1. We have added a Gift Upgrades feature that allows you to gift an account upgrade to another member, just in time for the holiday season. You can see the gift option when going to the Account Upgrades screen, or on any user profile screen.
    Dismiss Notice

1UPT - final verdict?

Discussion in 'CivBE - General Discussions' started by santoo, Oct 27, 2014.

?

One unit per tile (1UPT) or multiple units per tile (MUPT)?

Poll closed Nov 10, 2014.
  1. I started out with 1UPT (e.g. CIV5) and prefer 1UPT

    44 vote(s)
    10.0%
  2. I started out with 1UPT (e.g. CIV5) and prefer MUPT

    6 vote(s)
    1.4%
  3. I stated out with MUPT (e.g. SMAC) and prefer 1UPT

    244 vote(s)
    55.2%
  4. I stated out with MUPT (e.g. SMAC) and prefer MUPT

    148 vote(s)
    33.5%
  1. Decamper

    Decamper ..!

    Joined:
    Jul 28, 2014
    Messages:
    1,157
    I've only seen one person in this thread (Talcove), as far as I can tell, who has brought up multiplayer. 1UPT really shines in player vs player wars, and I've seen lots of great saves and manoeuvres that never happened for me in the the past MUPT games, where it generally seemed to just boil down to the biggest stack rolling over or chasing off any smaller stacks.

    So, given that I love multiplayer, I side with 1UPT. Of course, all the AI civs in multiplayer games are just fodder that preoccupies the players to stave off the great bloodletting that the endgame turns into.

    Edit: SMAC was were I jumped on the civ wagon.
     
  2. Perfxion

    Perfxion Warlord

    Joined:
    Oct 22, 2005
    Messages:
    168
    Location:
    Goldcoast, USA
    The problem isn't so much stack of doom vs 1UPT, it should be combat only versus any other win condition. Problem with Stacks of doom is that war was the only major win condition in Civ4 and Civ3. Having to churn out units in mass. If someone wanted to go tall and build major large cities, it was only done playing off easier levels.

    I would be okay with mix units, but has to be of some logical use. I am talking out of my rear end but may what if you can add 3 units to a tile, but after that the more units you use, the larger the area they take up so say 8 units takes off 3 tiles wide. Thus coming to a choke point, would have to break up stack, but can merge again in open field. So each unit can have an area rating. Melee units being the smallest, tanks being the largest. And the more you stack, the wider the needed room. Thus a city with a choke point in front has a terrain defensive advantage but the open field allows an easier counter attack.
     
  3. Calavente

    Calavente Richard's voice

    Joined:
    Jun 4, 2006
    Messages:
    2,794
    Location:
    France
    I'm on the MUPT side.
    I'm not in disagreement over 1upt.. if we only think about "actively militarily engaged units".
    however, for "moving units" or civilians, you need MUPT (if only to represent real warfare).
    Spoiler first stupid idea starting from 1upt :
    you could have civilians as MUPT;
    then military units being in an engage or disengaged state.
    Disengaged change their movement: movement is moved at 2 (for all units), reduce their defensive str / health by 2-3-4, and attack str by 5-10-20, can move through another unit, engaged or disengaged.
    Engaged give you back the unit as normal : 1upt, but can pass over disengaged units (at double movement cost)

    passing from engaged to disengaged takes a whole turn, need full movement points.
    Passing from disengaged to engaged ends your turn for this unit.

    Units in friendly territory stay in disengaged mode, otherwise they create negative influence points with friendly units
    ..Etc
    there are surely plenty of improvement possible, but you could thus move your units to the front lines, while still have frontlines.

    maybe "disengaged" can only be activated/kept inside own territory... or within 1tile of an engaged unit ?
    (so that you need to keep a "supply route" in wilderlands/ennemy territory ?)

    another idea: give units the possibility to join another unit : give some stats : +0.5str, +5withdrawal...etc depending on the unit stat...)

    2nd
    Spoiler how to boost terrrain effectiveness starting from 1upt :
    increase terrain boni/mali : forest give +50% defense; forester give +50%forest attack, hills give +20%defense, +20%defense to archer, +20%attack to archery... hills and forest give "ambush" : +50%attack to "no-hill/no-forest" tile
    (or 20% instead of 50%).
    increase movement by at least +1: so those pesky hills/forest that take double/triple movement matter to infantry.
    give "support factors" to nearby units (as in 1upt)
    give zoc (as in 1upt)
    give health to units so combat doesn't last 1 turn only (see civV)
    give "supply costs per unit in stack" applied immediatly, and un-moving during ennemy turn, even if you get some units that die: str = str*(n/(n+1)) (but that one might be too big a nerf): the more units in stack, the less they are individually powerful) or str = str (1- (n-1)%) : a 10 units stack get +10%str on units, effectively negating that combat 1 promotion in civ4.

    thus you'll get no "1-upt" issues for movement, but stacks would be less interesting, and using terrain will also be really important.

    However Ai might be a little dumb, but not more than for normal stack-using AI nor normal UPT.

     
  4. vandyr

    vandyr Prince

    Joined:
    Sep 5, 2007
    Messages:
    351
    Started with Civ 2, played every version since, and I much prefer 1UPT, to the point that I wouldn't play any version of Civ with stacks anymore. SoD is an antiquated, boring system that never needs to see the light of day again, imo.
     
  5. TomSawyer

    TomSawyer Chieftain

    Joined:
    Nov 5, 2005
    Messages:
    48
    Been playing since Civ 1.

    The more I chew on the 1 unit vs the stack o doom vs the army system the more I wonder if any of the programers will have a Sid moment and create something new and just totally mind blowing.

    /me waits patiently for Civ6
     
  6. Lieu

    Lieu Chieftain

    Joined:
    Aug 4, 2009
    Messages:
    48
    I'll argue the opposing side to this - I'm maybe one of the few people who will :p Tactical AI in civ is not an insurmountable problem. It is a matter of implementation, not an inherent impossibility in the game itself. As a side project I made a chess-style tactical AI for civ and it works. Incomplete, but at the core job it works. It does so for the same reasons computers can play chess: It doesn't brute force all possible outcomes, for the game tree is far too big. It wouldn't stand a chance. Computers can play chess because they use the same important heuristics we do. Yes, computers can search millions of positions per second but it means nothing without direction. If you had a trillion computers running for the current age of our universe it would still take about a million times longer to have a hope of reaching each position once. That is brute force. As it stands, computers search many more positions than we can but have lower quality (but fast) evaluation of how good each one is - at least compared to the top humans.

    It's all about the heuristics, although we rarely realise which ones we are using ourselves. Good tactical AI is possible in civ though.
     
  7. SammyKhalifa

    SammyKhalifa Deity

    Joined:
    Sep 18, 2003
    Messages:
    5,175
    Somehow I doubt this is going to be a Final Verdict.

    :D
     
  8. Becephalus

    Becephalus King

    Joined:
    Nov 30, 2005
    Messages:
    725
    1UPT is fine and I prefer it. They need to figure some things out though IMO.

    Units need to be fewer and more expensive. Have the units themselves represent actual armies. Ditch ranged units they make the AI unnecessarily terrible.

    These changes would make the game more realistic and more challenging (while perhaps more boring).
     
  9. SammyKhalifa

    SammyKhalifa Deity

    Joined:
    Sep 18, 2003
    Messages:
    5,175
    Don't they already?
     
  10. Lucius_

    Lucius_ King

    Joined:
    Dec 3, 2012
    Messages:
    806
    Interesting poll. Was it the expected result? (didn't read the thread so sorry if that has been answered.)

    I started with Civ I in 2007 (abandonware.) I then played a couple games of II and bought C3C. I played hundreds of hours of III. Never played 4 because my PC couldn't play it. Years later I played V and all exps.

    I prefer 1UPT. It adds tactical depth that the earlier Civs lacked. It's clear 1UPT is here to stay (based on the overwhelming support this poll shows), so it's up to Firaxis to get the AI up to working with it.

    It surprises me though. I've seen the AI make some very competent assaults and I've seen them make assaults that have been entirely flawed.
     
  11. Cromagnus

    Cromagnus Deity

    Joined:
    Sep 11, 2012
    Messages:
    2,272
    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.

    Compare that to having 20+ pieces you can move per turn on a board 5 times as big. Mathematically speaking, we just went from number of grains of sand in a sandbox to the number of atoms in the universe.

    An AI that was effective at troop movements, tacics and strategy in Civ 5 would require IMMENSE computing power. However, I'm not saying it's impossible. I'm just saying your desktop can't/won't handle it any time soon...

    However! If Civ 6 was implemented with a cloud AI backend, with hundreds of high-powered servers collaborating, then perhaps we'd see some truly intelligent AI.

    Of course, that assumes one other key factor. It would take a team of very savvy AI programmers to deliver on said cloud-based system. The hardest part of all would be developing and tuning the heuristics.

    All of this would cost more than Civ 6 is likely to earn in revenues. Translation: It'll never happen.

    But, I'd love to be proven wrong. :)

    IMHO, the only "intelligent" AI you're likely to see is one that cheats very cleverly.

    For example, let's say the AI has 10 units. All out of position, but if the *player doesn't know that*, teleporting units just inside fog of war is an incredibly effective means of simulating intelligent troop movements. Using this *in moderation* based on difficulty level and with some randomness thrown in, the AI could simulate military strategy. That is doable. And it's likely the only thing that's doable. The essence of good AI is cheating, and always has been.
     
  12. Pepo

    Pepo Prince

    Joined:
    Nov 27, 2012
    Messages:
    315
    SoD is much better than 1upt as it works while civ V and BE combat ai simply sucks. I want Mupt to return, but with changes like attricion like paradoz games. It would make much better work at simulating combat than unrestictive Mupt or 1 upt
     
  13. Strategist83

    Strategist83 King

    Joined:
    Dec 30, 2005
    Messages:
    625
    1UPT is superior to stacks. It offers better possibilities for depth of combat tactics. People dislike 1UPT for the wrong reasons. It isn't that the AI can't use 1UPT properly because of 1UPT, rather that the AI can't use it properly because the AI is sorely lacking. In other words, the problem is with Firaxis, not with 1UPT. They simply haven't executed it properly because they haven't committed sufficient resources to making it work. They really haven't tried. I don't buy the excuse that "creating an AI for a 1UPT game is too difficult." That doesn't match what we're seeing.

    To support this claim I simply present you Civ:BE. A game that has new ideas but at its base, the mechanical level, is really just Civ V again. Firaxis hasn't worked to correct any of the problems with the Civ V AI so they have simply carried over and are thus present in BE also. They have not addressed any of the AI issues V was criticized for with their new 50€ game. The game is completely unchanged in that regard.

    Should Civ VI go back to stacks or something simpler than 1UPT? That depends on Firaxis. If they want to actually try to create a decent AI, there's no need to return to stacks. But, they precisely don't appear to want to try.
     
  14. Kolbeg

    Kolbeg Chieftain

    Joined:
    May 10, 2011
    Messages:
    66
    1UPT is better hands down, more strategic combat. I´ve played since Civ1.
     
  15. joncnunn

    joncnunn Senior Java Wizard Moderator

    Joined:
    Mar 17, 2008
    Messages:
    8,621
    Location:
    Missouri
    The AI isn't good enough to properly use 1 UPT, which is why I voted MPT.

    A comparison of other games though shows you don't need unlimited stack per tile for the AI, most of the AI problems actually go away at 2 UPT.
    (Both the traffic jam problem and the keep a heavy ranged unit with weak melee protected by a strong melee unit are solved AI problems at 2 UPT)

    Now in an all human game with no AIs then I'd prefer 1 UPT.
     
  16. Drowsy Emperor

    Drowsy Emperor Warlord

    Joined:
    Oct 27, 2014
    Messages:
    102
    @Lieu, so what is your AI capable of doing?

    The key thing to remember here, besides the discussion of whether it can or cannot be done is what game developers aren't likely to say and that is, that AI programming is an afterthought.
    If it gets beyond workable (and sometimes not even that), then as long as they're concerned its fine.
    Its not one of the things that's immediately apparent and that will sell your product, like, graphics. Even though companies have been known to brag about their AI even when its junk (Oblivions, "radiant" AI - outdone by 20 years older Squaresoft RPGs), AI is essentially left to the personal initiative of the programmer for anything beyond "meh".

    So sometimes you get guys like the one who made bots for Unreal Tournament, who really worked hard at it and knew what he was doing.

    99% of the time you get garbage, which is amusing as AI programmers are really sought after and from what I read, paid rather well.

    Now the upsetting thing here is that with Civ BE they made no apparent effort to fix up the Civ V AI. I've read that people went through the code and saw that it was basically unchanged.
    What that really tells you is that they aren't interested in fixing Civ V mechanics, which is an implicit admission that its either not fixable or not worth fixing in its current state.
    Either way, its a bad move to ignore a known problem, take the cash and dump it on the players again.

    This is why I don't have much faith in Firaxis to do the right thing, regardless of which system they use in the next Civ. Without a determined and competent designer at the helm all we're going to get is another product that's "good enough", in AI and everything else.

    And being "good enough" is not what Civ is about. At least it wasn't.
     
  17. Krajzen

    Krajzen Deity

    Joined:
    Oct 23, 2013
    Messages:
    2,555
    Location:
    Poland
    My problem is, I don't like both MUPT and 1UPT :D


    MUPT - hard to say why but I hate it in turn - based strategies, really really hate it, everything IMHO becomes "amass bigger stack and punch enemy smaller stack, then you won the war and don't have to worry anymore as nothing can stand against it!"

    1UPT - prefer it over MUPT but honestly I am also tired by it, I mean battles are cool but this is a nightmare in terms of logistics. Every time I have to command my 632282 units for extremely awful terrain - and there are lot of awful terrains in Civ5, narrow jungle passages with lakes and rivers etc. - my enjoyment falls.


    Honestly, currently I would love to see Endless Legend's solution in Civ:

    - Armies walk in stacks

    - When two stacks meet they fight

    - But battle arena is not some abstract - on - another - map arena but armies fight on the World Map tiles!

    - Thus, logistical problems of 1UPT are solved while battles are awesome.


    I think this mixed with some neat general system and HoMM 'many soldiers on one battle tile' solutions would be the best.

    Moderator Action: Please stop using the wall of bold text. Removed the bold.
     
  18. Becephalus

    Becephalus King

    Joined:
    Nov 30, 2005
    Messages:
    725
    Um no, they are sort of cutesey arcade like abstractions with odd traits. No one builds and army of tanks. An army has tanks and men and air support.


    I also think it is always important to remember that the Civ 4 AI didn't really understand or use air units until an unofficial fan patch after the last expansion. There was no glorious past where the AI was actually good. All sticks did was make sure if you were totally and completely out classed you could not trick your way out of it. That is much easier to do with 1UPT. But the AI was also terrible at MUT.
     
  19. Lord Tirian

    Lord Tirian Erratic Poster

    Joined:
    Nov 30, 2007
    Messages:
    2,724
    Location:
    Liverpool, UK
    Slightly off-topic:

    I really need to check out Endless Legend, I think... did you play Endless Space? If so, how does it compare, atmosphere-wise?

    I found Endless Space competent but utterly soulless (much worse than Civ:BE's blandness), so I held off Endless Legend - impressions regarding that?
     
  20. santoo

    santoo Chieftain

    Joined:
    Apr 1, 2007
    Messages:
    83
    interesting, I hadn't thought of that. Should have included SP vs MP in the poll. Out of curiosity: are there any stats, which percentage of games are MP vs. SP or which percentage of players MP or SPonly? Any guesstimates?
    I can certainly see the advantage of 1UPT in MP - but, considering that civ already takes enormous amounts of time - isn't it even worse, when every battle involves a lot of tactical thinking/planning? Are we talking mostly asynchronous play (ie PBEM)?

    :D I meant "final" in the sense that 1UPT now has had a couple of expansions/patches/years to prove itself. comparing MUPT from a fully fleshed out CIV4 to 1UPT from unpatched vanilla CIV5 was never going to be fair, imho

    short/simplified answer: not at all. expected: players from MUPT-area preferring MUPT, players from 1UPT preferring 1UPT. expected ~50:50 split or maybe majority MUPT, based on reading threads.

    I really like that concept. I'm trying hard to remember where I know it from - I'm sure some game from the 90s(?) had something like that, at least for mechanical units IIRC. I'd love to see that implemented.

    I agree that EL and HoMM offer much better tactical combat, but: if we had the same thing in CIV, wouldn't we spend 80% of the time fighting? Not saying that's generally a bad thing - I enjoy HoMM - but would CIV benefit from being 80% fighting?
     

Share This Page