1 unit per hex: failed experiment

Discussion in 'Civ5 - General Discussions' started by ohioastronomy, Oct 3, 2010.

  1. QES

    QES Court Jester

    Joined:
    Jul 18, 2006
    Messages:
    2,139
    Location:
    Minnesota USA
  2. Xiao Xiong

    Xiao Xiong Prince

    Joined:
    Oct 15, 2009
    Messages:
    480
    I don't see why. Essentially it's just two changes plus associated tweaks to other elements:

    1. The scale of the map is actually bigger than it appears to be in most views 7x bigger, from the combat system point of view

    2. Unit movement in a turn is in groups of up to 7 and from one "combined" tile set to another.

    Underneath that's not such a big change. Most of the work would be in the UI.
     
  3. dexters

    dexters Gods & Emperors Supporter

    Joined:
    Apr 23, 2003
    Messages:
    4,182
    Gender:
    Male
    Location:
    Canada
    I think the most realistic solution is the relaxation of 1UPT rules for
    a) either AI only so they can move the required units around
    b) for both AI/human for certain units (ie: ranged can be stacked with 1 escort unit).

    That will still require a significant AI rewrite,but that's one area where as a big supporter of Civ5, I EXPECT, Firaxis will work on, if they want the $100 to $150 outlay I will be spending in the next 2-3 years buying their expansion packs and DLC.
     
  4. hewhoknowsall

    hewhoknowsall Warlord

    Joined:
    Nov 30, 2009
    Messages:
    294

    At the risk of sounding rude, this is rather blatantly hypocritical. One of the main problems with the 1UPT is the blatant and absurd unrealism and problems with scale. Real life archers cannot shoot across mountains or outrange modern infantry, real life task forces do not fight hundreds of kilometers away from eachother, real life longbowmen do not outrange artillery, etc. A pro 1UPT supporter will often say that realism in a game is secondary to gameplay as a rebuttal to this, which is a good point, but to try and argue for 1UPT on the basis of realism is rather contradictory.
     
  5. ash88

    ash88 Hail to the King Baby -DN

    Joined:
    Jun 26, 2009
    Messages:
    352
    Location:
    Castle Merlot
    Well thought out post that I appreciate.

    As a matter of subjective preference I prefer the problems that the CiV no stacking creates over the problems that Stacking created in Civ4. Both have problems, and you highlighted the good ones.

    As for your points about the AI - we both agree it is virtually non existant in release Civ 5. As I understand your post you have no hope that an AI can be programmed to account for the single unit limits and the "exploits" that the limit imposes. I can't fault you for this view, and I may soon share it. For now I remain optimistic though; I'm optimistic that they will improve the AI to the point where it recognizes and corrects for the issue.

    Some fundamental things could be added to further correct this - like attacks of opportunity (like you describe an entrenched artillery being able to react and shoot first instead of having to wait till all the damage is done). These would be awesome. Maybe an expansion to deal with that.

    But for now I prefer the single unit combat exploits versus the stacks of doom. I respect that you prefer the latter.
     
  6. tom2050

    tom2050 Deity

    Joined:
    Oct 12, 2005
    Messages:
    5,516
    Perhaps the 1upt with combining units to form 1 unit is the best. As was mentioned earlier, making them armies.
     
  7. Maxor127

    Maxor127 Warlord

    Joined:
    May 28, 2003
    Messages:
    229
    You ignore the obvious problem with a stack like that, which is a units attacking over and over from the same spot. So no, that's a terrible idea. The solution is to be able to select multiple units and move them in a formation like in an RTS.
     
  8. Demon Rob

    Demon Rob Chieftain

    Joined:
    Sep 8, 2005
    Messages:
    42
    Location:
    Melbourne, Oz
    Unfortunately its not the correct scale for formations of this type to work. Civ IS at the army level, not the unit level.

    Its all so dissapointing.
     
  9. Seanmac

    Seanmac Chieftain

    Joined:
    Feb 17, 2004
    Messages:
    7
    The simplest solution for cleaning up that mess is to impose realistic supply systems. I shouldn't say it's simple--it would be complex to model the difference in supply models for ancient armies and modern ones, for instance--but it would help impose some realistic limits on army size and on army movement.

    Honestly, coming from a wargaming perspective, the combat in Civ has always been so profoundly boring and unrealistic that it ends up being a major detraction. The addition of hexes and the renewed emphasis on combat has gotten my attention, but it does sound like the final implementation is pretty off.
     
  10. Seanmac

    Seanmac Chieftain

    Joined:
    Feb 17, 2004
    Messages:
    7
    In that case, the basic concept is correct, as armies most certainly do not stack.
     
  11. mrt144

    mrt144 Deity

    Joined:
    Jan 27, 2007
    Messages:
    11,121
    Location:
    Seattle

    It's not that simple though.
     
  12. Gatsby

    Gatsby King

    Joined:
    Jun 25, 2004
    Messages:
    655
    Some combination of the unit system used in CCTP and SMAC would be best imo. This could be done by combining elements from CCTP such as:

    - a upt limit that still allows multiple units to occupy the same tile (in CCTP, up to 9 units could occupy the same tile)
    - the ability to group and un-group units into an army which could then move, attack, and defend as one

    with elements from the SMAC system such as:
    - artillery units are able to engage in ranged combat with each other, and units in the same tile and on the same team as the losing artillery unit receive collateral damage
    - artillery units can bombard a tile with one or more enemy units, causing a small amount of damage to each unit in that square.

    Combined with unit production limits based on available strategic resources and/or population, this might work pretty well at avoiding the congestion and tedium problems of 1 upt while at the same time discouraging stacks of doom.
     
  13. JBConquests

    JBConquests Prince

    Joined:
    Nov 11, 2003
    Messages:
    342
    Location:
    North Dakota, US
    Fewer units will make it easier to make the AI smarter.

    I do agree that we want a smarter AI. It is funny to see an AI sent a lone archer to attack several ground units.
     
  14. Jolly Rogerer

    Jolly Rogerer Prince

    Joined:
    May 16, 2008
    Messages:
    402
    My thoughts exactly. I play the Civ games for "everything else". Combat has always been annoyingly unrealistic and dull for me too. There are better games that focus on combat. In fact almost every other game has a better combat system than the Civ series because they focus more on combat and in almost every case to it better.

    The truly irritating thing is that there are so many combat systems that have already been invented (30-40 years ago often as not) that would be a better job in that they would be more realistic as well as more fun. No need to do anything but choose one and implement it. My preference is for something that doesn't become the main focus of the game (like Total War series games), but even a complicated tactical sub-game would be fine if one were allowed to auto-resolve in order to spend one's time on the other parts of the game, or to avoid the inevitable huge advantage one gains against AI opponents in tactical combat.

    Another massive design flaw in the whole series is the inability to rein in military power as the ultimate arbiter of everything in the long run. (Caveat, have not played Civ Rev.) This line of thinking is perhaps understandable considering the influence of World War Two on the modern psyche, but people who study history should be well aware that massive conquests are the exception rather than the rule, and massive conquests that have a lasting impact are exceedingly rare. Note the demise of the Warsaw Pact and the devolution of the USSR / Russian Empire.
     
  15. dexters

    dexters Gods & Emperors Supporter

    Joined:
    Apr 23, 2003
    Messages:
    4,182
    Gender:
    Male
    Location:
    Canada
    I actually don't think processing resources is a restriction insofar as a rules based AI won't even cap out the cycles a modern CPU has. Now if they go with a completely new AI engine that tries to emulate human thought, maybe even just moving 1 unit would be a challenge. But as a consumer product Civ5's AI is as far as I know, rules based, and really those rules need some tweaking and refining now that they're getting feedback from their customers.

    The reason turns are slow and game is sluggish to some is the graphics, and all the other things going in the background. And maybe the game isn't just very well optimized at the moment.

    I think I immediately noticed are the save files 600-900k and maybe more. BTS save files averages 200-400KB both are standard maps. Same with Civ3, even when the average # of units are easily 10-20x those in Civ5. So the game is keeping track of a lot of extra information
     
  16. Psyringe

    Psyringe Scout

    Joined:
    Dec 7, 2001
    Messages:
    3,394
    Location:
    Berlin, Germany
    Not sure if you already accounted for that, but I saw mention of a bug that causes savegames to be inflated with junk data after loading one while a game is in progress (i.e., not from the main menu). Someone else who started analyzing the save data found lots of strings reading "FINAL VERSION". So the size of the save games might not yet be a good indicator of much information the game actually tracks.
     
  17. QES

    QES Court Jester

    Joined:
    Jul 18, 2006
    Messages:
    2,139
    Location:
    Minnesota USA
    What about making the Great General a Unit that can "hold" land units like the aircraft carrier can "hold" air units? (Similar but not the same as civ 3). The unit attacked in the general's stack would be whatever had the highest defensive value in the situation (civ 2). Generals could hold as many units as the era/age/technology offered. Maybe initially 2? Max of 5 in the end game?

    Only generals would allow stacking, this would make their utility quit potent.
     
  18. Venereus

    Venereus This Is Streamlined!

    Joined:
    Aug 8, 2010
    Messages:
    1,026
    You're back at the "3-unit-stack becomes the smallest unit you really use" problem. And you can easily get enough generals to have all your units like that. So why bother?
     
  19. QES

    QES Court Jester

    Joined:
    Jul 18, 2006
    Messages:
    2,139
    Location:
    Minnesota USA
    Because making it "general dependant" means you're not able to really guarantee getting them. Battling 1v1 would still be very common. Even if you were able to get 2 generals, that's still only two stacks of 3, You could, very reasonably, have a nation that did not focus on great generals (or simply currently lacks them) and field 8-9 individual units to try and pluck away at the stack.

    Containing an army also becomes more possible, rather than outright obliteration, as a stack cannot outflank anything. Ranged units should also damage every unit in the stack simultaneously, rather than one at a time like melee.

    The stack would be powerful, but there are drawbacks.
     
  20. Earthling

    Earthling Deity

    Joined:
    Nov 9, 2008
    Messages:
    8,518
    That's in no way necessarily the case. It might make the game run faster, but it won't necessarily mean the AI is better and such a design could often makes things worse as it kinda did here. You are likely conflating several other factors, but a more accurate list of what would make the AI better would really start with things like:

    -less dependence on luck or random chance
    -less distinction between different units and environmental/other effects on units - in our case, this would mean fewer and less important promotions, terrain bonuses, etc...
    -fewer possible actions for units to take (we're mostly good here, yes there are ranged and melee attacks but not like wargames where each unit has several types of attack/defense)

    So yes, a "simpler" system allows more easily for the AI to compete. This is not the same as "fewer units" - just aiming for "fewer units" could in fact end up with the exact opposite, with the human being even more effective with the tactical choices available and no AI advantage in scale or numbers.
     

Share This Page