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

No 1UPT

Discussion in 'CivBE - Ideas and Suggestions' started by nimling, Apr 25, 2015.

Thread Status:
Not open for further replies.
  1. Lexicus

    Lexicus Deity

    Joined:
    Aug 28, 2007
    Messages:
    23,768
    Location:
    Sovereign State of the Have-Nots
    Wrong. Stacks of Doom water down tactics, not strategy. But in my view, that's okay because Civ is (so I thought) supposed to be a game of grand strategy, not tactics.
     
  2. nimling

    nimling Prince

    Joined:
    Mar 25, 2007
    Messages:
    351
    This is a much nicer way of saying what I've been trying to say, or at least part of it. :)
    Although I'd argue that 1UPT doesn't really add significant tactics either, and 1UPT also constrains tactics in many ways too.
    I can think of MUPT implementations that do preserve tactics, or discourage stacking.
     
  3. Galgus

    Galgus Emperor

    Joined:
    Aug 22, 2012
    Messages:
    1,695
    They both tie in in terms of choosing spots with natural barriers, which stacking cheapens.
     
  4. Lexicus

    Lexicus Deity

    Joined:
    Aug 28, 2007
    Messages:
    23,768
    Location:
    Sovereign State of the Have-Nots
    I agree, I'm just telling you that's tactics, not strategy. With unlimited stacking the sole tactical considerations are very elementary things like "attack the enemy stack first with your siege units".

    Considerations like "how do I produce more units" or "with whom should I cultivate an alliance to avoid getting into a war I can't handle" become far more important than any tactical considerations.

    I don't see how you work this out. 1upt clearly allows for more involved tactics than SoD. As I've said previously, though, I am fine with stacking in a civ game. I prefer SoD to 1upt, not generally but in the specific context of a game of grand strategy like Civilization.

    If 1upt is absolutely necessary then I believe battles should play out as a hex-based wargame in a separate tactical layer. This would of course warp the Civ series out of all recognition but arguably the implementation of 1upt has already done so.
     
  5. Roxlimn

    Roxlimn Deity

    Joined:
    Feb 11, 2005
    Messages:
    3,526
    Not at all. The very Civvish Master of Magic involved a Civ-like strategic layer and a tactical sublayer very similar to Civ's 1UPT. In fact, within the sublayer, units like Mages had incredible range, and you could have ranged units of various ranges and strengths. Endless Legend also uses a tactical sublayer based on the overland map. GalCiv 3 has "hexes in hexes" - large strategic hexes directly zoomable into smaller hexes in the hexes for tactical fleet maneuver. Sid Meier's very own Starships has a very simple strategic layer and a variety of tactical maps for tile play.

    I think it can be plausible to design a pregenerated overmap with simple broad characteristics, and then assign a specific type of tactical map to each overland hex, which can be previewed. This way, you can build fortifications that will matter on the tactical layer, but some tactical maps are better or worse for defensive purposes. Cities themselves can exist on the tactical sublayer as a ranged unit without having them capable of killing units two tiles over.

    Players who prefer strategic play might opt to "auto-resolve" the tactical sublayer combat (a feature in MoM and GalCiv) so as to not play the 1UPT layer at all, if they didn't want to.
     
  6. Lexicus

    Lexicus Deity

    Joined:
    Aug 28, 2007
    Messages:
    23,768
    Location:
    Sovereign State of the Have-Nots
    Okay, it would twist my idea of what the Civ series should be totally out of shape ;) I agree with all else you're saying. As a big fan of the Total War series (at least up to Medieval II) I'm aware of many of the possibilities for a tactical layer.
     
  7. Galgus

    Galgus Emperor

    Joined:
    Aug 22, 2012
    Messages:
    1,695
    Tactics tie into strategy in city placement and from what angle one chooses to attack.

    Let's be honest, there isn't much strategic depth in spamming units. Aside that I'd generally like to see more variety in units for the tactics game.

    I'd like diplomacy to be more of a thing in warfare, which doesn't seem any more likely than making the AI better at 1UPT.

    If the AI is just terrible in rough terrain the best option may be to alter map generators to be more AI friendly.
    _______________________________

    I'd be fine with wars taking place on sub-hexes so that war tiles could be broader than city tiles.

    I love Rome: Total War with its duel layer system, though I doubt Civ would go the quasi real time combat layer route.

    Off topic I've enjoyed the beta of the Fourth Age mod for that game, taking place many years after the Lord of the Rings books.
     
  8. Roxlimn

    Roxlimn Deity

    Joined:
    Feb 11, 2005
    Messages:
    3,526
    First: AI will never be better than a human if that code has to run on a variety of machines well, including dinky little notebooks, requires adaptation to ever-changing rulesets, and has to process multiple turns in under a minute. Frankly, I'm surprised the Civ AI isn't dumber than it is.

    Second: The most glaring problem with 1UPT isn't 1UPT. It's the same problem in SoD, but it isn't as obvious because stack combat is just idiotic. It's lack of simultaneity. Imagine a ruleset in which one side in chess gets to move all their pieces once, and then the other side gets to move all their pieces once and so on. The game would be horrifically unbalanced and impossible to balance.

    The solution is to either boil down everything to essentially a one unit vs one unit concept or stagger turns so that each side gets to use one unit before turning over control. That will make it harder to FF a single unit into oblivion without consequence.
     
  9. joncnunn

    joncnunn Senior Java Wizard Moderator

    Joined:
    Mar 17, 2008
    Messages:
    8,621
    Location:
    Missouri
    That is actually tribal to show that with optimum play its a tie.

    First player advances all pawns by 2 squares. (And does whatever he wants with non pawns.)

    Second player advances all pawns by 1 square. (And does whatever he wants with non pawns.)

    All future moves by both sides involve pieces behind the lines with everybody keeping their pieces behind it (nobody will enter "no mans land") until the game ends in a draw with no pieces taken by either side.
    First player has one row more territory, and while to an in progress game its a factor used as a heuristic, all ties are recorded equally.
     
  10. Roxlimn

    Roxlimn Deity

    Joined:
    Feb 11, 2005
    Messages:
    3,526
    Simultaneous piece moves means that an opposing black player can arrange his pieces to threaten an instant loss for white if he captures an opposing pawn. This will allow black to advance and attack the white line, especially when they're all unsupported! I think you don't understand chess very well if you think that a line of pawns all abreast makes any kind of defensive sense.
     
  11. nimling

    nimling Prince

    Joined:
    Mar 25, 2007
    Messages:
    351
    I don't know why people are bringing up chess, it's a very different game with a fixed board.

    Maybe one of these days I should just build a mockup to show what I mean, rather than try to explain myself in a lengthy post and receive snide remarks.
     
  12. Roxlimn

    Roxlimn Deity

    Joined:
    Feb 11, 2005
    Messages:
    3,526
    I think, nimling, that what you mean is basically trumped and trampled over by the many very functional, essentially 1UPT games that already exist and are pretty good. They stand fairly strongly on the idea that 1UPT is a workable mechanic, if not very well implemented in Civ V.
     
  13. nimling

    nimling Prince

    Joined:
    Mar 25, 2007
    Messages:
    351
    Some are functional, but they often use 1UPT out of necessity, or because their focus is narrower (for instance, a WW2 scenario with fixed unit types might do fine with 1UPT). Most often those games succeed not because 1UPT is preferable, but because it is not too much of a liability.
    Using the board game example - if a board game player could track tens or hundreds of units without making the game incredibly tedious, and playing pieces were versatile enough, would it be a major issue? Board games necessarily need to abstract a lot to prevent the game from bogging down with too many details, hence why 1UPT is usually preferred in such a game (and even then not universally).

    The state of Civ5 and BE though shows why 1UPT doesn't work, especially in multiplayer - and when other games in the series didn't have the problems presented by 1UPT, for what are similar enough games, I don't know why people want to fix what wasn't broken.

    Master of Magic used a 9UPT system, but this also led to some balance issues that could have been prevented; and part of that was because the game used a tactical combat sublayer, and inserting 100 spearmen into a fight would be incredibly unwieldly without automating and abstracting a lot of details.

    I could see a tactical sublayer of sorts working in Civilization, though to keep the game moving fast, it should be limited to automated battles (which was an option in MoM and MoO). This could possibly set up situations where a bigger army doesn't always win by statistical analysis, if you can simulate a catastrophic tactical <snip> like the Battle of Cannae in tactical mode rather than having say a 1% roll to <snip> up.

    Another pet peeve of mine is healing units for free, which trivializes any combat that doesn't destroy units outright. The necessary changes to fix that go beyond my original scope, but would definitely be something to consider.

    Moderator Action: Please do try to avoid the autocensor. Offending words snipped.
    Please read the forum rules: http://forums.civfanatics.com/showthread.php?t=422889
     
  14. Velasti

    Velasti Warlord

    Joined:
    Sep 26, 2014
    Messages:
    103
    Abstracting combat actually really annoys me. Starships is fun, so is Star Wars Empire at War... but the abstract combat is annoying.

    1UPT also really annoys me, but that is because the AI is really useless.

    I like Fins's idea.

    On the other hand, there are some mods that correct issues with the AI to some extent (e.g. BNW Smart AI) There are other times when the AI can dominate with its carpets of doom (I don't mind carpets of doom... but the logistical issues are a pain in the neck, and I wish I could just combine 6 units into an army and move the army. I.e. the AI sorts out the pathing and keeps the army together)

    The comments about AI due to performance aren't correct, download firetuner and play the game on fast forward. The AI plays at speed. The game is "slow" because of graphics, which are guaranteed to be much more computationally intensive than anything that the AI can do. The AI doesn't do as much maths. The AI's turns in normal play take a long time to play out because of the animations. I also suggest that the animations during the turn and the delays are deliberately slowed down by the developers so as not to make the player feel bad (imagine clicking end turn and have the AI complete its turn immediately without "thinking")

    For anyone who doesn't understand programming, computers do maths incredibly fast. A 1GHZ CPU makes 1 000 000 000 calculations per second. The AI won't be that resource intensive.


    I think Fin's idea should be simplified though - just make it 4UPT, as opposed to 1UPT, 4 TPT. You can keep the existing damage calculations no problem.

    The AI will still have issues with rough terrain (but I hate fighting over rough terrain too) I sincerely doubt that the AI has any kind of heuristic related to identifying chokepoints and rough terrain whatsoever. The AI needs improvements in those areas anyway.
     
  15. nimling

    nimling Prince

    Joined:
    Mar 25, 2007
    Messages:
    351
    AI routines can be surprisingly intensive. Things that seem easy to a human brain are quite complex, like identifying chokepoints and understanding just shy embarking units is often a bad idea. (1UPT doesn't help due to unit jamming, but it wouldn't do for the AI to basically never embark units since it is useful in many spots.)

    Teaching an AI to play 1UPT without too many mistakes can happen - BNW Smart AI is a vast improvement over Vanilla - but part of the problem is that 1UPT has inherent flaws, even for a human player.

    So far we have a few options...

    MUPT, single unit combat, ranged units no longer exist aside from air units and missiles - i.e. similar to Civ 4, but with few anti-stack options unless those are hacked in. Nukes are obviously devastating to stacks.

    limited UPT, single unit combat, ranged units may or may not exist - some of the models proposed here. Still has the problem of imposing an arbitrary limtiation and some drone-jamming, but is better than 1UPT,

    MUPT, single combat but stack damage when a unit is defeated outside a fort or city - this is a variant on the system used in Alpha Centauri. May include ranged units like AC's artillery, which are pretty effect stack damagers.

    In any event, the absurdity of archers that can hit over oceans is a major balance error in Civ5, and that error extends to BE with Rangers; however, Rangers aren't as unbalancing as Archers due to melee in BE actually meaning a damn, whereas ranged units in Civ5 stomp most melee units and have brokenly powerful units like chariot archers with no real counter.
    Siege in Civ5 is only weak because there is no real reason not to use overpowered archer units to take cities / function as field artillery instead. If ranged combat were limited to the level siege units function at, it would at least be less infuriating. Siege units also show how 1UPT is incredibly obnoxious.

    Another gripe with Civ5 that BE thankfully reversed - Naval warfare. I had a bad feeling Rising Tide will make the same bad decision Civ5 did to allow melee naval units to capture cities.
     
  16. Velasti

    Velasti Warlord

    Joined:
    Sep 26, 2014
    Messages:
    103
    Yes they can be... but it isn't an issue in Civ, and playing the game through firetuner makes that clear, the AI calculates its turns really quickly.

    Identifying chokepoints isn't that big an issue - it's influence maps, which have been around in strategy games forever.

    How much do you know about cognitive science? The human brain doesn't think rationally, it uses heuristics. You program computers to think in a similar way and also apply heuristics (rules of thumb). You're correct, things that seem easy to humans which are complex are easy because your brain figures out ways to simplify the process. They're still complex,you're just programmed to make things into habits and routines so you don't have to waste energy.

    Besides which, they can always just set up 50 games in a lab somewhere with machine learning algorithms and run combat scenarios over a weekend. There's evolutionary algorithms for that kind of thing. Just teach the AI to write itself, and let it do so.

    https://www.kickstarter.com/projects/jonshafer/jon-shafers-at-the-gates/posts?page=7
    The issue is really map size, as Jon Schafer identifies. I think that stacks should not be infinite, there should be a logistical limit to how many units you can put on one tile. It's a neat way of correcting the issue. You make it army vs army, but you limit the sizes of the armies, so that it doesn't become ridiculous.

    There's no issue with units having ranged attacks. The problem there is that archers have mobility and siege units don't, and mobility is always better than extra damage. I remember there's an RTS game called Praetorians, and your archers are absolutely devastating, your siege units, not so much. But the advantage your siege units have is that they have more range than your archers, so they can attack a position where archers are entrenched. To that end, increasing the relative range of siege units (and cities) wouldn't be that bad a deal. I quite like the ranged units in Civ V in the gunpowder era, that make their ranged attack just one tile over.
    The AI can't capture cities without melee naval units though? (Then, the AI can't capture cities, full stop) I'm really excited about bringing melee naval units back, it means I'll actually have to bother paying attention to Naval.
     
  17. nimling

    nimling Prince

    Joined:
    Mar 25, 2007
    Messages:
    351
    I'm not speaking about the AI, but humans playing against other humans, and looking strictly at how 1UPT and Civ5 mechanics fail. Playing enough MP games should show anyone with a sense of good game design what I mean; and that's also why the state of Civ5 MP is so bad, because decent gamers want no part of the behavior on display. Bad design generally leads to bad behavior.

    I don't have a lot of MP experience in BE but what I did play was slightly more tolerable - largely because gunner-line units get stomped by melee, unlike CivV compbows and xbows. It's still pretty bad and BE has its' own flaws, but it does show that a great deal of the problem is that ranged combat itself is a large reason why MP winds up sucking so much.

    The Naval units capping cities thing is a bad idea as well, since it trivializes amphibious assaults and makes any coastal city (and especially capitals) too risky to even consider without an ultra-secure harbor. Ranged units being broken is the largest part of that equation (frigate / gunboat spam), but a melee unit that can swoop in from 7 tiles out or whatever, that just sucks.

    I don't know why the game doesn't penalized ranged units vs. cities. Vanilla did it to a small degree, but it didn't nerf archers vs. cities enough. It would be better if archer spam wasn't such a huge factor in any game, and the way to start doing that is to do away with 1UPT.
     
  18. Velasti

    Velasti Warlord

    Joined:
    Sep 26, 2014
    Messages:
    103
    I don't really see that the Civ series are a good example of PVP games to be honest. Civ tactics fail because Civ tactics fail. I don't really see how you could fix the game, because there are aspects of unit movement in Civ which make it inherently damaged beyond repair. I think I agree with the idea that it would work if maps were bigger and had less vicious terrain.

    Well, yes. In the real world, you would have to secure a harbour using military force. I don't think think that melee vessels trivialise amphibious assaults - you have a dedicated amphibious unit conquering a city, much as would actually happen in the real world (remember we don't even know what BE amphibious warfare looks like yet)

    Part of your issue is that melee units shouldn't be swooping in from 7 tiles out. 7 tiles from your city really isn't that far over sea and ocean, because there are no movement penalties or terrain factors. Part of the issue from a game mechanic perspective is the lack of visibility over the sea. 7 tiles is only one turn of movement.

    However, an amphibious naval fleet parked off New York - What's going to happen? In reality, the aircraft carriers and destroyers and what have you would bombard the city from afar with missiles to knock out the infrastructure and defences and so on, and only then would the amphibious landing force leave the relative safety of open ocean to close in for the kill. Why would you be allowing this to happen?

    The AI trivialises combat, so anything that allows for the AI to actually be a combatant is excellent.

    Your third paragraph seems like a circular argument, because you acknowledge that Beyond Earth does melee vs ranged well.
     
  19. nimling

    nimling Prince

    Joined:
    Mar 25, 2007
    Messages:
    351
    BE does ranged combat better than Civ5, which is like saying Iceland is warmer than Greenland.
    If you play MP Civ5 and get Frigate rushed you should know what I mean. It sucks and with simultaneous turns it's a bloody nightmare to fight naval wars. Even with sequential turns it's an exercise in frustration.

    In the same way, most of the "tactics" from 1UPT are in fact added micromanagement, rather than strategic choices. The only strategic choice is to spam ranged, and domestically the only strategic choice is to max out science due to the lack of tradeoffs (in BE and Civ5). The rest is just a matter of will to engage in clickfests, and in MP play a really bad RTS due to commonly accepted rules.

    The other alternative is to do away with units and tiles altogether, which is what i suggested in OP.
     
  20. Velasti

    Velasti Warlord

    Joined:
    Sep 26, 2014
    Messages:
    103
    I like little improvements. One step, then another, now you're running.

    I don't think designers know what tactics and strategy are to be honest - you're right, the game is a micromanagement simulator. There is very little strategy, and very little choice, in that learning the best techniques reveals which choices are strategic and which are naive.

    But this is a common theme in many strategy games, that there isn't sufficient strategic depth.

    And yes, but most of the "strategy" and "tactics" in strategy games are, in fact, just micromanagement and will to engage in clickfests. . . it's why MOBA's are popular. I wish they'd automate it all away and add some depth and complexity, other than just 1. Learn the game mechanics
    2. ????
    3. Profit.

    I think we only differ in that I don't think it's 1UPT that is the cause of this, but more design decisions permeating every level of the game. E.g. clickfest combat is due to interface and logistics problems.
     
Thread Status:
Not open for further replies.

Share This Page