Build, gather, brawl, repeat: The history of real-time strategy games (2017) | Hacker News Reader

Build, gather, brawl, repeat: The history of real-time strategy games (2017)

arstechnica.com

84 pointsTomte61 comments

The new AOE never hooked me. I desperately wanted it to. Feels like there is no RTS salvation possible anymore.

I don’t know what it is about the older titles, but I can still play supcom (as FAF now) all day and enjoy the hell out of it.

I also played the shit out of BFME2 until the servers got taken down.

Sometimes I get an itch for StarCraft 2, but it’s a hard game to play competently. Requires a certain… sweatiness.

Re: StarCraft 2, I think it's a shame that playing at lower speeds never really caught on. Both SC1 and SC2 have (or at least had?) options like "normal", faster, and fastest speed, but the community normalized on the highest speed as the standard, where mechanical skill is much more important to winning, and you don't get much time to plan anything.

You plan ahead of time. Of course, no plan survives contact with the opponent and then you improvise. There is, as you correctly observed, no time for conscious thinking then. It's all 'muscle memory' so to speak. Hence you won't be competitive until you have played thousands of games.

The match-making algorithm ensures that you win ~50% of the times, unless you're at either extreme of the skill level. Every so often (or quite often) you're matched with a vastly superior opponent and blasted to pieces, but it's already much better than SC1 w/o such match-making, where beginners never saw the light. Also Blizzard has a good handle on cheating in SC2 (unlike SC1).

The early game on normal speed is insanely boring but mid/late game on fastest can be overwhelming

I think it's a bit of a shame Archon mode (2v2 but 2 players control a single "player") never took off and eventually got cancelled

Archon mode still exists though, for example Harstem (SC2 pro/streamer) posted a video of a 3vs3 with some of the best players in the world just 2 days ago:

https://www.youtube.com/watch?v=eThKoRNB6Wc

You're right that the archon ladder is a shadow of the 1v1 ladder though.

Optimal play doesn't really involve much novel planning unless you're an absolute god at the game. Until then you're likely to see more improvement just improving your execution... which sucks

That is how most sports are. There's not all that much strategy involved in amateur league tennis or football either, execution rules supreme. I don't think you can separate out execution from strategy unless you explicitly separate out the time pressure, such as in (non-bullet/blitz) chess. But RTS without the RT part is a very different game.

Yeah, it's not that it's unique or anything. But the S is supposed to be the value prop of the RTS genre. Most people are fine with the focus being on execution for sports and FPS and what not.

Well compare and contrast with turn-based strategy games like Civilisation - I don't think the S is the value proposition per se. I think the combination of RT + S is what drives people to the genre. There are plenty of excellent turn-based strategy games where time pressure is effectively removed, and only the planning component is important.

I'd recommend turn-based strategy for anyone who is not particularly interested in the execution part.

Well, that's for the amateur sports, but I can tell you that if we're talking about professional sports where players are way more skilled, the strategy there plays a huge role. I didn't understand that before I started keeping track of the sports with http://www.televisionet.tv/ , but at the same time, it's pretty obvious that execution rules supreme anyway, there's no doubt

Same here. Age of Empires II is still the cash cow. Neither AoE III or Age of Myth hooked me. AoE II is the best RTS in my opinion. Timing, game play, balance, fun with others.

I even - GOG - reinstalled Warcraft 2 lately and enjoyed its simplicity. ;)

AOE2 is wonky, but that's what makes it good.

Villagers collide with each other, forcing you to find new tree-lines (compared to Starcraft/Starcraft2 SCVs/workers which "stack" on each other for maximum mineral collecting at all times).

Units in AOE2 move terribly, minimizing what you can do with micro (sure, micro is still possible, but this ain't no "magic box Psi Storm", nor a 11-muta harass here). AOE2 "micro" is largely delegated to patrol-shennanigans and "zvzvzvz" (staggered formation / flank-formation) for dodging magonel shots.

-------

SC2 had units deathball too easily with perfect flocking behavior. The 90s-style crappy flocking behavior / collisions (aka: idiotic Dragoons from SC:Brood War or AOE2 pathing issues) somehow balances the game for the better.

Once units "deathball" too easily, you end up with well... deathball-style combat. By making units get less-efficient the bigger the group (ie: Dragoons going the wrong direction or getting stuck in paths, or AOE2 units running around not understanding how to attack the damn enemy cause the group is too big), it forces the players into controling smaller groups and more strategically using them.

Which seems like more fun? Its annoying because the "units don't do what you want them to do" (and SC2 units really "do what you want them to do"). But... somehow its less balanced with the SC2 approach.

-------

The key to a good strategy game is to accept and recognize the duality of Lanchester's Linear Law (ie: Risk rules), and Lanchester's Quadratic Law (ie: Axis and Allies).

If you overly focus on the Quadratic law, you end up with death-balls.

If you overly focus on the linear law, it feels like additional units don't support each other and the game feels very flat (ie: Risk-like).

But if you have a rich continuum between the two laws, you end up with great gameplay (Starcraft: Brood War, AOE2, and Hearts of Iron 3).

Let units get quadratically stronger as they flock together... but only to a point. (IE; the size of the front in Hearts of Iron 3. Or until the "Magic Box size" in Starcraft:Brood War). Once the units grow beyond that, they need to be hampered and grow more linearly, encouraging a more diverse set of combat situations.

nice writing.

I think you have a good point with the "death-ball" approach. I haven't seen it put it that way.

I'm a big SC:BW fan. I haven't played in a while but I love watching it, and I feel like the competitive RTS market is something that is waiting for a hero. SC2 really isn't it, for a multitude of reasons. (balled groups of units being one of them, imo)

Frost Giant's first technical demo (presented in "the pylon show" forget which episode) was a "perfect flocking simulator" showing how the machine could multi-threaded handle 100 million units around a complex maze. I hated it.

Give me the chaos of units bumping into each other and trying to find the optimal way with other moving agents in the way. Maybe not quite as bad as Dragoons and SCVs, but similar.

There's the certain aspect of suspense and excitement that BW pro games have compared to SC2. "we all know that the unit has to move there and shoot, but will it get there in time? have they given it orders fast enough? is something going to block it?" -- it really ads to the atmosphere of excitement, knowing that the perfect play isn't going to be possible.

> competitive RTS market is something that is waiting for a hero

Really try Age of Empires 2. Its having a mild renaissance right now.

https://www.youtube.com/watch?v=2u3HupyXyKQ

Sure, its over 20 years old now, but its still getting new DLC, updates, and balance patches today.

I'm 8 minutes in and the presentation level is astounding. And with the red-bull sponsorship I'm sure this has got a bit of the cash behind it to keep such a great presentation.

Thanks for the link, I will definitely keep an eye out for this. I'll be watching the rest of this video through the day I think.

I think the biggest difference from Starcraft vs AoE2 is that AoE2 has random maps.

So while a map archetype has preset elements (ex: Arabia always has 8 sheep, 2 boars, 6 berries, 4-woodlines, 3 gold groups and 2 stone groups near your start), you don't actually know where they are.

Both players also start in a somewhat random position (players know the general area to scout), but the game is one of discovery. You need to scout the map, understand and explore, and then decide where to place strategic castles (or other defensive buildings), and where / how you'll fight.

-------

Beyond that, AoE2 has a lot of bonus damage, especially compared to Starcraft. A Halberdier normally 6 damage, but will deal +32 bonus damage vs Cavalry (for a total of 38).

Unit matchups are therefore more complex than Starcraft (sure, there's bonuses like Banelings vs light armor, or Immortals vs armored enemies), but nothing quite as gross as the bonus damage you have to memorize. (Skirmishers beat Archers. Scouts beat monks. Halberdier beats cavalry. Swordmen beat trash)

I really think that the randomised maps definitely add a component of strategy and depth to AoE2 I definitely don't think that unit matchups are more complex than in Starcraft, there are simply more interactions and possible matchups than starcraft because of the larger number of different units.

Starcraft's heavy emphasis on execution and APM means that very subtle changes to units can dramatically change matchups. Environmental positioning, micro management of units, healing, flanking etc. All change engagement outcomes. AoEs damage bonuses are easier to understand and requires less practise and expertise to squeeze every last point of damage out.

I'd love to see a version of Starcraft that has randomised maps instead of the fixed rotation that we currently have. It'd make playing at lower levels more fun since smart thinking would play a heavier role than rote memorization of build orders and times. And at high levels of play I think we'd see more exciting clutch moments as the best players with the best execution improvise their way through fights.

Unfortunately it would mean that the tight balance between all the different units in Starcraft would necessarily be relaxed and made more resilient to small changes. Right now, whenever new map rotations are introduced, some maps are considered unbalanced because small variations in widths of bridges or placements of obstacles favour one race over others.

I know some of those words.

AoM was pretty solid in my opinion

Dark Reign was a sleek, fun RTS that never got the attention is should have. It's balance was awful - but you can solve that with a gentleman's agreement. Total Annihilation was still played as of ~8 years ago.

https://www.beyondallreason.info/

Keep an eye out for Stormgate: https://youtu.be/0UqxgaWuYss

A handful of the original devs for Starcraft 2 got together and created Frost Giant Studios. I'm hoping it jumpstarts the RTS genre again

I still feel like 0AD (https://play0ad.com) moves at just the right speed and isn't as chaotic as everything else. With the source open, soundtrack/art/etc, it's really a more relaxing play, especially when you find the right opening play/map/options.

In SC II, you can switch to the grid layout. That helps with the keyboard side of things.

FAF is good. A shame the gigantic maps are impossible to really enjoy though, there's a sweet spot with supcom between too big and too small where it works really well, but outside of that it falls apart. I find the late game incredibly boring because of how slow it can be, but the mid-game to the first cheaper experimentals is super fun.

Also, balancing air. It seems to be impossible to make it fun or balanced at the final tier.

Did you play AoE2:DE? It is the classic, but with quality of life improvements and regular balance updates.

FAF is great but a journey in development: - Supreme Commander - Forged Alliance extra - Forever community and the fantastic extra MAPS, Units, extension of capabilities.

That is a LOT of development time. Other features I like:

- It will scale up as computing power gets better.

- If one player has a computer struggling to keep up, it doesn't desync or drop them, it slows down. Sometimes to an unplayable point but one nuke can remove enough units to make it better again.

Ah, Battle for Middle Earth 2 (BFME2)! Now that brings back fond memories (and countless wasted hours defending Helms Deep / Minus Tirith from 2 other players!). Perhaps not an iconic example of an RTS, but certainly a fun one for any LoTR fans

There's nothing literally close to supcom(faf). I reckon faf, and maybe cnc3/ra3 will be the only viable rts games for casual players in the next 5 years.

Beyond All Reason gets pretty close

+1 for FAF, really the only game I've been playing for years. And most of that has been camped out on Seton's Rock playing the same build order and yet it is still deeply satisfying watching a 2 hour game unfold

I'd be remiss not to link to https://www.beyondallreason.info/

A community made spiritual successor to the total annihilation/supreme commander series. It's notionally in alpha, but it works perfectly, and I have to say it's pretty amazing (hesitate to call it prefect). Yes, it's complicated, but the balance, options, automation and strategies available are almost perfect, as is the presence of strategic zoom so you don't feel like you're viewing the Mona Lisa through a toilet roll and forcing you to battle more with the interface than the game itself (I'm looking at you starcraft).

on the article front, I didn't see references to Sacrifice: a bizarre hybrid mix between third-person action and RTS. It didn't leave a legacy (as far as I can tell), but by God (pun intended) it was a weird and wonderful game.

How does this relate to TA spring? I'd spring the engine and this game is built on that?

i don't know the specifics, but it's had a long road and evolved out of prijects that came before.

I believe it's currently running on a modified version of the spring engine.

My friends and I still do comp-stomp company of heroes. That game had such sophistication of cover, unit movement, morale, and resource capture without "drones". I love it immensely.

Very excited for the third installment.

Also, Homeworld 3 should be at least a re-skin of Homeworld 2 + a few new mechanics. I'd encourage RTS fans to give HW2 a revisit, and maybe deserts of kharak as well, for savory lore at least.

These "non standard" RTSs are much more enjoyable than the more copy-paste base builders, at least to me.

Try Original war. Some pretty nonstandard ideas - battery powered vehicles, limited number of people assigned to roles,...

I used to love RTS games as a kid. I kind of fell off of them when they started to really emphasize the e-sports angle. Mostly because I wasn't really good enough to enjoy them and there seemed to be an ever increasing focus on "more APM". However, I recently picked up "Starship Troopers: Terran Command" and loved it. It was built to just be a campaign focused single player RTS. It's such a fun game, and doesn't overstay its welcome.

The lack of realistic single player RTS games really makes me sad. I want to escape like a kid playing with army men, not compete in high speed competitive chess.

All the single player games are fantasy or sci-fi, or otherwise unrealistic. Just give me a game - of any genre - based on reality and I will be pleased.

you missed some gems.

try company if heroes. world in conflict. the wargame series. even total war games have enough complexity to feel realistic.

I loved playing red alert and command and conquer online when I was younger. It seems like the genre has fallen off.

I recently discovered They Are Billions which I found to be a fun variation of traditional RTS. The goal is to manage resources to build up defenses and defend against large zombie attacks. Single-player only, but I've played it for over 100 hours!

https://store.steampowered.com/app/644930/They_Are_Billions/

They fell off because they weren't single player anymore. RTS games feel very strategic in singleplayer. You're often working to overcome a seemingly more powerful asymmetric force. In multiplayer you're just grinding through a hyper optimized meta or you're losing.

I've felt this for years but never been able to express it so well.

I think the article is a bit too dismissive of Real-Time Tactics; they are definitely a spin-off genre, and some like the Total War series have significant strategy elements - you have to grow your armies, not just lead them.

I also like the Combined Arms RT from Eugen games - the Wargame and Steel Division series are great and Wargame: Red Dragon has a very enthusiastic multiplayer community. The games are very competitive and yet the APM is quite low. Here you can see top ranked player zoom out and examine the game map at a leisurely pace compared to Stacraft players: https://youtu.be/HBynkUYmH8E?t=2164

Their game in early access - WARNO has various smart orders to automate some common actions.

What sets Combined Arms RTT from others is the focus on Combined arms and realism. They have a deck-building element where you choose the units available to you to match your strategy of control of the map (see why I think the article was too dismissive?). You have all kinds of interesting tradeoffs. In example you might find that cheap but highly available tanks are great for fire support, even if they can't face the best of enemy armour.

This genre potentially will have a really good year, with three competing titles coming out - WARNO, Regiments (single player) and Broken Arrow.

I much prefer RTT games to play competitively, be it the Wargame series or, my personal favorite of all time, World In Conflict. Minus the base building and production minding, you can actually focus on the combat (and still have a platoon of tanks wiped off the map in a flurry of ATGMs when you glance away for a second) and map control, positioning, etc.

I really only play RTS games solo, or co-op comp stomps.

In terms of revolution I keep hoping for more developments with RTS+FPS merging.

Like Natural Selection. A great blend of tech tree research and FPS fighting. The RTS bit was a little limited. I hoped for bigger maps, more RTS depth, better bots. But the Natural Selection team never quite got there.

The Battlefield series doesn't do a great job but does it - kinda limited though.

I think the only way it could work is with fairly severe information asymmetry. The RTS player would need to have a lot more information than the FPS players. And I don't think you could do both.

This but on a smaller scale was one of the things I really loved about the original Ghost Recon.

I recently got back into the C&C franchise and realized how much I'd missed that part of my childhood. It was pretty cool to realize there's still an active community around these games -- check out Sybert on YT, he's doing a brilliant job casting matches across all of the still actively played C&Cs.

Admittedly I've never been good at RTS games in general, and maybe never will be, but it's still fun to take a break from 'single focus' games like FPSs once in a while and switch to the (literal) bird's-eye view.

This article doesn't feel like it was written by anyone paying attention to the scene for the last decade. It downplays the role MOBAs have played in the downfall of the RTS by calling them "subgenre offshoots that snuck off" while ignoring that MOBA player and viewer numbers dwarf what RTSs ever achieved [1]. They've completely eaten the mindshare that RTS games used to occupy. Unsure if this article is spinning history or clueless, but MOBAs killed the RTS.

There are still good entries getting added to the RTS genre. This article completely ignores They Are Billions, Northgard, and others...maybe because they aren't culture defining the way Starcraft and Age of Empires were. However, if that's their criteria for a good RTS game, then those days are long over.

1. 124 million monthly active players https://activeplayer.io/league-of-legends/

SC2 also killed itself by focusing solely on ladder and esports and abandoning the casual gamer. The genre might have had a chance otherwise.

The referenced article is five years old, so of course it's going to miss some of the biggest trends of the last decade. They Are Billions and Northgard were both released on Steam sometime after this article was written.

The author also made it a point to narrowly define an RTS game as "one that involves base building and/or management, resource gathering, unit production, and semi-autonomous combat, all conducted in real time (rather than being turn-based), for the purpose of gaining/maintaining control over strategic points on a map (such as the resources and command centers)," which leaves MOBAs out of the discussion.

I think the author succeeded in his goal of providing a light history of the RTS genre during its peak years (early 90s to mid 2010s). The criticisms you have may be valid, but they are largely outside of the article's scope.

I want a browser RTS:

- Simple. Few units. The reason starcraft 2 has so many units is that every expansion pack had to include new ones. Does anyone think go or risk need more units?

- Fast. Maybe a few minutes per game.

>it had never worked well with a controller. And nobody wanted to take a chance at figuring it out.

Perhaps they'll see a revival with VR? Some sort of eye level control to select units seems plausible

Children play Age of Empire (I and II, but never III (for very long)).

Teens play Rome Total War (I, off course, we do not talk about II).

I don't like RTS games, even though there are some really fun games out there like the Red Dragon series. The end game of RTS games is muscle memory clickfest which leaves little space for immersion. A pause feature is nice but, not frequently seen in multi-player games.

Well, I'd never say something like immersion is the goal of rts games.

I wonder if you could make a good RTS with a "command limit", or a delay between issuing a command and the army moving. This could be simulated by a horseback messenger, or the delay of lightspeed communications across large stellar distances. There's probably a reason nobody's made a game like that before though...

I can't name one, but I do have a feeling I played game(s) like this at some point.

Good point, this is another way to slow the gameplay down a bit. The following come to mind, Command Ops 2, Armored Brigade, Graviteam Tactics. Non are multi-player.

I've recently read some deliberations on that here:

https://acoup.blog/category/collections/total-generalship/

Many pre-modern armies couldn't really be commanded around the battlefield much or at all, and indeed if they could (to limited extent, eg the Roman armies) then there'd be severe limits on command propagation (horseback rider reaching a centurion who then needs to get some trumpets sounded, but there's only so many trumpet sounds you can make and train troops for), plus the general (and everyone else) have an extremely limited view of what's going on to start with. The author states that it wouldn't be a very enjoyable game if all such dynamics would be simulated.

But I myself would like to see some studio take a stab at something like you suggest, and see what comes out!

Yea, I felt the same way after playing a few of them years ago. RTS games are twitch games that pretend to be strategy games. If I want actual building/moving/fighting strategy games, I stick to turn-based, where I can actually think about my strategy and have the time to manage all the complexity. If I want a muscle-memory twitch game, I can play a FPS or fighting game. RTS occupies that weird area in between where it does neither well, but eventually the twitchy players always dominate because few if any of those games figured out a way to not give advantage to high "Actions Per Minute" playing.

Reply on news.ycombinator.com