Why did Real Time Strategy genre die out?

[Damned Registrations]

Yeah, if meta strategies were truly the most effective, they'd never change, barring changes to game mechanics. The fact entire metas get thrown out because a single player pokes a hole in them and everyone starts copying him, or playing to counter him instead of what was previously dominant, reveals that chasing the meta is basically a copout when it comes to strategy- you've given up on thinking about the problem and just rely on superior execution and luck to see you through.

Something I've realized from playing a bunch of chess lately is that what constitutes a 'good' strategy in it is rather nebulous. Often while checking back for mistakes, what it reveals as a huge advantage at some point was unrealistic- if you play perfectly for over a dozen moves you'll certainly win, but if you play even the second best move even once during those dozen moves, you'll lose to even the 5th best move the opponent can play. In chess, these situations can be very interesting; things like an early queen sacrifice that eventually trades back all the lost material, or leaving your king terribly exposed but applying constant pressure so it can't be exploited. You get a lot of very tense situations one after another.

This doesn't work in other games.

This generally takes the form of a 'glass cannon' strategy in other games, that snowballs unless you miss a shot and get counter attacked. But it's boring because there's no complex situation to unravel, it's just a matter of twitchy execution or really obvious choices in a turn based game. And you see this play out in RTS all the time, and it's encouraged- static defenses are intentionally made to be utter crap so the best strategy is to build nothing but army and economy and be very aggressive with it. I think that's a major flaw in design- you can see in the evolution of fighting games that they used to play out that way (and some still do, with easy 'Touch of Death' combos with no counters) but many now often have comeback mechanics, ways to interrupt combos, multiple characters in a single round, and generally a lot more interaction beyond 'well he hit me, guess I'll go make a sandwich while I die.' I think RTS needs stuff like that- a one time use spell that kills everything on your half of the map. A hero, army, technology or just pile of resources that arrive when you start losing. Rebellions or ambushes that get triggered when you're winning. Ways to avoid the snowball effect and diminish how optimal a glass cannon strategy is, which at the same time would make different factions more unique and interesting. I think chasing the esports market hasn't helped- developers seem eager to make matches last in the 5-30 minute range reliably and have single exciting moments where everything goes to shit and the whole game is decided, rather than a measured back and forth of hard to quantify advantages.

Nobody wants to play a fighting game where every character dies to a single punch, so why do developers think that's how RTS should work?

 

[civac2]

What games are the ones with bad defensive structures? Dawn of War maybe?

AOE2 has walls as defensive structures which are used a ton. Not sure about defensive towers. In Broodwar all races use their defensive structures a lot. They are very strong and cost effective if the opponent decides to engage them.

 

[Camel]

And you see this play out in RTS all the time, and it's encouraged- static defenses are intentionally made to be utter crap so the best strategy is to build nothing but army and economy and be very aggressive with it.

Such as? SC bunkers, photon cannons and zerg static defense are pretty good and are often used offensively in so-called bunker rushes. You seem to like turtling and effective turtling strategies lead to boring matches and generally to boring RTS games.

 

[tritosine2k]

You seem to like turtling and effective turtling strategies lead to boring matches and generally to boring RTS games.

To me the best Men of War match ever (by far- out of hundreds) was for instance turtling with center of map forming a divisive line and reinforce.

It ticked all the boxes & never occured again. CTF in reinstallments didn't exactly help.

 

[Gahbreeil]

Company of Heroes 1 gave me Red Alert 2 vibes. The series still exists, there was a sequel.

Can't get more RTS than Red Alert 2.

 

[Damned Registrations]

SC bunkers, photon cannons and zerg static defense are pretty good

What games are the ones with bad defensive structures?

WTF are the two of you smoking? Cannon/tower rushing cheese doesn't work because static defenses are strong. They work because they let you reinforce your position instantly instead of waiting for a dude to walk across the map, putting you behind the curve. Defensively, they're pretty much exclusively used for detection or because you're sitting on thousands of minerals at the supply cap. Nobody worth a shit builds a bunch of cannons do defend an expansion, let alone your main base because you can just outrange them or move past them. They're wildly overcosted for what they do. Who the fuck is building bunkers instead of siege tanks?

Walls get used in AoE2... almost not at all. Wooden walls get thrown up mostly as an early warning system or to prevent low effort harassment, they present no real obstacle to an actual attack. People build houses in place of walls precisely because walls are so shit. Castles get used to control territory sometimes but only assuming your opponent is behind you in tech, and pretty much never defensively.

Effective defenses aren't necessarily about turtling, but about being able to establish territory and map control independantly of your death ball of ranged units. Yes, if static defenses are too strong there's no incentive to attack and it becomes a stalemate. But that's easily handled by making siege weapons available without also making the siege weapons insanely good at killing units.

Against an opponent of roughly equal skill, I can't recall ever playing an RTS where static defenses were a good choice. They're crap in all the Blizzard games, AoE2 and 4, Total War, and Company of Heroes.

Games where static defenses are good would be... Master of Orion (Assuming you've invested in missile and shield tech), EU4 (More for forcing engagements and territory control), AI Wars, and some of the Civilization series. Of those, only AI Wars is an RTS, and it's an oddball that wasn't built for player vs player. In the case of all but MoO, they're more about controlling territory and influencing where enemy attacks can arrive from than actually stacking them up to become impenetrable while you get stronger.

I think another problem with static defenses is that while they can serve as a deterrent in a FFA situation as is commonly seen in TBS, nobody plays that shit in RTS, so there's no deterrence factor. There's no compelling reason to attack a turtle in a 1v1 RTS match when you can just landgrab instead. In a FFA, static defenses might convince a pair of enemies to fight each other rather than you.

 

[Jaedar]

reveals that chasing the meta is basically a copout when it comes to strategy- you've given up on thinking about the problem and just rely on superior execution and luck to see you through.

The odds that you can beat the meta with mere strategy is pretty optimistic. Superior execution may let you execute strategies, and the strategies that are optimized for high level execution skills might not be optimal in the lower tiers, but you're exceedingly unlikely to beat the meta by thinking about it for 30 minutes: if it were that easy, someone else would already have done it. Unless the game is played by very few people, and they are not very dedicated.

WTF are the two of you smoking? Cannon/tower rushing cheese doesn't work because static defenses are strong. They work because they let you reinforce your position instantly instead of waiting for a dude to walk across the map, putting you behind the curve. Defensively, they're pretty much exclusively used for detection or because you're sitting on thousands of minerals at the supply cap. Nobody worth a shit builds a bunch of cannons do defend an expansion, let alone your main base because you can just outrange them or move past them. They're wildly overcosted for what they do. Who the fuck is building bunkers instead of siege tanks?

Bunkers are frequently used in professional star craft though? Same with sunken colonies and photon cannons. And not just for rushing, frequently for defending your bases. Protoss like walling off with a photon cannon+zealot for example. Since bunkers cost pure minerals, they don't really compete with siege tank production... Not to mention massive lines of missile turrets to fend of mutas.

Against an opponent of roughly equal skill, I can't recall ever playing an RTS where static defenses were a good choice. They're crap in all the Blizzard games, AoE2 and 4, Total War, and Company of Heroes.

Let me speak then of the RTS I know best: Dawn of War Soulstorm. While it is true that mines and turrets are mostly memes (necron turrets being the exception), the whole listetning point system is basically static defense (even the listening post without a gun is still defending the point with it's chunky hp and building armor). Admittedly the fact that they also generate income muddles the point since they are not pure static defense, but even if they didn't you'd still build at least the basic post very frequently.

 

[Average Manatee]

WTF are the two of you smoking? Cannon/tower rushing cheese doesn't work because static defenses are strong.

I must have missed the advanced micro that allows you to kill a cannon with a dragoon then.

 

[Damned Registrations]

Superior execution may let you execute strategies, and the strategies that are optimized for high level execution skills might not be optimal in the lower tiers, but you're exceedingly unlikely to beat the meta by thinking about it for 30 minutes: if it were that easy, someone else would already have done it.

Completely true, trying to start a new meta on your own is a pretty hopeless endevour. Doesn't change the fact that it means you're effectively memorizing answers to the test instead of figuring them out yourself.

Protoss like walling off with a photon cannon+zealot for example.

I don't think a single cannon/bunker really qualifies as 'effective static defenses.' Realistically the wall is doing all the work there, and a single cannon is just the fastest way to get a range weapon defending it and abusing the high ground mechanics of broodwar to prevent cheeses.

Maybe I'm just out of touch for BW, but they're certainly not seen in SC2 and I don't remember them seeing any real use years ago when I watched BW being played. Certainly doesn't help that they gave zerg insanely fast access to siege alongside terrans, and protoss the means to teleport past them in multiple ways.

I must have missed the advanced micro that allows you to kill a cannon with a dragoon then.

Right click past the cannon. You now have a damaged dragoon attacking a line of probes, (or the pylon powering the cannon )while the opponent has... a cannon defending nothing. If you built 3 cannons in one spot, you can kill a dragoon but now have 2 dragoons past the cannons. If you spread the 3 cannons out, 3 dragoons kill one cannon easily. This is the reason static defenses need to not just have a slight edge against infantry, but a massive one, or they're a waste of resources. I don't need to kill your cannon. If I can kill literally anything with my army or expand without fear of yours, it's a win for me.

 

[Jaedar]

Completely true, trying to start a new meta on your own is a pretty hopeless endevour. Doesn't change the fact that it means you're effectively memorizing answers to the test instead of figuring them out yourself.

If you actually want to be good you kinda have to understand why though, rarely is the execution so rigid you don't need to adjust it on the fly to respond to the enemy. There's no real shame in looking up the current best and studying that. If you enjoy rederiving all of maths and science, be my guest, but you're probably not going to be a competitive engineer. Figuring out a game isn't as complex as that, but even so.

I don't think a single cannon/bunker really qualifies as 'effective static defenses.' Realistically the wall is doing all the work there, and a single cannon is just the fastest way to get a range weapon defending it and abusing the high ground mechanics of broodwar to prevent cheeses.

If the photon cannon couldn't shoot, zerglings would just kill the buildings. The wall is doing a lot of work though, you're not wrong about that.

 

[Average Manatee]

Right click past the cannon. You now have a damaged dragoon attacking a line of probes, (or the pylon powering the cannon )while the opponent has... a cannon defending nothing. If you built 3 cannons in one spot, you can kill a dragoon but now have 2 dragoons past the cannons. If you spread the 3 cannons out, 3 dragoons kill one cannon easily. This is the reason static defenses need to not just have a slight edge against infantry, but a massive one, or they're a waste of resources. I don't need to kill your cannon. If I can kill literally anything with my army or expand without fear of yours, it's a win for me.

That doesn't make the dragoon stronger than the cannon. That makes it more mobile.

Also a bunch of probes will easily kill a dragoon, assuming the cannon wasn't in range of the mineral line and that the probes don't bodyblock the dragoon from escaping the cannon.

 

[vota DC]

SC bunkers, photon cannons and zerg static defense are pretty good

What games are the ones with bad defensive structures?

WTF are the two of you smoking? Cannon/tower rushing cheese doesn't work because static defenses are strong. They work because they let you reinforce your position instantly instead of waiting for a dude to walk across the map, putting you behind the curve. Defensively, they're pretty much exclusively used for detection or because you're sitting on thousands of minerals at the supply cap. Nobody worth a shit builds a bunch of cannons do defend an expansion, let alone your main base because you can just outrange them or move past them. They're wildly overcosted for what they do. Who the fuck is building bunkers instead of siege tanks?

Walls get used in AoE2... almost not at all. Wooden walls get thrown up mostly as an early warning system or to prevent low effort harassment, they present no real obstacle to an actual attack. People build houses in place of walls precisely because walls are so shit. Castles get used to control territory sometimes but only assuming your opponent is behind you in tech, and pretty much never defensively.

Effective defenses aren't necessarily about turtling, but about being able to establish territory and map control independantly of your death ball of ranged units. Yes, if static defenses are too strong there's no incentive to attack and it becomes a stalemate. But that's easily handled by making siege weapons available without also making the siege weapons insanely good at killing units.

Against an opponent of roughly equal skill, I can't recall ever playing an RTS where static defenses were a good choice. They're crap in all the Blizzard games, AoE2 and 4, Total War, and Company of Heroes.

Games where static defenses are good would be... Master of Orion (Assuming you've invested in missile and shield tech), EU4 (More for forcing engagements and territory control), AI Wars, and some of the Civilization series. Of those, only AI Wars is an RTS, and it's an oddball that wasn't built for player vs player. In the case of all but MoO, they're more about controlling territory and influencing where enemy attacks can arrive from than actually stacking them up to become impenetrable while you get stronger.

I think another problem with static defenses is that while they can serve as a deterrent in a FFA situation as is commonly seen in TBS, nobody plays that shit in RTS, so there's no deterrence factor. There's no compelling reason to attack a turtle in a 1v1 RTS match when you can just landgrab instead. In a FFA, static defenses might convince a pair of enemies to fight each other rather than you.

Wooden palisades are worse than houses. Stone walls are better, the problem is that stone is rare.
Stronghold have good defenses.

 

[civac2]

Here is a recent Broodwar ladder game between a pro (Hero, playing Zerg) and some highly ranked ladder player (Protoss).



Protoss opens with a forge into second nexus, making 2 cannons to defend his fast expand. He later adds 2 more cannons to the defenses. (These fast expansion builds are weak to hydralisk all ins and Protoss won't have psionic storm in time. Cannons are the only thing which can hold the hydralisk attack. You sometimes see 7, 8 or even more cannons in that scenario. In the above game, though, Zerg does not opt for a (serious) fast hydralisk attack. Note that the very approach of a Protoss fast expand vs Zerg is only possible because cannons are very strong.) When Zerg adds a group of mutalisks to his army (for sniping high templars and harrassing) Protoss also makes a couple cannons near the main nexus. Protoss routinely defends additional expansions with 4 cannons each, seen around 13:30 (12 o'clock expansion) and 15:27 (mineral only expansion).
There is also a point in the game where Protoss gains the upper hand in army balance by picking off the mutalisk group with a dark archon (around 13:00). Zerg immediatly starts hardening his positions with some sunkens. I spotted 7 of them in total.

Obviously, this depends on matchups and flow of the game. But like in this game Protoss tends to fortify expands with 2-4 cannons vs Zerg. Terran builds a ton of missile turrets vs Zerg and also vs Protoss if Protoss goes for mass speed shuttles or arbiters. Zerg also uses sunken colonies liberally against both terran and protoss. They are particularly nasty when interspersed with lurkers. A wall of sunkens is also one of the few things Zerg has that can stop a marine/medic ball in the early game. This has huge implications on how the whole matchup plays out, similar to the example of cannons enabling Protoss fast expands vs Zerg.

 

[tritosine2k]

The thing about photon cannon is normal dmg into small target like hydra and marine vs. dragoon thats half so not interchangeable early.
Also sunken is half.

 

[Damned Registrations]

I think another issue with static defenses is that they generally get balanced for campaign use. And obviously, if they're going to be a balanced but not outright overpowered option in a map where you never need to leave your base for the first 20 minutes, they're going to be utter dogshit in one where you do.

There's no real shame in looking up the current best and studying that

Yeah, despite my tone there, I'm not trying to imply it's shameful. I'm just bemoaning the lack of meaningful decisions in the earlygame of RTS. MOBAs suffer the same issue, especially if you're forced into playing a meta that often isn't relevant because you're a casual scrub and not the pros your teammates pretend to be when raging over who picked what. Many other competitive games have much more interesting opening phases, including chess if you're not a GM.

Also a bunch of probes will easily kill a dragoon, assuming the cannon wasn't in range of the mineral line and that the probes don't bodyblock the dragoon from escaping the cannon.

Assuming you've got a full line of probes, and pull them all. But how many minerals did they fail to mine while chasing a dragoon kiting them around? Did you lose even a single probe? That's going to be way more costly than the dragoon over the course of 3 minutes, let alone 10. My probes were just happily mining away, because your cannon wasn't doing anything to them. Now I've got enough money for another dragoon, and you don't have the money for your own dragoon. Hell, I might have enough for 2 dragoons before you can even afford 1 more dragoon or cannon yourself.

Arguing static defenses are good because they kill an equal value of infantry in a toe to toe fight is meaningless. They need to be worth the opportunity cost. And the math just doesn't work out for the vast majority of the game.

Here is a recent Broodwar ladder game between a pro (Hero playing Zerg) and some highranked ladder player (Protoss).

So I'm at minute 7 so far, and protoss has made a total of 4 cannons against seemingly a very cheesy zerg rush that failed. Zerg has zero defenses. Map allows for a very easy wall at second expand which is new to me, if such maps are the norm these days I can see static defenses being a lot more useful. But still, we're in a situation where even when he has no intention to expand, the player investing in static defenses has made 4 cannons and 5 zealots and 2 corsairs. Even in a completely turtling position, less than half his resources are invested in cannons as opposed to army, and zerg needs none. That doesn't strike me as a healthy balance. 12 minutes in, and zerg is completely dominating until he trades 11 mutalisks for a a high templar. Protoss needed a 2000 resource swing to get back in the game. Definitely a weird game, and I think the use of static defenses on both sides made it a lot more interesting. They're definitely more useful than they used to be on that map- the zerg expansion with the sunkens being on a hill like that was not a thing in the old days. Caster also described it as a 'unicorn' of a game, which makes me think it was cherry picked to show off so many cannons being used. I suspect I'd see a lot less if I clicked next on the playlist.

The thing about photon cannon is normal dmg into small target like hydra and marine vs. dragoon thats half so not interchangeable early.
Also sunken is half.

Did they change unit sizes and damage types since I played? Cannons used to be normal I think, while Dragoons and Hydras were explosive (half vs marines, 3/4 vs hydras) Sunkens were explosive though.

 

[civac2]

So obstinate. Of course, players try to use as little static defense as possible. But the trade off is often worth it because Broodwar defence structures are so strong. I even wrote down some more general info in my second parapgraph above. Protoss vs Zerg will almost always defend the natural expansion with some cannons, have some (usually 3) at each additional expand and have a single one at their stargate. You will find games where they make less but that is the most common case. Zerg has the most mobile army and often can get away without static defense. But they do use it sometimes to deny entire areas of the map as you picked up on. Against Terran Zerg does have to make sunkens in some situations. Terran builds tons of missile turrets. Look at any TvZ where Terran goes infantry (which is the most common approach).

 

[Jaedar]

Yeah, despite my tone there, I'm not trying to imply it's shameful. I'm just bemoaning the lack of meaningful decisions in the earlygame of RTS. MOBAs suffer the same issue, especially if you're forced into playing a meta that often isn't relevant because you're a casual scrub and not the pros your teammates pretend to be when raging over who picked what. Many other competitive games have much more interesting opening phases, including chess if you're not a GM.

Maybe? I can't speak to other RTS that well, but in dawn of war multiplayer I feel I am making meaningful decisions within the first minute in most (but not all) matchups.

 

[Borelli]

Static defenses in starcraft are generally strong early game (sunken / photon cannon in a right place can save a game) and bad late game (3/3 units just delete them).
One exception is guarding expansions but even then it's main purpose is that you force your opponent to attack with more units.

 

[Borelli]

There is no strong reason to believe the meta has found a global optimum in terms of strategies (the fact that even something as old as brood war still has new strategies, let alone something as ancient as go still evolving shows this), but I agree that every strategy in the meta is proven effective.

Last time i followed the scene was 15 years ago i think, when bisu invented fast corsair / dark templar vs Z and fantasy invented 1 port valkyrie into mech vs Z.
I thought that the games was "solved" since then. What new strats have popped up since then?

 

[Jaedar]

There is no strong reason to believe the meta has found a global optimum in terms of strategies (the fact that even something as old as brood war still has new strategies, let alone something as ancient as go still evolving shows this), but I agree that every strategy in the meta is proven effective.

Last time i followed the scene was 15 years ago i think, when bisu invented fast corsair / dark templar vs Z and fantasy invented 1 port valkyrie into mech vs Z.
I thought that the games was "solved" since then. What new strats have popped up since then?

Hmm, I don't recall ever seeing a valkyrie 15 years ago, if it's that old it's waaay less new than I thought.

 

[civac2]

Dunno how old some of the strats are. Protoss will often fast expand with a gateway opener vs Zerg which I don't think Bisu did back then. Are you familiar with Nexus first (PvT), mech switch (TvZ, going infantry at first), mass speed shuttle play (PvT), mass queen (ZvT, vs Terran metal)?

Valkyries were bugged in Broodwar. Remastered made them more usable.

 

[Saldrone]

Former Age of Empires developer says the RTS genre's stuck in a rut: 'you're still playing the same game'

Apparently the problem with new RTS games is "they’re still really largely based on that formula."

Dave Pottinger has programming credits on all three Age of Empires games, was director of technology on Age of Mythology, and was lead designer on Halo Wars. He probably knows what he's talking about when it comes to the RTS genre. Speaking to Videogamer to promote his new studio Last Keep's upcoming RTS, Project Citadel, Pottinger pointed out the genre hasn't done a lot of evolving in the last two decades.

"It hasn't changed much," he said. "You know, you're still playing the same game we're playing 20 years ago and looking at some of these new games—Stormgate and others like that—and they're still really largely based on that formula. It works, it's an old, golden set of rules, because they were good back then and they're still good now, and it’s nice to see that stuff still works but at the same time I want to do something new, we want to do something new."

As for why the RTS is so indebted to its past, he mentioned that back in his Age of Empires days there was a fear that players would push back against attempts to experiment. "There were some times on the Age franchise where we flew a little too close to the sun," he said. "We had to pull back and take some very innovative things out of the game—I'm talking particularly about formation-based combat in Age of Empires 3. Hell, we demoed that at E3, and took that out of the game because we were afraid it was going to alienate too many of the existing Age fans."

I'm the guy who fruitlessly votes for Dawn of War: Dark Crusade in the PC Gamer Top 100 every year, so I'd agree the RTS enjoyed its peak a good long while ago. Pottinger isn't just talking the talk, however, and Project Citadel (as it's currently codenamed) is aiming to shake things up. It'll be a space-based roguelike with procedural maps, and sounds a bit like a cross between Planetary Annihilation and FTL. "Battle your way through Voltari space on a mission to free the sector from their rule" says the description on its Steam page. "It won't be easy, but you have a secret weapon to aid you in your voyage—a secret relic that allows your mothership and crew to regenerate at the edge of Voltari space—again and again as you uncover the story’s twists and turns."

Other tweaks to the formula include a "Turn-based overmap" and "Fast-paced action with skill-shot special abilities". I dunno if permadeath and procedural generation is what it'll take to get me excited in a new RTS, but it's definitely a niche in need of a change. For once I agree with our cynic-in-residence Fraser Brown, who summed up recent attempts at revivals by writing that real-time strategy almost came back from the brink of death and then fell flat on its face.

Project Citadel is on its way to a release in early access, where the campaign will be fully playable. "Through Early Access," the developer says on its Steam page, "we plan to expand the number of playable factions, variety of units, and add related content in stages. We also plan to iterate on balance, gameplay rules and expand content based upon community feedback."

 

[Nutmeg]

Project Citadel (as it's currently codenamed) is aiming to shake things up. It'll be a space-based roguelike with procedural maps, and sounds a bit like a cross between Planetary Annihilation and FTL. "Battle your way through Voltari space on a mission to free the sector from their rule" says the description on its Steam page. "It won't be easy, but you have a secret weapon to aid you in your voyage—a secret relic that allows your mothership and crew to regenerate at the edge of Voltari space—again and again as you uncover the story’s twists and turns."

This sounds pretty lame.

 

[Borelli]

Dunno how old some of the strats are. Protoss will often fast expand with a gateway opener vs Zerg which I don't think Bisu did back then. Are you familiar with Nexus first (PvT), mech switch (TvZ, going infantry at first), mass speed shuttle play (PvT), mass queen (ZvT, vs Terran metal)?

Valkyries were bugged in Broodwar. Remastered made them more usable.

I remember 14 nexus (disgusting strat) and mech switch, protoss used shuttles for harass or contain busting i don't know if that is considered "mass shuttles", queens were not used often when i last played.

Valkyries were only bugged in 4v4 because of the sprite cap or something IIRC.

 

[civac2]

Yeah, it's about the sprite cap. I'm not entirely sure but my impression was that the sprite cap already reared its ugly head in 1v1 if you made many valkiyries.

Shuttles for harass and busting where used vs Terran since forever but Protoss these days often gets a bunch of them. Up to 5 or 6 or even more with shuttle speed upgrade, instead of more advanced tech like arbiters which was more common back in the day. Such refinements are what is common as strategic innovations these days and are still happening. I don't remember anything entirely new, no. There was a time when Terrans routinely used mech switches or just straight mech vs Zerg to counter defilers. But it turned out that queens with broodling is just too efficient against mech over time. So the meta is mostly back to SKTerran in TvZ. Flash is still experimenting with mech builds though.