1UPT - final verdict?

[ahawk]

My thoughts with Mupt would be for the purpose of maneuverability only. It will be easier to arrange 3 siege units if they are moved in at the same time, but does not inflate combat. Tactical positioning would still be required. Brute force tactics would be minimized but possible if done in waves.

Yeah, this is a problem with 1UPT. Though I prefer 1UPT, I have always thought that it might have made sense to allow certain combat units to occupy a tile together, even if the limit was always just 2 units. In other words, maybe 2UPT would have made more sense than 1UPT, because then you would still have to consider positioning and terrain while having some ability to have units move 'through the lines' in order to bring up siege units or whatever.

As it is, I prefer 1UPT to stacks, but stacking did have some advantages and fun decision making, and at the same time 1UPT is frustrating because of having to move units for 5 minutes per turn in the late game.

 

[Aeson]

I agree with most of your points and would also like to see XUPT in a sandbox/mod (it's been a long time since CTP, but I think I liked armies well enough).

A question for the people who understand AI better than me:
Are there any fundamental reasons, why an AI would do better/worse in e.g. 5UPT vs the 1UPT and infinteUPT-cases? Moving would obviously be quicker/easier, and having more than one unit per tile would open up more combat options (and probably make positioning-mistakes by the AI more forgivable) - but would an exponentially increase search-space not only make it that much harder for the AI? My intuitive analogy: chess as a XUPT-game

Pathfinding is much easier when paths are not being dynamically blocked/unblocked by other paths which themselves are being dynamically blocked and unblocked. Increasing the UPT lowers the chances of such happening, but as long as there is a reachable hard limit the case has to be dealt with. With infinite UPT there is no such blocking. So it becomes vastly easier to say, move a bunch of units though a chokepoint other units are also passing through while keeping their formation.

 

[Aeson]

Realistically, you have a limited front against any enemy. Even if you are able to fit in 4 divisions, the terrain is often such that you can at most have 1 division or even only part of that division at the front. If you want all 4 divisions to engage at the same time, you need to surround that enemy from other directions, which translates to attacking from other adjacent tiles.

If one wants stack composition, I would rather prefer how Heart of Iron 3 does it. Each unit (division) consists of 4 brigades. Two types of brigades exist: combat brigades and support brigades. One would think that 4 combat brigades would give you the strongest division, but not so.

Each battle has a limited frontage. This can change depending on terrain, on target size, on whether you are surrounded or whether you are flanking. On a wide frontage, you can send the whole division to the line. On a narrow frontage, however, a division with say an artillery brigade will actually provide far higher firepower because the support brigade does not need to be on the line.

Also, realistically it takes time to rotate divisions off the line. Once on the line, they need time to do their work, meaning when a division is in combat, you cannot rotate the next division up. It's not an assembly line. Engaged divisions also cannot be rotated off the line because it creates chaos. This is why it is unrealistic to let every stacked unit attack in the same turn.

With the brigade concept, each division is in effect a stack of 4 brigades, and they work in unison just like how a stack of catapults plus swords plus archers would. The difference being that you have to micromanage a stack, but not a division.

If one really wants to talk about realistic combat, check out Hearts of Iron series. MUPT as it was in Civ IV is as unrealistic as it gets.

Civ is on a very different scale. Each tile is a huge area and each move is years or even decades of time.

 

[santoo]

1. Because 'soft' units need to be squishy. If you can stack glass canons with defensive units, they aren't glass canons anymore. The whole point of 1upt is to make positioning important. You allow stacks different types and it is no longer tactical but instead simply brute force.

3. Moving in (lets assume 3 per stack) 24-48 units attacking in the same turn, city strength would have to be extremely inflated. Cities would need 2000 hp to survive a single turn. Inflating everything serves no purpose.

Thanks for explaining your points!
@1: I've seen that argument a number of times and I disagree. It is based upon the premise, that you have infinite resources and can build units and therefore compose stacks as you like/as is optimal at any time. This is obviously not the case, so you will have to think about stack composition - even if you know what the optimal composition for a given scenario is, you will still be restricted by numerous factors. Balancing those does not seem like brute force.

@3: You are correct wrt city combat. At least under the assumption that the CIV5-way of city combat, with both HP and attack-strength of cities is kept (I think I'm neutral on that). However, this does not apply to battles in the field, afaics. And slowing down *all* combat, just because city-combat needs to be slowed down (in this scenario) seems suboptimal.

My thoughts with Mupt would be for the purpose of maneuverability only. It will be easier to arrange 3 siege units if they are moved in at the same time, but does not inflate combat. Tactical positioning would still be required. Brute force tactics would be minimized but possible if done in waves.

It seems to me, that that could be achieved by the model Clavente proposed in post 83 (i.e. having an "engaged" status which is 1UPT for military units participating in combat, and a "disengaged"/"packed up" status that is MUPT).

 

[Mizar]

In a game you have to do what works and not whats the most realistic.

On a strategic layer, 1upt doesn't work well.
It works on a tactical level, where you have room to maneuver and (somehow) caps for numbers, the old PG proved that on a very high standard.

So as long as you're on a highly simplistic strategic layer with most fronts being only a few tiles, 1upt kills a strategy game like this.

 

[Drowsy Emperor]

Realism is not important. There's absolutely nothing realistic about Civ as such. Mechanics first, theme integration second.

Limited stacks seem very problematic (3-4 units), and would in mid to late standard Civ IV game lead to a carpet of doom/bottle-necking, when producing 3-4 units a turn is nothing. In a Civ V game it would allow the benefit of speeding up the slow rate of production in the game, but from a tactical point of view it might take away more from the game than add to it. Hypothetically you'd have a rounded stack of bowman/swordsman/horseman/catapult or a several of a single unit. The problem is, without previous knowledge of enemy troop composition, once they get to the front lines you'd have to shuffle everything to get the best possible stack combination for each battle, making the whole thing no better than guesswork and a mess to boot.

1UPT has its logic, and for the most part its a simple one - infantry leads, ranged follows. In limitless MUPT, you can have the benefit of quite a few units in a stack to deal with most situations. So in limited MUPT you lose both the predictability of 1UPT and the flexibility of MUPT.

Really, we aren't going to get anywhere without knowing whether Firaxis considers hexes and 1UPT set in stone or not.

 

[Drowsy Emperor]

@3: You are correct wrt city combat. At least under the assumption that the CIV5-way of city combat, with both HP and attack-strength of cities is kept (I think I'm neutral on that). However, this does not apply to battles in the field, afaics. And slowing down *all* combat, just because city-combat needs to be slowed down (in this scenario) seems suboptimal.

You could solve the city problem by ditching the current system. I personally think its stupid and unexciting to have cities shoot down units out of thin air and city conquering boil down to lowering a number.

I like the Endless Legend system (not necessarily its implementation) - having cities take over hexes and expand over the map and you get to work the surrounding area of your expansion. That would let us increase map size substantially (in fact it would be necessary due to city size) to allow for 1UPT combat and increase unit movement rate (due to larger maps). Additionally, it would allow you to defend a city by forming garrisons within its hexes. A city would fall when there are only enemy units on its hexes.

 

[ariefbud]

We can make unlimited MUPT with tactical positioning by limiting only 1 attack per turn per tile.

 

[mnf]

Civ is on a very different scale. Each tile is a huge area and each move is years or even decades of time.

So it's moot to talk reality in Civ. The scale, as you say, changes over the course of one game, too.

It is wrong to rebut putting limitations on MUPT saying it's not realistic, while the application of the system in question (MUPT vs 1UPT) itself is not realistic.

MUPT vs 1UPT is purely a gameplay decision. Reality has no place in Civ. One would not be playing Civ if reality is an issue anyway, because while Civ is a strategy game, it is not a simulation.

 

[glider1]

Why do we even need tiles for this game? They are unnecessary.

Does anyone remember Rise of Nations? They had a beautiful 1UPT concept without using tiles. Armies could be sent together in a variety of interesting formations as one group that the AI was good at using, and the armies moved through complex terrain without issues. The pathing for the army was done by vectors using real numbers and because there were no tiles, the army formations were properly aligned and could be better calculated by the game engine.

That was back in 2003. Why are we still have these ancient debates about civ 1UPT/MUPT in the year 2014?

EDIT:
If anything, just put a grid around cities and leave the rest of the map gridless.

 

[Thormodr]

Why do we even need tiles for this game? They are unnecessary.

Does anyone remember Rise of Nations? They had a beautiful 1UPT concept without using tiles. Armies could be sent together in a variety of interesting formations as one group that the AI was good at using, and the armies moved through complex terrain without issues. The pathing for the army was done by vectors using real numbers and because there were no tiles, the army formations were properly aligned and could be better calculated by the game engine.

That was back in 2003. Why are we still have these ancient debates about civ 1UPT/MUPT in the year 2014?

EDIT:
If anything, just put a grid around cities and leave the rest of the map gridless.

Rise of Nations was an excellent game and Brian Reynolds did a great job. Civ II and SMAC were superb, as well. Brian Reynolds knows how to make great games.

Too bad he is doing mobile games now.

Anyway, he is very creative and I believe he was the first to implement embarking in a Civ style game, IIRC. Solved the problem of the AI not being able to launch naval invasions.

 

[PhilBowles]

Civ V is full of that as well.

Yes, that's very true. Fundamentally it's something that comes from the tension between a game like the original Civilization - which was feature-light and didn't demand 20+ cities on a map - and the feature-bloated sequels that have been made first to appeal to people who want new features, and latterly to appeal at least in part to people who want the new entries to be at least as complex as their immediate predecessors. It's like trying to add half a dozen new types of pieces to chess, but without changing the board size or the rules of movement.

That having been said, I do feel Civ V's features are handled in a more interesting way (for instance, buildings and Wonders are much more distinct in their effects than just percentage boosts to a given resource), and among other things you have fewer cities so fewer instances of spamming the same behaviour over and over.

Lots of people have maligned the game for removing penalties for most actions, but in their place the game has tried (with varying success) to make trade-offs more meaningful: duplicating buildings everywhere, or building too many units, hits your economy (certainly early in the game); social policies are in limited supply and become more expensive as time goes on, so that every selection restricts future options. Particular religious choices, spy missions or trade routes prevent you using s a different belief, or the spy or trade route being used with another target/mission until they can be reassigned, and so forth.

That balance issues and the game's general linearity have given these trade offs well-established solutions is a fault with the game, but not with the underlying design philosophy. When you make these decisions, usually they have some impact on your game progression in a way that connecting another health resource or building another duplicate unit to keep public order in Civ IV doesn't.

Half the time you're building base improvements that do nothing but raise your happiness in an effort to outpace the inevitable decline with a growing population.

But not in every city, and you don't have two resources to manage that do pretty much the same thing. Global happiness proved not to be the correct solution to the tedium of make-work micromanagement, but I think the designers correctly identified the problem with gameplay that demands repetitive, strategically useless behaviour every so often just to avoid game penalties.

Penalties should be something that accrue from mismanaging your empire - which may, indeed, include letting populations spiral out of control - or as a balancing factor to place brakes on runaways. They should not be something preprogrammed to happen at particular junctures saying "Manage me now, or else!" regardless of playstyle, game success to that point, or indeed anything else. Civ V is pretty much as guilty of failing to do this on a conceptual level as Civ IV.

Its sub-optimal to settle anywhere that doesn't have a luxury resource you don't have - its probably the most restrictive Civ game regarding city settling.

You usually want access to at least one luxury resource, but given how common they are in the landscape in practice that's not too restrictive in most cases - there won't be many spots that are otherwise prime real estate but lack a resource. Unless you're on a dual map, setting resources you already have can be as good as setting one you don't since you can trade them, and can be better with certain religious beliefs or if the resource happens to have a particularly good yield (such as salt).

And worst of all, global happiness essentially ruins the fun of war and conquest, the better you do at it the worse you're off.

To some extent Civ IV did the same, since war weariness accrued at the same rate empire-wide, and newly-conquered cities would be unhappy and often lost their happiness and health buildings during the conquest.

 

[Aeson]

So it's moot to talk reality in Civ. The scale, as you say, changes over the course of one game, too.

It is wrong to rebut putting limitations on MUPT saying it's not realistic, while the application of the system in question (MUPT vs 1UPT) itself is not realistic.

MUPT vs 1UPT is purely a gameplay decision. Reality has no place in Civ. One would not be playing Civ if reality is an issue anyway, because while Civ is a strategy game, it is not a simulation.

No, it may be irrellevent in regards to what is best for the game, but realism is certainly in favor of MUPT. Any military force you could fit entirely into one tile even on the largest of game maps.

 

[glider1]

Rise of Nations was an excellent game and Brian Reynolds did a great job. Civ II and SMAC were superb, as well. Brian Reynolds knows how to make great games.

Too bad he is doing mobile games now.

Anyway, he is very creative and I believe he was the first to implement embarking in a Civ style game, IIRC. Solved the problem of the AI not being able to launch naval invasions.

We have to remember that the Brian Reynolds of then might not be the same genius today. Ability is not constant.

That said, the solution for Civ6 is to dispense with tiles altogether, as Brian realised all those years ago. Just overlay tiles around the city if we want Civ to play the same way it always has with working the surrounding land.

Without the restriction of tiles, armies would not need to scale with everything else. When you zoom out, the army looks like just one icon. When you zoom in, you see the individual units. Armies path by vector maths using real numbers not integers.

EDIT: I hear people say "the AI won't be able to play it". Well, buy Rise of Nations on Steam and you will see that the AI can play it.

 

[Thormodr]

We have to remember that the Brian Reynolds of then might not be the same genius today. Ability is not constant.

That said, the solution for Civ6 is to dispense with tiles altogether, as Brian realised all those years ago. Just overlay tiles around the city if we want Civ to play the same way it always has with working the surrounding land.

Without the restriction of tiles, armies would not need to scale with everything else. When you zoom out, the army looks like just one icon. When you zoom in, you see the individual units. Armies path by vector maths using real numbers not integers.

EDIT: I hear people say "the AI won't be able to play it". Well, buy Rise of Nations on Steam and you will see that the AI can play it.

I still have Rise of Nations on my hard drive. The AI is pretty good, I agree.

You've definitely got some good ideas there. You should post that in the Civ VI ideas thread if you haven't already.

 

[mnf]

No, it may be irrellevent in regards to what is best for the game, but realism is certainly in favor of MUPT. Any military force you could fit entirely into one tile even on the largest of game maps.

Are you talking about absolute numbers, to say that each tile represents an area big enough to fit entire armies in?

In that case we must think carefully about what it means. It means that area itself is considered the smallest granularity for combat in Civ. Every combat action you take, is at the scale of that whole area.

Hence this actually supports the 1UPT system because one unit is the smallest granularity for military units. If the land is abstracted that much, it creates problems when you don't abstract the actors (military units) to the same scale.

In other words, we should work towards an 1UPT system that has the properties of the MUPT system---i.e. each unit is built up from smaller parts, but must move, attack and defend as one.

 

[Aeson]

Are you talking about absolute numbers, to say that each tile represents an area big enough to fit entire armies in?

I'm saying that the entire military of a civ could fit within the dimensions represented by a single tile. Not that it has to. You can find many cases in history where a specific force of military was at times more concentrated than at other times. Different situations call for different approaches.

If the land is abstracted that much, it creates problems when you don't abstract the actors (military units) to the same scale.

That doesn't follow. Neither does it mean that units aren't already abstracted at the same level. All a unit on a tile means is that there is a unit in that area. It doesn't require that unit to be the only one (unless you make a gamey rule to require it) or that that unit covers the entire area.

Civ allows many units and many tiles. Reality does the same ... the same military force can be spread out or concentrated.

In other words, we should work towards an 1UPT system that has the properties of the MUPT system---i.e. each unit is built up from smaller parts, but must move, attack and defend as one.

Combining units is just a method of management of units on the same tile. It isn't 1UPT anymore than stack movement is 1UPT. Once you get to the conclusion that more than one unit should be allowed on a tile, you've lost any reason for or claim of 1UPT. Then you're just talking about how to manage stacking.

 

[sammoraj]

Are you talking about absolute numbers, to say that each tile represents an area big enough to fit entire armies in?

In that case we must think carefully about what it means. It means that area itself is considered the smallest granularity for combat in Civ. Every combat action you take, is at the scale of that whole area.

Hence this actually supports the 1UPT system because one unit is the smallest granularity for military units. If the land is abstracted that much, it creates problems when you don't abstract the actors (military units) to the same scale.

In other words, we should work towards an 1UPT system that has the properties of the MUPT system---i.e. each unit is built up from smaller parts, but must move, attack and defend as one.

Assuming 1 unit equals 10,000 men or roughly a division(in the modern era, ancient and early eras would be less) and Stalingrad equals a city.
Fit 300k German troops into the battle, plus the Russians. If you can do that with 1UPT, let me know. Let me know if you can do that with 20 tiles. Fact is 1UPT is a failure as a strategic concept.
MUPT is needed as a straegic concept, and this is a strategic game.
We are playing a game where the entire span of human history is played. 1 settler builds an empire to rule them all. 10 units for the entire game is not empire building. Its barely village building.
This is a strategy game, not tactical. If you desperately want tactical games, go play close combat or maybe one of the cheap turn-based phone games.

 

[BlueTemplar]

I'm going to repeat what a few other people already wrote here, but I feel that this must be insisted upon :
- A lot of people seem to hate MUPT because how (supposedly) it made Civ4 (and Civ3?) just a game of "who has the biggest stack", with no real tactical considerations.
- Whereas, when you listen to experienced Civ4 players that have played multiplayer extensively, it appears that using a single stack is an easy way to get your ass kicked by a player that knows how to use collateral damage units.
- So one issue is that the Civ4 AI doesn't seem to be able to use the same tactics well enough - therefore "a single stack of death" is a viable strategy against it.
- But the Civ4 way is far from being the only way to do MUPT! People seem to forget/ignore that Alpha Centauri doesn't really have this "stack of doom issue"!
- One reason is because of the automatic collateral damage to all the stack when one of the units in the stack dies (if not in a base or bunker). This combined with ranged artillery units (and all sea combat units being also artillery) means that even the AI (which is, as usual in 4X games, generally incompetent) can severely punish you if you stack too many units in the same spot in the open.
- And if you add terrain that matters (weapon values being stronger than armor values, positioning your defenders well is crucial), complex Zone of Control features, special abilities, unit workshop, psi combat, etc... I'd say that "tactical" combat in Alpha Centauri is heads and bounds better and deeper than in any other Civilization game!
(It's a shame that 'Copters are so overpowered, so you only really get to play with the last half of the game features if you restrain yourself from winning with them - the SManiaC mod did IMHO a good job by nerfing them :
http://alphacentauri2.info/index.php?action=downloads;sa=view;down=144
- Meanwhile, 1UPT suffers from the AI being even more incompetent at it, tactics being limited due to small "battlefields", and whole game pacing being affected to reduce the "carpet of doom" issue.
- This might seem somewhat arrogant, but I get the impression that we get this whole discussion because the tactics are much more apparent in Civ5/BE than in Civ4/SMAC, so the "casuals" get to feel like real armchair generals without having to learn a lot of combat mechanics - while with only a superficial glance they fail to see the tactical possibilities of SMAC (and Civ4?). (And people that don't like warfare can almost ignore it and focus their efforts on empire building.)
- The general "dumbing down" was a big factor in attracting so many people to Civ5... while the (pre-Civ5) Civfanatics (sadly, for us) are just not Firaxis target group anymore (weirdly, BE's tech web is probably one with the most strategic possibilities / the hardest to get into among 4X games)
- It may still be possible to make a deep 1UPT 4X game, but that would be probably a 4X/Wargame hybrid (including features like bigger maps, logistics, morale...)

 

[Teproc]

Here's my view on 1UPT.

The AI will never be good at it, I've accepted that. And I'm completely fine with it. CiV is the civ game I've played the most by far (and I've played every game in the main franchise, + the first Call to Power), because of 1UPT. Combat with 1UPT is fun. It makes you feel clever. The AI may suck at it but when you're beating massively superior forces, it feels good.

In CIV, I barely ever went to war because combat was so boring. I was mostly playing at lower difficulties so it wasn't that i was more challenging, it was just completely unappealing.

Ultimately this poll shows that most people agree with me that, despite its obvious flaws, 1UPT is simply the more enjoyable system. You might say it's biased because it's in a 1UPT game forum, but it's even more biased by being on Civfanatics, which probably has a much higher concentration of people who still play CIV for dislike of 1UPT than anywhere else.

CiVI should have 1UPT, and it will. Hopefully they can make the AI better at it, but even if they don't, it beats the alternative.