1 unit per hex: failed experiment
- Thread starter ohioastronomy
- Start date
1upt works just as well as stacks. The problem's the AI.
I always think of this when I see a thread like this. Go back and play Civ4, and examine on the war AI. It's the most pathetic thing I've ever seen. The fact that the AI in Civ5 is at least somewhat competent and doesn't require a 10:1 ratio (probably more like 3:1 or 5:1) to win against me is good. Of course I want the AI much better, but it's leagues ahead of Civ4.
I always think of this when I see a thread like this. Go back and play Civ4, and examine on the war AI. It's the most pathetic thing I've ever seen. The fact that the AI in Civ5 is at least somewhat competent and doesn't require a 10:1 ratio (probably more like 3:1 or 5:1) to win against me is good. Of course I want the AI much better, but it's leagues ahead of Civ4.
I disagree, 1UPT is the best feature of CIV 5. The fact that it favors the use of small number of stong forces over a mass of weaker forces is a bonus.
The only way to have a better army before was to have more units in your stack. Spam more units than the other guys and you will get a larger stack and larger stacks win.
Now a small but effective force using terrain and tactics can deal with a larger forces that is either weaker or using poor tactics. I love the new mechanic and I'm having a LOT of fun with it.
---
That said, if you want to have unlimited stacking, but also make it so that its bad to put all your units in one stack, you would make it so that EVERY unit does collateral damage to a stack. For example, if your stack gets attacked, all units in that stack take damage. This would be best for range attackers, especially seige units, that would effectively get to do "free" damage to all units in the stack.
Secondly, you could implement an "overcrowding penallty". The simple way is a flat -15% to all units in the stack (and no flanking bonus). Better may be -10% for each unit in a stack, which mean could put 10 units in a stack, but would be at -100%.
The only way to have a better army before was to have more units in your stack. Spam more units than the other guys and you will get a larger stack and larger stacks win.
Now a small but effective force using terrain and tactics can deal with a larger forces that is either weaker or using poor tactics. I love the new mechanic and I'm having a LOT of fun with it.
---
That said, if you want to have unlimited stacking, but also make it so that its bad to put all your units in one stack, you would make it so that EVERY unit does collateral damage to a stack. For example, if your stack gets attacked, all units in that stack take damage. This would be best for range attackers, especially seige units, that would effectively get to do "free" damage to all units in the stack.
Secondly, you could implement an "overcrowding penallty". The simple way is a flat -15% to all units in the stack (and no flanking bonus). Better may be -10% for each unit in a stack, which mean could put 10 units in a stack, but would be at -100%.
The AI needs to compute values per hex-tile:
- Moving there puts the unit in a position that it can receive an amount of damage (city bombardment, artillery, units that can reach it.
- Moving to a certain tile can create a more dominant position: artillery can hit oponents, units can reach other units with favorable results, terrain give odds bonuses if they get attacked, etc...
It should then determine a move, or decide to not move, which the AI is not capable off atm.
At the moment I think the AI doesn't do this at all. They only look at enemy units, and the battle odds. Besides that they completely ignore each tile as a possible important spot to go to, or to not go to. Then if the AI would calculate tile values, it could base unit movement on that principle.
I think that's the easiest approach for the developers for now, I think it's still too hard to make it actually plan out a campaign. Although I've had the AI successfully conquer one of my cities, in quite an impressive way recently, 2 archers shooting at the city, infantry on the attack. But if I would've had a real good defense around that city, they would surely have failed. I had almost no defense there at the time.
- Moving there puts the unit in a position that it can receive an amount of damage (city bombardment, artillery, units that can reach it.
- Moving to a certain tile can create a more dominant position: artillery can hit oponents, units can reach other units with favorable results, terrain give odds bonuses if they get attacked, etc...
It should then determine a move, or decide to not move, which the AI is not capable off atm.
At the moment I think the AI doesn't do this at all. They only look at enemy units, and the battle odds. Besides that they completely ignore each tile as a possible important spot to go to, or to not go to. Then if the AI would calculate tile values, it could base unit movement on that principle.
I think that's the easiest approach for the developers for now, I think it's still too hard to make it actually plan out a campaign. Although I've had the AI successfully conquer one of my cities, in quite an impressive way recently, 2 archers shooting at the city, infantry on the attack. But if I would've had a real good defense around that city, they would surely have failed. I had almost no defense there at the time.
A recent poll already stated the vast majority of civ players considers the hex system and the 1UPT the best feature ever seen on last 5 years.
Told that, I can understand someone is missing Final Fanatsy but I am sorry for you but this is a strategic game.
Told that, I can understand someone is missing Final Fanatsy but I am sorry for you but this is a strategic game.
Usually on small maps my army is around a dozen of units within middle eras, I can reach the double on late stages so I can't understand how you can maintain 150 units and simoultaneously having a solid economy that leads you to win.
Prince level BTW
Not just the human player, but the AI players too.
http://img801.imageshack.us/img801/8246/civ5screen0023e.jpg
http://img814.imageshack.us/img814/4739/civ5screen0024.jpg
Yep, those AIs (Deity level) are using lots of strategy right there.
Looks like time to bring out the nukes to me...
jeffreyac
Mostly Harmless
I think the OP was a pretty well thought out and constructive criticism of Civ V. Note it's not just "This game sucks!" but a detailed and well thought out reasoning of why the poster feels how he (or she; sorry, didn't look!) feels.
And, frankly, though I'm trying to like the game, I'm finding that what is said in the original post is pretty much spot on...
And, frankly, though I'm trying to like the game, I'm finding that what is said in the original post is pretty much spot on...
Not just the human player, but the AI players too.
http://img801.imageshack.us/img801/8246/civ5screen0023e.jpg
http://img814.imageshack.us/img814/4739/civ5screen0024.jpg
Yep, those AIs (Deity level) are using lots of strategy right there.
Ehehe it is the typical AI cheating.... I have never seen such a mess BTW but I usually play on Prince. On Deity ... considering the number of units the AI can mainatin I would never play against such a cheater, it is like to play chess where the opponent has 26 bishops , 134 knights and of course 16 Queens.
Antmanbrooks
Prince
Sounds like a queue for a club in Canal Street.
cephalo
Deity
I completely disagree with the OP. As far as I'm concerned, late game tedium is fixed, which I think it's a huge acheivement for TBS. The complaints of 'realism' all apply equally to the stack system. There might be some room for improvement in the AI and so forth, but by no means is this system a 'failed experiment'.
In particular, I have to take issue with point number 5. 'Surface area' is a hugely important military concept. Those traffic jams have doomed many large armies in history. At Cannae, the superior Roman force was surrounded such that the guys in the middle could not join the fight!
The whole plan for operation Overlord was a question of how to get your superior force into France in such away that they can exploit their superiority? Landing everyone on the Beach would be doomed to failure precisely for such traffic jams as a large army is prone to have.
The AI is having trouble dealing with these issues now, but I bet by the end of the Civ5 era you'll be facing the 'Deep Blue' of civ combat, and wish for the days when you could win! (Ok, thats an exaggeration. I gotta call myself out there.)
Never again will I assemble 50 units over 20 turns and drag them around the map in a stack. That is not a failed experiment, it's a no brainer.
In particular, I have to take issue with point number 5. 'Surface area' is a hugely important military concept. Those traffic jams have doomed many large armies in history. At Cannae, the superior Roman force was surrounded such that the guys in the middle could not join the fight!
The whole plan for operation Overlord was a question of how to get your superior force into France in such away that they can exploit their superiority? Landing everyone on the Beach would be doomed to failure precisely for such traffic jams as a large army is prone to have.
The AI is having trouble dealing with these issues now, but I bet by the end of the Civ5 era you'll be facing the 'Deep Blue' of civ combat, and wish for the days when you could win! (Ok, thats an exaggeration. I gotta call myself out there.)
Never again will I assemble 50 units over 20 turns and drag them around the map in a stack. That is not a failed experiment, it's a no brainer.
I hate being myself definitely not a mothertongue, I miss quite a few shots like this.
What does that mean ?
skallben
Diplomat
The absolutely acute issue that needs to be fixed is that AI workers, scouts and settlers blockade movement, open borders should fix that. . Especially on small maps but even on bigger, narrow corridors get blocked by workers who stand there for like 10 turns when they build stuff.
Moving a decent sized army takes ages through friendly territory, trying to move your units in a decent formation can sometimes just be a no-go.
Moving a decent sized army takes ages through friendly territory, trying to move your units in a decent formation can sometimes just be a no-go.
Antmanbrooks
Prince
It's not a problem with your use of the English language. It was just a cheap joke.
A Queen is another word for a cross-dresser or a transvestite.
Canal Street is usually a popular Gay, Bi and Tranny zone in Manchester and other UK Cities.
Therefore 16 Queens could quite easily be seen queuing for a club in Canal Street.
It's not a problem with your use of the English language. It was just a cheap joke.
A Queen is another word for a cross-dresser or a transvestite.
Canal Street is usually a popular Gay, Bi and Tranny zone in Manchester and other UK Cities.
Therefore 16 Queens could quite easily be seen queuing for a club in Canal Street.
I was missing Canal Street ... you are right.....
Glad to see you enjoy that club BTW
I disagree with the OP.
1) It makes the mechanics of moving clunky and the game run very slow.
I haven't noticed this but the point about workers needing time to build a road is rather poor. Workers stack with friendly military units, so they are only hindered by other workers and unfriendly military units. Stacking workers to mass produce a road in one turn doesn't make sense. A better solution imo would be public works anyway, so no worker at all.
2) Single NPC units can perma-block roads in neutral territory, and frequently do.
Which is very realistic. Ask a military what he'd expect of a simulation where one road is blocked by a lone unit. exactly that. Traffic jams exist.
3) Civ was designed around a different paradigm
No. Play Civ I and stack your units outside cities. The ai will slaughter you. This argument doesn't make sense.
4) It is inappropriate for the scale of the game.
So are almost all wars. Again this is not a valid argument. World War I took something like 4 turns in Civ. And WWII for America would be 3, or maybe 6? The scale of military operations has always been totally distorted, so this point is as irrelevant as point 3.
4) It's prone to artificial tactics
I would rather have a "we go" system myself but that's never been the civ way. If you changed that, you could counter it with your own argument number 3.
5) The new problems created with 1 unit are worse than the big stack problem they solved.
This opinion I disagree with.
6) The above problems cripple the AI as an opponent.
The AI has other problems. Give it chokepoints and it can defend them properly. Civ AI has never been good at warmongering, I don't think this makes the problem worse.
1) It makes the mechanics of moving clunky and the game run very slow.
I haven't noticed this but the point about workers needing time to build a road is rather poor. Workers stack with friendly military units, so they are only hindered by other workers and unfriendly military units. Stacking workers to mass produce a road in one turn doesn't make sense. A better solution imo would be public works anyway, so no worker at all.
2) Single NPC units can perma-block roads in neutral territory, and frequently do.
Which is very realistic. Ask a military what he'd expect of a simulation where one road is blocked by a lone unit. exactly that. Traffic jams exist.
3) Civ was designed around a different paradigm
No. Play Civ I and stack your units outside cities. The ai will slaughter you. This argument doesn't make sense.
4) It is inappropriate for the scale of the game.
So are almost all wars. Again this is not a valid argument. World War I took something like 4 turns in Civ. And WWII for America would be 3, or maybe 6? The scale of military operations has always been totally distorted, so this point is as irrelevant as point 3.
4) It's prone to artificial tactics
I would rather have a "we go" system myself but that's never been the civ way. If you changed that, you could counter it with your own argument number 3.
5) The new problems created with 1 unit are worse than the big stack problem they solved.
This opinion I disagree with.
6) The above problems cripple the AI as an opponent.
The AI has other problems. Give it chokepoints and it can defend them properly. Civ AI has never been good at warmongering, I don't think this makes the problem worse.
ohioastronomy
King
- Joined
- Dec 14, 2005
- Messages
- 714
I like the appeal of a tactical system - I played face to face wargames for many years and still do occasionally. The problem for me is that Civ 5 just didn't seem to learn from the known drawbacks to this method. Letting me move all of my pieces at once, for example, gives enormous advantages to the offense (since the defender can't react and withdraw wounded units.) You don't then add the ability to heal on the fly or attack and retreat out of range - those are "I win" buttons, not strategy. If you want no stacking limits you need room to move pieces around so that you don't get logjams - or you end up replacing strategy by "killing units moved foolishly because of the pathing algorithm".
One unit per tile might work well if the battles were resolved on a tactical map. You could have want enough units on the board for tactics, and at the same time a low piece density. It just doesn't work for the scale of the game as it is.
