuke or two worth of a metal lead for free.
Building AA is part of any fledgling bases' healthy breakfast. Just like anti nuke silos and walls in front of laser turrets, and umbrellas That goes for team games, FFAs and 1v1s (the ones that get past the dox stage, if any)
Everyone knows flak stops T1 snipes... that's what flak is for. I'm more interested in the T1 tower numbers! Your current data suggests that every 1 T1 turret increases the critical mass of bombers by 2. However, I'd like a clearer definition of that critical mass; Is it 20 bombers each making one attack to kill the commander in a single blow, or is there some looping around for a second pass involved? Does this 1-2 ratio get altered at higher numbers of bombers? MORE DATA! MOOORE!
Data! Well, the testing of T1 AA towers was actually pretty interesting. While the T2 AA towers exhibited a sort of exponential/polynomial growth, where small increases in flak towers increased their total efficacy by a huge amount, the T1 AA towers actually exhibited the opposite effect. How I tested: For small numbers of AA towers I tried to be relatively neat. But once I hit around 45-55 AA towers, I just decided to stack them (unrealistic, but we're more interested in general trends so it's okay) Lots of towers: impossible to stack realistically. I used a bomber snipe squad with a 3:2 bomber:fighter ratio. So for example, 30 bombers and 20 fighters (Ratios like this are what generally are used in successful bomber snipes, since fighters are good at absorbing incoming missiles) Data: On the left we have towers, and on the right we have the number of bombers (not including fighters). i.e. 45 bombers means there were 30 hummingbirds as well. Towers- --edit: the data table got messed up-- Towers -- Bombers 9 21 16 30 25 36 36 50 45 52 50 55 60 65 70 70 80 75 100 80 250 115 As you can see, small numbers of bombers can be dealt with by T1 towers, but we get immense diminishing returns. Around T = 60-80, we start to see the effectiveness of the hummingbirds in absorbing incoming missiles become too overwhelming for the towers to handle. Evidently, the more towers you have, the less effective they are at clearing airspace, which makes sense because T1 towers fire homing missiles which can sometimes fail to hit their target on the first run, meaning a handful of fighters can waste dozens of missiles. (All the more reason to build T2 flak!) But this does explain why some people may find bomber snipes to be too hard to counter. They can build crazy amounts of T1 towers and hummingbirds (which also fire homing missile that can miss), expecting that a large amount of T1 AA defense is comparable to a small amount of T2 air defense. So I guess what we can take away is that T1 exhibits diminishing returns, while T2 air defense exhibits increasing returns. Comparing results to the T2 flak data, we roughly find that 9 flak cannons (14400 metal), is about as effective as 250 T1 turrets (56000 metal). I estimate the critical point where flak cannons are better than T1 towers by metal cost to be around 3-4 flak towers. -- It's understandable that people get mad at bomber snipes, especially if they think they're actively protecting themselves by building massive amounts of T1, but, as the data shows, T1 is horribly ineffective at taking down swarms.
Lovely expansion of your research! Yet every answer begs more questions: for example, if T1 turrets suffer diminishing returns from overkill, would spreading the samr number of turrets over a wider area increase their effectiveness? Another dangerous question, one i ought to know; do Commanders die instantly if they're riding a Pelican and it's destroyed? If you have T2 for pelicans then you obviously have flak, but one wonders how safe one might be simply by lifting the comm a few feet off the bomb-able ground...
I think T1 turrets would be more effective if they were spaced out, rather than in a tight formation like you have there. The flaks work either way, but having the missiles so tightly packed, everything comes into range almost at the same time, so you have large masses of missiles going after the same targets. Space out the missile turrets in a spiral and I bet it'll be more effective. Not flak effective, but perhaps not a diminishing returns situation.
So there's basically no reason to complain about bomber snipes, because you only need to mass about 30-50 T1 missile turrets to protect your stationary commander while you tech to T2 flak cannons?
I didn't spread them out in my tests because doing so is hard to standardize. If you maintain even spacing of turrets, you get to a point where a portion of the turrets are no longer in range of the approaching bombers.
No, early game bomber snipes are very ineffective and can lose you the game. A basic calculation can show this. It takes 20 bombers to minimally kill an unprotected commander. Which is 4800 metal. Reaching T2 costs 5000 metal, so by the time your opponent can snipe you,even if you're completely undefended, you can have t2 up. No need to build t1 towers en mass. If your opponent goes for such a risky strategy, they'll probably lose.
Please go forth and try this tactic when testing the Flak If the attack is broad enough, you need far more than just 9 Flak to achieve full 360° coverage (or at least full 180° required to deal with the outcome of above pattern). And if you want to go completely mad: Send in some fighter from the opposite side first to force turn the Flak turrets and with 1 second delay send in the main force. Requires some delicate timing, but can reduce effectiveness of non-tracking defenses by far.
