original creator of generals.io here (late to the party)
really cool to see people still having fun with this game I made in college! I sold the game a few years back because I didn't have time to properly maintain it, and I'm glad the new owners have kept it running.
Linking to some past HN threads on this:
- https://news.ycombinator.com/item?id=13145781 original generals.io post
- https://news.ycombinator.com/item?id=13562866 launch of the Bot API
I used to play the 8-player version of this game a lot. Here's an analysis of the strategy. I put it in rot13 in case you prefer to discover for yourself.
Gurer ner 3 onfvp fgengrtvrf va gur rneyl tnzr:
1. Encvq nffnhyg. Jnvg sbe n qrprag-fvmrq nezl ba lbhe pncvgny, gura fgneg punetvat nebhaq gur znc naq ubcr gung lbh ner noyr gb pncgher nabgure cynlre'f pncvgny juvyr vg'f cbbeyl qrsraqrq.
2. Encvq tebjgu. Sbphf ba pncghevat greevgbel naq arhgeny pvgvrf. Ihyarenoyr gb encvq nffnhyg, fvapr pncghevat pvgvrf jvyy qrcyrgr lbhe sbeprf va gur fubeg grez naq yrnir lbhe pncvgny ihyarenoyr.
3. Qrsrafvir tebjgu. Fybjre naq fgrnyguvre pncgher bs greevgbel naq arhgeny pvgvrf, jurer lbh tebj pnhgvbhfyl naq sbphf ba fheivivat encvq nffnhygf. Vg graqf gb ybfr gb encvq tebjgu, ohg jva ntnvafg encvq nffnhyg.
Fb gur guerr fgengrtvrf unir n ebpx/cncre/fpvffbef eryngvbafuvc. Vs lbh abgvpr gung n cnegvphyne fgengrtl vf cbchyne va gur pheerag zrgntnzr, lbh pna nqwhfg lbhe bja fgengrtl nppbeqvatyl.
Va rvtug-cynlre, lbh xabj gurer jvyy or frira ybfref. Vs lbh jnag gb or gur bar jvaare, lbh unir gb trg yhpxl. Fb vg znxrf frafr gb yrna gbjneqf n uvtu-inevnapr fgengrtl yvxr encvq tebjgu be encvq nffnhyg.
Va gur zvq naq yngr tnzr, cynlref graq gb or birezngpurq ntnvafg rnpu bgure. Lbh'yy unir whfg 2 be 3 rzcverf erznvavat. Hfhnyyl bar rzcver jvyy or fvtavsvpnagyl fgebatre. Vs lbh ner gur jrnxre rzcver, lbhe orfg fubg gb jva vf trarenyyl gb sbez n znffvir nezl jvgu nyy bs lbhe sbeprf, punetr vagb lbhe bccbarag'f greevgbel, naq frr vs lbh pna gnxr gurve pncvgny qhevat n zbzrag jura vg'f cbbeyl qrsraqrq. Vs lbh ner gur fgebatre rzcver, lbh xabj lbh'er tbvat gb jva ol qrsnhyg qhr gb gur fabjonyy angher bs gur tnzr. Fb gur vzcbegnag guvat vf gb qrsraq ntnvafg gur fbeg bs fhecevfr nggnpx V whfg qrfpevorq. Gurer unir bayl orra n srj gvzrf jura V ybfg sebz n cbfvgvba bs fgeratgu, jura zl jrnxre bccbarag znantrq gb bhgcynl zr ba gur onfvp zrpunavpf bs greevgbel naq pvgl pncgher. Nernf jvgu n uvtu qrafvgl bs pvgvrf ner rfcrpvnyyl inyhnoyr gb gnxr pbageby bs, vs lbh pna znantr gb qb vg, juvpu graqf gb or qvssvphyg.
Deciphered if you don't feel like doing it and want to read it here:
There are 3 basic strategies in the early game:
1. Rapid assault. Wait for a decent-sized army on your capital, then start charging around the map and hope that you are able to capture another player's capital while it's poorly defended.
2. Rapid growth. Focus on capturing territory and neutral cities. Vulnerable to rapid assault, since capturing cities will deplete your forces in the short term and leave your capital vulnerable.
3. Defensive growth. Slower and stealthier capture of territory and neutral cities, where you grow cautiously and focus on surviving rapid assaults. It tends to lose to rapid growth, but win against rapid assault.
So the three strategies have a rock/paper/scissors relationship. If you notice that a particular strategy is popular in the current metagame, you can adjust your own strategy accordingly. In eight-player, you know there will be seven losers.
If you want to be the one winner, you have to get lucky. So it makes sense to lean towards a high-variance strategy like rapid growth or rapid assault.
In the mid and late game, players tend to be overmatched against each other. You'll have just 2 or 3 empires remaining. Usually one empire will be significantly stronger. If you are the weaker empire, your best shot to win is generally to form a massive army with all of your forces, charge into your opponent's territory, and see if you can take their capital during a moment when it's poorly defended. If you are the stronger empire, you know you're going to win by default due to the snowball nature of the game. So the important thing is to defend against the sort of surprise attack I just described. There have only been a few times when I lost from a position of strength, when my weaker opponent managed to outplay me on the basic mechanics of territory and city capture. Areas with a high density of cities are especially valuable to take control of, if you can manage to do it, which tends to be difficult.
There are many problems here. Taking a neutral city costs 40 troops, taking a long time to recuperate. You're better off looking for the enemy and attacking immediately, only taking neutral cities in the late game when it takes 20+ moves to deploy men against the enemy.
Attacking the enemy is better than taking neutral territory, n.b. you want to do this towards the end of the turn so your conquered territory will respawn.
I don't propose taking a city, waiting for recuperation, and then attacking an opponent. I propose taking a city, then soon taking another, then another, then another. Basically moving along an exponential growth curve. If you get lucky and no one attacks you during that period, you stand a very good chance of winning. Even if someone does attack you, the rapid regen from all your cities gives you a fighting chance. This strategy works especially well if you discover a lot of cities in your region during the early game.
The problem with attacking an enemy is that it's easy to get into a war of attrition where you both fall behind other players.
I think it's important to emphasize that you won't win against random luck without the rush strategy:
- If you begin in a corner (best) or side (ok), you can win if the side opposite to the one you are rushing towards does not rush you. If you begin in the center there is no way to win.
- To rush wait until ~17 because then you'll fill a full run before your lands get 2, enabling a second rush if the first fails.
If you win against one player before 50 seconds the chances get exponentially higher to win. Beat 2 in the first 100 seconds and you probably have 90% chance of steamrolling the entire board.
Was going to reply to say the same thing. It's tempting to take neutral cities at the start and sometimes I do it, but it's almost always a mistake, and I can usually tell when other players do it and they make easy pickings. It takes a long time to recover from taking a city, and the payoff isn't worth it until you have a lot of them.
It's better to cover a large swathe of territory. This can hide where your general is, and every 25 turns you can collect armies from the squares. Build up that way and then you wait for someone next to you to get into a fight with someone else, or take a neutral city, and then bomb in on them in force when they can't defend, and take advantage of all of their hard work.
There's also a bit of a "dark forest" aspect to the game, where you can benefit from being hidden for a long time at the beginning of the game, but you have to trade this off with the disadvantage of not being able to build adequate forces in that time.
I'm "bubbles" in the game.
In high level play you have to manage your land, the hard thing is how to find the optimal route to collect your army: It's advisable to take around 15-18 turns to collect your army from an area and attack in the remaining 10-7 turns, so you can attack before the enemy numbers increase. In high level 1v1 play what matters most is the area you have, if you can get +1 area per 25 rounds from your enemy you'll probably win.
But! As both of you collect more and more area it gets really time consuming and after a point you just can't collect your army optimally. I'd argue that the biggest difference between skill levels at the highest levels is how efficient the collecting is. At that point you'll need cities. Always look for the opponent's counter to know how many cities your opponent has.
Not that it matters but I was a top 5 1v1 player quite a while ago.
If the GP wanted to hide it, it’s a pretty strange thing to explicitly unhide it.
I appreciate it
I appreciate it too. The deciphered version is clearly noted as such, which even respects the OP intent in giving readers the choice to skip reading it.
It's just one person's perspective on strategy not some sort of spoilers or cheat codes.
To be honest it's a strange thing to rot13. I get why OP did it, but I'm really not sure the benefit outweighs the negatives or that it accomplished what it was supposed to. Those aren't exactly spoilers in the same way as visual media, where a fraction of a second can reveal the whole story.
Brb painting a shed.
Oh, well done mate! gratz. Maybe he ROT'd it for a reason.
Yes, and I unROt'd it for a reason too
Or you could have posted a link to an unrot site instead of possibly spoiling it for others.
Seriously just post the strategy with a spoiler alert.
This game is a dopamine trap that has been bad for my brain (since 2020). I have a bogus entry in my /etc/hosts to keep me from playing it, and yet I too often disable it because I can't help myself. It's like crack.
Yeah I had the same issue with it in like 2017-18. Very addictive and ultimately unsatisfying gameplay. Juice not worth the squeeze. If you struggle with distraction and dopamine loops just avoid it. You aren’t missing much.
I also did the hosts file block and removed it. It’s just like any addictive behavior though: break the habit for a week or two and you won’t miss it. For me I was visiting family for 2 weeks and they had terrible internet so I couldn’t win the game anyway.
Had the same issue with Supremacy 1914 and I ended up getting the shakes from it. Had to go cold turkey
No it honestly is - the only productive thing to come out of my 5-hour long gameplay session was this blog rant: https://www.cirex.dev/blog/dopamine-hits-hard
the game is a big trap... why is the post still on the HN home page...
I keep losing games because "you went AFK" even though I've been moving units. What is considered "AFK"? Pretty frustrating tbh
in FFA, you need to take at least 10 tiles by turn 60. That means 11 tiles including your general. But if you encounter another player before turn 60 it won't make you afk no matter what.
(And note that to play well you'd want more like one tile every two-or-less seconds in FFA.)
What do you think are they using for the map?
HTML table
One of the original devs here. The map is just html with position offsets!
This might have the best UX for onboarding I've ever seen. The game is pretty damn good too. Turn based but real time is really interesting.
EDIT Ok this is addicting. Mobile version when? It is playable on mobile but having a touchable direction pad would work wonders.
You need WASD and E... and mouse. I have bluetooth keyboard + mouse on my tablet. XD
Check out Utopia-Game! MUGA also had a hand in that and it's a personal favorite.
It appears as though, for the keyboard shortcuts, they are matching on the "key", as in the actual letter, instead of the "code", which is more accurate regarding the actual position of the key.
Just a minor annoyance for alternative keyboard layout users. Cool game, though!
This is something I never thought about. So key codes are based on the physical location of the key on a standard layout/qwerty, and changing your layout will cause the key value to change but not the code?
Yes, approximately. In X11 at least, the hardware codes are called keycodes, and the human readable names are called keysyms. Example of the XkbKeycodeToKeysym function in use: https://github.com/anko/xkbcat/blob/8abc3402cb679027a3bd0313...
Keysyms don't necessarily strictly match "key location"; keyboards are allowed to output whatever codes they want for whatever key they want (see e.g. QMK firmware; often used in custom keyboards to do complex conditional key remapping), but they're in practice relatively consistent between common keys on most keyboards, and consistent on the same keyboard even if you switch keyboard layouts in software, unless you have some custom keyboard firmware which functionality is stateful.
This game frustrates me because city positions and explored/unexplored information aren't retained on the map (vs Age of Empires or any game with a "fog of war"). You can be scouring the map in a tight race to find your opponents capital with no idea what tiles you've already explored or where previously held cities are.
I actually really enjoy that aspect of the game. It gives an extra layer of skill to the game, where you can outwit your opponent by quickly capturing back territory to restore the fog of war before they notice that they walked right by your general. But you don't know for sure that they missed it, so you're still on edge even if you're actually in the clear.
Having to keep track of where the enemy general was in the moment of crisis (as you scramble to get a larger army out there) is also honestly kind of fun.
I saw this the first time around. Loved it. Then I started losing quickly…
Anyway. Would do again.
Just tried this out for the first time and I'm hooked! It even works OK on mobile
I just played a few matches on my phone and there seems to be some bug that can detect you as afk and loose you the game for no apparent reason (i.e. while making moves). It's really a good game otherwise.
"This username is not okay." c'mon "barfnuggets" is not that offensive
rematch
This used to be the game I played during long nights at my previous startup when I had downtime.
Both good and bad memories.
Blast from the past! I played this a bunch when it was first posted here, rose to #2 on the 8v8 US leader board, then went on to make a similar game called Starjack which hit ~1k concurrent players in 2019 (although it's no longer available).
Luck plays a role, but the game is almost entirely about reading what other players are doing and thinking and reacting appropriately. It's poker, not chess. I had a win rate significantly over 50%, and hit a greater than 10 win streak at one point in 8v8.
To add to some strategies posted here - although I haven't played in five years (around the time that movement started to be buffered), so things may have changed. I played a quick game to jog my memory while writing this (and won it of course!)
The army count list tells you even more than exploring, and correctly reading what's happening is key to high level play.
1) An AFK player will slowly accumulate units on a predictable schedule. They may or may not come back. AFK to build up units and suddenly attack is a poor strategy, so you don't have to be worried about them, but there are edge cases where you can get stung.
2) Players that have army counts that drop together are adjacent - use this to understand where on the grid players are.
3) If only a single player has an army count drop, they're attacking a neutral city. It's a good time to attack as they've made a large, long term investment.
4) Players that have expanded but not lost any armies in a while are looking for a chance to attack - be careful if you're next to them.
5) When players fighting a war of attrition, stubbornly trading armies when they're not the only two left, they're probably not very good. A well timed swoop in at the end of the fight will capture all of the production. Don't get involved before that, because they're not very good and will start a war of attrition with you :)
6) At the start, neutral territory has an immense ROI. Capture as much as you possibly can. Don't worry too much about cities.
7) Generally, neutral territory with high numbers at the end of the game correlates with starting locations.
8) Pay a lot of attention to the star ratings of the players to know who is good and who isn't. Picking on beginners is fantastic strategy, because you can generally capture their armies intact.
9) Pay just as much attention to how well people are playing. Expert players will maximise neutral ground, and after the first pop at turn 25, you'll immediately see who you have to pay attention to.
10) If a player has launched a massed attack on you, it's often a better strategy to counterattack if you think you know where their base is better than they know where yours is. If you win, you'll capture their army intact and this often guarantees a win.
11) If you're not in the lead, throw everything you have into conflict where you think you have advantage. You are at this point "default dead", and if you play conservatively you'll lose.
12) If you are in the lead, you can now profitably consolidate cities without the power drop opening you up to attacks. Keep a good chunk of an army close to your base to prevent sneak attacks, and expand out to neutral territory as quickly as you can, preparing to jump on anyone that's weak. You're "default alive".
I am loving this
I used to play this game all the time - great blast from the past seeing it posted today. It's like a mini RTS.
I love how easy it is to jump into/out of a game quickly, at most, you're only ever committing 10 minutes.
I love this game, got pretty into it around 2016/17 and rose to the top of the global 1v1 leaderboard for a while. Really cool to see it still running :D
I'm always amazed how a simple game can be fun
Great game, great UX.
On suggestion for the UI: in replay the POV checkbox should be unfocused after click, or event bubbling stopped, as space-bar toggles both the autoplay and the POV now.
This is pretty fun
I thought the turns are fixed
I tried playing but it booted me for being "AFK" even though I was tapping the whole time
Neat clone of empire attack (now defunct, https://www.ianandrew.com/empire-attack). Used to play the 10 day variants there which were great fun, although more defense focused I think.
nice little game!
I wonder why it doesn't offer any kind of higher level controls. Surely it gets old to have to click all the tiles repeatedly to collect the army
How much does a game/IP like that sell for, if I may ask?
2-3x yearly net revenue, more if you're lucky.
Revenue is ultimately a function of DAU, either via ad RPMs or IAP / subscription conversion rate.
Also most sales agreements have a confidentiality covenant.
And what is the yearly net revenue of a game like generals?
Yeah that's super annoying. I'm mainly just interested in the ballpark. $10k? $50k? $100k? $500k? >$1m?
Just curious, how would that ballpark number be helpful?
This thread has inspired me to do further research and explore monetizing/selling some of my side projects. I am curious how a ballpark sales value is useful...
Because it's interesting to know if my hypothetical video game similar to this is worth a few beers on a night, a fancy dinner, a year's salary, or enough to set me for life a la Microsoft's purchase of Minecraft. Depending on that information, someone could choose to build a game as a labor of love, build a game as a for-profit venture (haha), or decide to go do something else with their life.
Kind of rude to ask, if I had to guess less than $50k a year.
Kind of thing you put up and maintain because of the community you foster. Labor of love, not business.
Minecraft sold for 2,5B when having sold 1B, now it sold 10B.
Largely due to Microsoft's strategic direction. I doubt Minecraft would have grown 10x under notch's leadership.
What big changes has Microsoft made? The whole Java/Bedrock split seems like a mess, and the updates have been enjoyable, but certainly nothing to write home about.
The only vaguely interesting thing I can point to is Minecraft Dungeons, which was fun for a month for my friend group, and then it wasn't.
Go walk around any toy section...
+1 also curious
ballpark is $10k - $100k. Nothing too crazy but still nice to walk away from the project with something
hi victor, good to see you here! hope you're well :)
wow took me a bit to remember your username (time flies) but great to see you here too! :)
Did you ever consider to make the game persistent and MMO?
If so what made you resign that?