1UPT - final verdict?

[Callonia]

1upt is still good idea, firaxis just didn't put enough work in testing to make AI use it good.

After all, there is AI fixes in mod that firaxis has yet to do.

 

[Vaclac]

Maybe knick Europa Universalis 4's attrition/manpower to punish stacks of doom - you can push a big stack together for a one off battle, but without supplies they'll soon start starving and you'll lose men?

Anyway 1 upt has to go.

I think this is the way to go, could have all the units take 10 points per unit beyond the first ending its turn on the same tile, kind of like miasma right now. Perhaps the limit should be higher than just one - you could even increase it over time with some logistics-type tech or make it terrain-specific, plots with more food allowing more units to finish a turn on them before the attrition damage kicks in. But anyway, some variant on this system would be helpful to avoid the extremely annoying absolute logjam that occurs with carpets of doom, while still encouraging armies to be fairly spread out and having something other than SOD warfare.

 

[santoo]

Something I found interesting in the discussion so far:
MUPT is mostly equated with stacks of doom - historically, that's understandably.
However, there have been a lot of developments in the last years to discourage SoDs in CIV4 mods:

• collateral damage: bombard a stack with an artillery-type unit and damage several/all units in the stack. alternatively: suicide-dive 3 or 4 artilleries into an enemy SoD, and it's gone. Whoops.
• flanking: attack a stack with a cavalry unit & damage several/all of the slow artillery units at once
• surround & destroy: attacking bonuses, based on how many tiles surrounding the target you have units on

In my experience, SoDs have become much rarer - mostly only appearing, where they make sense, ie defending -but not attacking- cities, guarding fortresses, at choke points or on that single forested hill tile, overlooking the endless plains

 

[kirbdog]

As a side project I made a chess-style tactical AI for civ and it works.

This sounds interesting. Can I study the source code?

 

[jokii]

well... that was surprising.
seems like the vast majority of 1UPT-proponents have in fact started out with MUPT - which I never would have guessed.
Also: despite very vocal criticism of 1UPT (and related AI & scaling issues), most people prefer it to MUPT (thanks for the explanations offered for this divergence, btw!).
I'm glad so many have participated - and thanks for elucidating both sides

Many people play only occasionally, and many play economic part of game without ever going to war. So they like the game.... and they would probably like it even if there is no opponent in the game, but only the planet with aliens to settle and build an empire...

But I played Civ from Civilization 1. (skipped Civ3 and SMAC )
The war part of the game in early Civ games was something that was the foundation of game and now is something that is hidden, and in the background, because it is not functioning.
1UPT- not functioning in single player mode. Play on normal-war game and test it. AI is not weak, but completely lost.

The great old Civilization game has become Civilization-builder game.
And profit will almost certainly push it further in that direction.

 

[PhilBowles]

Something I found interesting in the discussion so far:
MUPT is mostly equated with stacks of doom - historically, that's understandably.
However, there have been a lot of developments in the last years to discourage SoDs in CIV4 mods:

• collateral damage: bombard a stack with an artillery-type unit and damage several/all units in the stack. alternatively: suicide-dive 3 or 4 artilleries into an enemy SoD, and it's gone. Whoops.
• flanking: attack a stack with a cavalry unit & damage several/all of the slow artillery units at once
• surround & destroy: attacking bonuses, based on how many tiles surrounding the target you have units on

In my experience, SoDs have become much rarer - mostly only appearing, where they make sense, ie defending -but not attacking- cities, guarding fortresses, at choke points or on that single forested hill tile, overlooking the endless plains

My own point seems to have been somewhat neglected, but for me, while SoD is a convenient moniker for the system as a whole, the issue is less with the stacks and more the game mechanics they need to support them. Multiple units per tile = a need to build more units. Build times need to be shorter and build queues repetitive. Unit production in quantity is forced by the system, and repeatedly building the same units over and over really isn't a very engaging playstyle. The height of strategy it promotes is just to specialise one city as a unit factory - obvious and simple-minded enough an approach to army management that even Total War: Rome 2 does it. There's been a lot made of the tediousness of moving units in 1UPT, but moving and rearranging stacks is at least equally tedious once they're spread across your empire to garrison and suppress unhappiness; you'll usually have more stacks in Civ IV than units in Civ V, and without suboptimal unit combinations that excise slower units they'll be moving at a very slow rate.

 

[kirbdog]

If I had time, I'm certain I could mathematically prove that you are wrong.

In Chess, with an 8x8 board (less than 1/5th the size of a civ board) and the ability to move *only one unit per turn*, the number of possible outcomes is so vast that it takes a super-computer and TONS of heuristics to beat a skilled player.

The goal is to make the game fun, or more fun, not to beat Magnus.

A look-ahead function of 1 turn (2 ply) or 2 turns (3-4 ply), even with a cheap eval function and some dirty heuristic tree pruning, is all this game would need to become much more fun.

Also, a cell phone can beat 99% of skilled human players at chess.

 

[santoo]

My own point seems to have been somewhat neglected, but for me, while SoD is a convenient moniker for the system as a whole, the issue is less with the stacks and more the game mechanics they need to support them. Multiple units per tile = a need to build more units. Build times need to be shorter and build queues repetitive.

I can see where you are coming from, but I think that balance/pacing has been mentioned a couple times in this thread. It is, however, the first time I saw the game mechanics necessary for 1UPT/MUPT being used as an argument for 1UPT.
I remember in the early days of Civ5 there were some really well thought through analyses of 1UPT vs MUPT and the most important aspects seemed to be not the tactical advantages, not the AI, not SoD vs CoD, but how the core-mechanics of CIV had to be changed to allow for 1UPT and how that was detrimental to the game.
If you are interested: look e.g. here - skip far far down to the 2nd time you see yellow text (or CTRL-F for "I believe that these problems"). A lot of the issues mentioned have been solved or at least worked around in patches/DLCs/expansions, but I wonder whether the core criticism does not remain valid.
This does not negate your point - just shows that you can approach the issue of mechanics from both sides.

 

[Thormodr]

People have mainly discussed the problem that the AI can't handle 1UPT. Fighting wars is literally like taking candy from a baby. I just don't see them devoting the resources to fixing it.

From the mouth of Jon Shafer:

One of the biggest challenges unearthed by 1UPT was writing a competent combat AI. I wasn’t the one who developed this particular AI subsystem, and the member of the team who was tasked with this did a great job of making lemonade out of the design lemons I’d given him. Needless to say, programming an AI which can effectively maneuver dozens of units around in extremely tactically-confined spaces is incredibly difficult.

The reason why this wasn’t an issue in Panzer General was that their AI didn’t actually need to do anything. It was always on the defensive, and a large part of that game was simply solving the “puzzle” of how to best crack open enemy strongholds. It was plenty sufficient if your opponents simply ordered a single tank to stir up some trouble every so often.

What made Panzer General fun was you blitzkrieg-ing through Europe while your enemies quickly and dramatically fell before your might. However, in a Civ game, the AI has to be capable of launching full-scale invasions, sometimes on different landmasses. Needless to say, we’re talking about a challenge on completely different scale.

Speaking of scale, another significant issue with 1UPT was that the maps wasn’t really suited for it. The joy of Panzer General was pulling off clever maneuvers and secretly encircling your helpless enemies. Unfortunately, in Civ 5 nasty bottlenecks aren’t uncommon and this tempers much of the natural value added by 1UPT. Ultimately, there just wasn’t enough room to do the fun part.

http://www.gamasutra.com/blogs/JonShafer/20130218/186843/Revisiting_the_Design_of_Civ_5.php

They certainly forgot the fun part. Incidentally, Jon Shafer's upcoming game At the Gates will not be using 1UPT. I wonder why that is?

However, there is another problem that hasn't been discussed much. A few people may have touched on it a bit.

In order to avoid the "Carpet of Doom" (Having units occupy literally every hex of the map) which nobody wants, you are forced to make changes in many areas to accommodate 1UPT. You need to vastly slow down production, reduce hex yields, fiddle with science, etc. This destroys the pacing for the game and makes the game tedious.

This critique sums up very well the huge problems that 1UPT creates:

http://www.garath.net/Sullla/Civ5/whatwentwrong.html

EDIT: While I was writing my post I see I was beaten to the punch.

All in all, Firaxis just has to get creative. No 1UPT but no SOD, either.
There is a good discussion on alternatives to 1UPT here. Valkrionn has some excellent ideas in particular:

http://forums.2k.com/showthread.php?106922-alternatives-to-1upt

 

[Lord Tirian]

In order to avoid the "Carpet of Doom" (Having units occupy literally every hex of the map) which nobody wants, you are forced to make changes in many areas to accommodate 1UPT.

Or implement unit upkeep factored into unit production. You have 5 units? Units are 5% more expensive. You have 10 units? Units are 20% more expensive. You have 20 units? Units are 80% more expensive.

I think the economic impact of 1UPT wasn't solely to 1UPT alone but also due to Firaxis' will to make the game a bit more "boardgame-like", i.e. have less cumulative effects and more transparent mechanics.

Blaming 1UPT for that is easy but not guaranteed to be the case - and in G+K and BNW, they tightened up the pacing a lot without touching unit production and overall 1UPT too much. Designing the system as it was in Civ5 vanilla was a choice not an inevitable outcome of implementing 1UPT.

 

[PhilBowles]

I can see where you are coming from, but I think that balance/pacing has been mentioned a couple times in this thread. It is, however, the first time I saw the game mechanics necessary for 1UPT/MUPT being used as an argument for 1UPT.
I remember in the early days of Civ5 there were some really well thought through analyses of 1UPT vs MUPT and the most important aspects seemed to be not the tactical advantages, not the AI, not SoD vs CoD, but how the core-mechanics of CIV had to be changed to allow for 1UPT and how that was detrimental to the game.

I remember those threads, but back then as well considered the reverse to be true: the key change 1UPT enabled was a slowdown in building or tech choices, which in the context of a game with the same number of turns made tech order and trade-offs between building options more important strategically, while in past games building anything you wanted was feasible. When I made this point I was met with the objection: if you spend too much time building duplicate buildings, you aren't building enough units. Which, in the context of my argument, is circular reasoning.

Having to constantly click 'spearman' (unless it's queued), build a grocer in each of X cities, or change techs every two or three turns gives the illusion that there's more to do in past Civ games than there is - it's mostly repetitive busywork that doesn't do a great deal to progress gameplay, and much of which is forced (such as 'do something to manage health in this city at pop X, manage happiness at pop Y'). Far from 'destroying' game pacing, I feel Civ V's approach improves it. I wouldn't disagree that Civ V doesn't add enough 'active' behaviour that you can engage in while waiting for the next tech or building to pop, but I wouldn't blame this on 1UPT or the changes it enabled. Game actions should be relevant to achieving an objective and management should exist to promote decision-making, not to set fixed, arbitrary penalties with repetitive solutions.

Of course, Civ V threw a spanner in the work by throwing gold purchasing into the mix, and despite making good strides in curtailing the economy at least in the early game, for some reason never simply removed or limited this feature. But that can't be blamed on 1UPT.

 

[Thormodr]

Or implement unit upkeep factored into unit production. You have 5 units? Units are 5% more expensive. You have 10 units? Units are 20% more expensive. You have 20 units? Units are 80% more expensive.

I think the economic impact of 1UPT wasn't solely to 1UPT alone but also due to Firaxis' will to make the game a bit more "boardgame-like", i.e. have less cumulative effects and more transparent mechanics.

Blaming 1UPT for that is easy but not guaranteed to be the case - and in G+K and BNW, they tightened up the pacing a lot without touching unit production and overall 1UPT too much.

Perhaps they did want the game to be more "boardgame-like". I do actually like playing boardgames that have some meat to them, I'll admit. In fact, I have played a lot more boardgames in the last few years due to Civilization 5. Still, I think they are chasing the wrong audience if they are going after boardgamers. One doesn't have to spend much time on Boardgamegeek to see that.

Computers can do so much more than boardgames and their relatively simple rules and I'd prefer they took advantage of that. I don't want Civ over a lunch hour. I want a Civ game that lasts for days upon end.

 

[SupremacyKing2]

All in all, Firaxis just has to get creative. No 1UPT but no SOD, either.
There is a good discussion on alternatives to 1UPT here. Valkrionn has some excellent ideas in particular:

Valkrionn has some good ideas. I think every complaint against MUPT could be solved without going to 1UPT:

1) Stacks of Dooms
- Give units a maintenance cost is gold, hammers and food. This would make it impossible for cities to just spam units indefinitely since at some point, your cities would completely stagnate.
2) Battles always taking place at cities
- When city takes damage from bombardment, have city also lose a building each turn. Also, auto destroy any tile improvement that a battle takes place on. This would encourage the defender to try to meet the enemy away from cities to avoid losing buildings and tile improvements.
3) MUPT has no strategy or tactics: just brute force your stack against another stack.
- Give units more special abilities like pikeman beat cavalry so that if I try to just brute force my cavalry against a stack that has 1 pikeman unit it, I will take a lot more damage. Give stacks bonuses so that stacks have more strength when using combined arms.
-Keep flanking bonuses and such so that players can still move a smaller stack of cavalry around to get a bonus strike against infantry for example. Have stacks take collateral damage based on size. Large stacks could be withered down by well placed bombardment.
- Make terrain matter more. Rivers can give defensive bonuses. Hills can give offensive bonus to cavalry attacking from a hill to a plain. Choke points would give combat penalty to stack based on size.
- Have units take attrition in enemy territory so that the defender that has a large territory has a viable strategy of playing a waiting game (think russia in WW2).

But by allowing some stacking but not SOD, units could still move around a lot easier and it would solve the terrible AI path finding problems of 1UPT.

 

[PhilBowles]

I think the economic impact of 1UPT wasn't solely to 1UPT alone but also due to Firaxis' will to make the game a bit more "boardgame-like", i.e. have less cumulative effects and more transparent mechanics.

That may have been part of it, but I think also there's a graphics issue. The designers have made a decision to favour improved animation, while pre-Civ IV graphics were mostly static. Animators naturally want their animations to be long enough to be visible, and with the large stacks in Civ IV mostly you wanted to play with animations off because battles just took too long between stacks otherwise.

The designers don't want to put the effort animation requires into a feature that players will turn off by default, and as animations become longer and more detailed, armies have to become smaller. As it is, Civ V players complain about the aircraft animations taking too long when aircraft are stacked. Look at Civ V unit animations, and now imagine just how long battles would take with, say, three times as many units per side.

 

[Tinker]

I like 1UPT, it's more some design philosophies that kind of frustrate me about it - like civilian units not being able to pass through opponents, civilian units not being stackable, being able to move a group of units together but having to click every single one separately, etc. Those aren't problems inherent to 1UPT, but rather things that could (and should've, with BE) been improved on to make 1UPT more feasible.

 

[Thormodr]

I like 1UPT, it's more some design philosophies that kind of frustrate me about it - like civilian units not being able to pass through opponents, civilian units not being stackable, being able to move a group of units together but having to click every single one separately, etc. Those aren't problems inherent to 1UPT, but rather things that could (and should've, with BE) been improved on to make 1UPT more feasible.

Yep. Like when you have a great person in the city and you build a space ship part and the game squawks at you to remove the dastardly stack. How does that make any sense?

 

[Leathaface]

I like both 1UPT and MUPT. Neither is better than the other for me.

 

[Lord Tirian]

Computers can do so much more than boardgames and their relatively simple rules and I'd prefer they took advantage of that.

On the other hand, just because a computer can do it, it doesn't mean you understand what it's doing (intuitively). I think something they really tried (and 1UPT makes sense in that context as well) is going for simple rules with complexity arising from the interactions.

An example are the Civ4 (and earlier) sliders: shifting around the culture slider causes complex changes, because it recalculates outputs, affecting culture, affecting yields again. But the effect is quite simple: you get more culture.

The Civ5 system is going exactly the opposite route: building up individual culture buildings and getting great artworks is in terms of rules, but due to the interaction with other rules systems (which are also fairly straightforward), you can get long-lasting effects.

Probably not the best and most concise examples, but lot of Civ5 decisions make a lot of sense if think about it that way: they wanted something like chess. Simple rules, complex play.

 

[lemony]

I'm really torn on this issue, but in general I absolutely prefer the multiple units per tile system.
The only problem with it is depending on how it's used, it makes for very little challenge end game. One of the easiest ways I've toppled AI's in Civ 4 was just by making a massive stack of the same 3 units and just mowing over all of their forces and cities.

However the IUPT is absolutely so annoying that I can't vote for it being a good system. Putting a cap on units per tile is understandable, but I think one unit is far too small and far too annoying a cap to put in place.

It's great that they tried to make it more strategic, but I don't think this is the best way to go about it. I'd prefer to lose the stack of doom strategy, but I think that a higher cap than one would be much more effective and garner a lot less unhappiness from the average player.

 

[Thormodr]

On the other hand, just because a computer can do it, it doesn't mean you understand what it's doing (intuitively). I think something they really tried (and 1UPT makes sense in that context as well) is going for simple rules with complexity arising from the interactions.

An example are the Civ4 (and earlier) sliders: shifting around the culture slider causes complex changes, because it recalculates outputs, affecting culture, affecting yields again. But the effect is quite simple: you get more culture.

The Civ5 system is going exactly the opposite route: building up individual culture buildings and getting great artworks is in terms of rules, but due to the interaction with other rules systems (which are also fairly straightforward), you can get long-lasting effects.

Probably not the best and most concise examples, but lot of Civ5 decisions make a lot of sense if think about it that way: they wanted something like chess. Simple rules, complex play.

I certainly respect your opinion. However, your conclusion seems to be a bit backwards.

Taking your chess analogy, I will say that if you want a game that was easy to understand but hard to master, that is cIV. If you want a game that easy to understand and easy to master, that is Civilization 5.

Anyway, even if 1UPT appears to be chess like, it is in effect like having a Grandmaster playing an average 5 year old. The combat in Civilization 5 is easy to learn, frustrating and tedious to operate but easy to master.