On Failing To Build A Wonder

[Deadstarre]

Can't we just have it as an option when setting up the game? I try to avoid mods other than CBP - it gets too messy.

Well the question becomes does that feature actually belong in the cpp or not. they can't just add options to suit everybody, but if they feel it belongs as part of the mod it of course would get done. its a hard argument to make that it really belongs though

to me vassalage is currently kinda a thing that doesnt exactly belong, it was added in apparently due to a strong following from civ4 users who wanted their favorite features of that game present (im guessing). we'd need a strong showing of civ2 fanatics here heh. did anybody else play the other civs actually know if notifications were in place for them or not? im curious when that got scrapped, i dont have enough history of this franchise though.

 

[Gazebo]

Well the question becomes does that feature actually belong in the cpp or not. they can't just add options to suit everybody, but if they feel it belongs as part of the mod it of course would get done. its a hard argument to make that it really belongs though

to me vassalage is currently kinda a thing that doesnt exactly belong, it was added in apparently due to a strong following from civ4 users who wanted their favorite features of that game present (im guessing). we'd need a strong showing of civ2 fanatics here heh. did anybody else play the other civs actually know if notifications were in place for them or not? im curious when that got scrapped, i dont have enough history of this franchise though.

Vassalage wasn't created by me, though - it was part of C4DF, and was integrated because it was already part of whoward's DLL (and some features are nice).

The problem with wonders returning investment in production is that it is fundamentally broken as a concept. It'd end up being a stockpile-leapfrog model that just really doesn't work, and would need lots of lengthy exceptions to make functional. It's just not a great idea once broken down into it's exploitable parts.

G

 

[Deadstarre]

The problem with wonders returning investment in production is that it is fundamentally broken as a concept.
G

i agree wholeheartedly. i should have specified, i was talking about notifications for when work on a wonder has begun by another civ- thats the feature that was in civ 2 i meant. even civ 2 gave no return on invested production =)

 

[Gazebo]

i agree wholeheartedly. i should have specified, i was talking about notifications for when work on a wonder has begun by another civ- thats the feature that was in civ 2 i meant. even civ 2 gave no return on invested production =)

There's a mod for that (Wonder Race). I think it overlaps with spies too much, so I don't personally recommend it.

G

 

[Gidoza]

The problem with wonders returning investment in production is that it is fundamentally broken as a concept. It'd end up being a stockpile-leapfrog model that just really doesn't work, and would need lots of lengthy exceptions to make functional. It's just not a great idea once broken down into it's exploitable parts.

But how is it an exploit, really? You only have as much production as you've actually generated. The exploit in Civ 2 was that you could keep "building" a Wonder that was already constructed - however, if you had saved up for something you had not finished and changed production type, all saved up production would be used up immediately on the next project. And no - I can't say how I see why switching to another Wonder is an exploit, because whatever you're getting still ends up as 2nd class from what you wanted; the only exception being that you could "prepare" production in advance before you have the right technology for a future Wonder: that's the only thing I'd see as worrisome.

 

[PurpleMentat]

But how is it an exploit, really? You only have as much production as you've actually generated. The exploit in Civ 2 was that you could keep "building" a Wonder that was already constructed - however, if you had saved up for something you had not finished and changed production type, all saved up production would be used up immediately on the next project. And no - I can't say how I see why switching to another Wonder is an exploit, because whatever you're getting still ends up as 2nd class from what you wanted; the only exception being that you could "prepare" production in advance before you have the right technology for a future Wonder: that's the only thing I'd see as worrisome.

Two scenarios. In Scenario A, you can use saved wonder production on anything and lose nothing. In Scenario B, you have to use wonder production on wonders.

A: It is bad play to ever build anything in your capital that is NOT a Wonder. If you get beat to the wonder, then you swap to whatever you need for as long as you get 1-turn production. Start of game, you research Wheel, start building Stonehenge. If you get it, GREAT! You'll get first religion. If you don't, GREAT! You get a one-turn Settler. After, you start building Pyrmids. If you get it, GREAT! You get a settler and a worker boost. If you lost it, GREAT! You get another one-turn Settler. During all the time you are building, your capital is growing. You never have to stop growth to get your settlers.

B: You try for Stonehenge. You miss it. What happens if you haven't researched another Wonder tech yet? We're not allowing 'you lose it' so it gets banked for future use while you work on something else. Research Calendar and start on Pyramids, you start near half done. If that fails as well, you are now basically guaranteed your choice of Ancient or Classical wonder from what you have banked. Roman Forum, Hanging Gardens, Oracle, Great Library, Angkor Wat, take your pick.

Both of these scenarios seem like something that only benefits the human player, and see a lot 'gamier' than the wonders being a race. In both scenarios, building wonders goes from being a calculated risk to something you should ALWAYS try to do.

 

[Gidoza]

Allright, those I can certainly agree on.

But not providing when someone else has started a wonder I can't - if it's a calculated risk, provide me with the actual information to make the calculation. Doing it blind is silly.

 

[Funak]

Doing it blind is silly.

But that's the whole point about wonders? If you could guarantee them every time there would be no point in a wonder system at all.
If you want to use that outside mod that tells you if you're winning a wonder-race you're absolutely welcome to do that, but adding it to the CPP makes no sense at all.

 

[PurpleMentat]

Allright, those I can certainly agree on.

But not providing when someone else has started a wonder I can't - if it's a calculated risk, provide me with the actual information to make the calculation. Doing it blind is silly.

If you know, for certain, someone else has started it or how many turns they have to complete it, it is no longer a risk. It is just a calculation. How many hammers you got? How many hammers they got? Who has more? How long have they been working? How long have you been working? All boiling down to who will finish it first. You can figure this the moment you get the notification. There is no risk there, just arithmetic and tedium.

As it is, the question isn't a solvable "will I build it first?" Instead, you have to ask if anyone you know has researched the tech, what is the likelihood they have started the wonder with what you know of their civ, how fast can you finish it, how much will it hurt you if someone beats you too it, and other factors.

 

[Deadstarre]

If you want to use that outside mod that tells you if you're winning a wonder-race you're absolutely welcome to do that, but adding it to the CPP makes no sense at all.

but he never asked for a mod that told him hes winning he was asking about being told whether other people have begun or not

If you know, for certain, someone else has started it or how many turns they have to complete it, it is no longer a risk. It is just a calculation.

knowing they started it and knowing how many turns till they complete it are 2 entirely different things. "atilla has begun work on the louvre" - ok so where am i? attila has 8 cities. even if i knew where it was, i have no way of knowing what his hammer output is, theres a lot more factors than just his terrain and really, nobody is counting up the hammers in his terrain- i dont even know what hes working. youre assuming too much knowledge is being gleaned from this simple notification, nevermind that the AI spawning an engineer just completely throws that all out already

in civ 2 you got notifications but that never meant you didnt lose a wonder- not even close i lost them all the time. they didnt have great engineers but there was something similar that added instant production. so i knew america had begun work on the wonder- if i needed that wonder i'd go for it anyway. sometimes i got it, sometimes i didnt- but it was actually a race. sometimes id get an alert they begun a wonder i had already started! this lit a fire under my ass and i knew i'd better hurry up- it was an element we're currently largely missing, theres no question on that really.

i still dont think we really need this function in the cpp, but the current system doesnt provide a race outside of spies function giving you this notify already, and that seems to be a part of whats bothering Gidoza. its worsened by the fact that we have some very lame duck ways (wonder skeletons and snooping through the tech tree) of telling whether the work has begun or not, but nothing straightforward or practical. personally i cant remember having a spy alert me about other civs wonders so im not really sure how it even works right now. most of the time my spies are in my cities or city states, but thats not important =)

 

[Edaka]

The thing with spies notifying you about foreign wonder construction is that they don't always do it as soon as the city with your spy in it starts constructing a wonder. I often get these notifications, but when I check to see how far Attila has to go until the completion of Louvre it turns out he's been working on it for good 10+ turns and is about to finish it in just a couple, even though my spy has been sitting there for that whole time.

 

[Gidoza]

But that's the whole point about wonders? If you could guarantee them every time there would be no point in a wonder system at all.
If you want to use that outside mod that tells you if you're winning a wonder-race you're absolutely welcome to do that, but adding it to the CPP makes no sense at all.

You misunderstand me. I don't want ALL the details - the start time is fine. If someone else starts a wonder, and then I have access to it 3 turns later, I might still consider running the race. If I fail then, it's my problem - the point is that I had some kind of meaningful information to use to actually choose to engage in the race, and I chose to engage in the race anyways because I figured that between my gold and production, I could win. This does not seem unreasonable to me.

And regarding comparing hammers: the only meaningful way to do that is to have a Spy, which isn't available in the Ancient Era. And one of the reasons I don't like or play with EUI is because it gives too much information: all I want is a single, generic piece of information. The information we do have available is either way too specific, or way too useless. I'm quite content to just know a start time without any other info, and I'll play the guessing game risk from there. The "risk" factor as it currently stands isn't risk, it's a guessing game. Am I supposed to know what tech an undiscovered civ has in order to build a Wonder? If I can't build a Wonder because some civ I never knew about built it first (and therefore neither should I know about it, because I don't know them), then I should be able to know (even though I can't know it) that they're building it. If we're not going to make sense here, at least let's be consistent about how little sense we make.

i still dont think we really need this function in the cpp, but the current system doesnt provide a race outside of spies function giving you this notify already, and that seems to be a part of whats bothering Gidoza. its worsened by the fact that we have some very lame duck ways (wonder skeletons and snooping through the tech tree) of telling whether the work has begun or not, but nothing straightforward or practical. personally i cant remember having a spy alert me about other civs wonders so im not really sure how it even works right now. most of the time my spies are in my cities or city states, but thats not important =)

Thank you.

 

[Llednar Twem]

The production refund is actually pretty much exactly how it works in Alpha Centauri. If someone else finished a secret project before you, you could switch to a different one and keep accumulated minerals. You would only loose some to re-tooling if you switched to a different type of thing like a unit or regular building. I thought it worked rather well.

 

[Gazebo]

The production refund is actually pretty much exactly how it works in Alpha Centauri. If someone else finished a secret project before you, you could switch to a different one and keep accumulated minerals. You would only loose some to re-tooling if you switched to a different type of thing like a unit or regular building. I thought it worked rather well.

Worked fine in SMAC, sure, but that's an utterly different game, with very different production modeling for cities and armies. Not really a great comparison.

You misunderstand me. I don't want ALL the details - the start time is fine. If someone else starts a wonder, and then I have access to it 3 turns later, I might still consider running the race. If I fail then, it's my problem - the point is that I had some kind of meaningful information to use to actually choose to engage in the race, and I chose to engage in the race anyways because I figured that between my gold and production, I could win. This does not seem unreasonable to me.

This piece of information rewards the human at the expense of the AI - I don't particularly like that the spy can do it, but it's there, and it is marginal enough that it doesn't make a big difference either way. In any case, that's why I won't implement it in the CP as you describe. You are welcome to solicit another person to do so, however.

G

 

[Gidoza]

This piece of information rewards the human at the expense of the AI - I don't particularly like that the spy can do it, but it's there, and it is marginal enough that it doesn't make a big difference either way. In any case, that's why I won't implement it in the CP as you describe. You are welcome to solicit another person to do so, however.

The game is multiplayer, too.

And why is it only at the AI's expense? Have the AI run a check or even let it "cheat" a little to approximate real approximation - let it know the rate of production and where it's being produced, then it won't bother starting it if it's a pointless endeavor. I find this a much more sane system than the one we have now.

And sorry, I have to disagree - to have a system where even the spies wouldn't be allowed necessarily implies that we need to get rid of the other factors as well: be rid of the skeletons (because they can tip you off, like spies); be rid of all technological cues (because they could tip you off). The situation, even as it is now, is that since spies don't report every turn, then if I want to build a Wonder, I need to check the city screen of every city every turn, just in case someone might catch up to me or bypass me or has started it first; then, I use my exploration knowledge and check every city for the right skeleton and see if something is under construction recently; then, I check my opponent's tech, and calculate based on my best guess what is going on and what this AI is likely to build.

This is stupid. Civ 5 is a game - could it be a game, please, and not a research paper? To need to peel through this much information when instead the game could just plainly tell me is idiotic. If the only thing the AI loses is some advantages in Wonders, then we might even call it balanced on Immortal, because I'm quite confident it'll still be winning many Wonder races, anyways, with all its starting advantages and the others.

 

[Gazebo]

The game is multiplayer, too.

And why is it only at the AI's expense? Have the AI run a check or even let it "cheat" a little to approximate real approximation - let it know the rate of production and where it's being produced, then it won't bother starting it if it's a pointless endeavor. I find this a much more sane system than the one we have now.

And sorry, I have to disagree - to have a system where even the spies wouldn't be allowed necessarily implies that we need to get rid of the other factors as well: be rid of the skeletons (because they can tip you off, like spies); be rid of all technological cues (because they could tip you off). The situation, even as it is now, is that since spies don't report every turn, then if I want to build a Wonder, I need to check the city screen of every city every turn, just in case someone might catch up to me or bypass me or has started it first; then, I use my exploration knowledge and check every city for the right skeleton and see if something is under construction recently; then, I check my opponent's tech, and calculate based on my best guess what is going on and what this AI is likely to build.

This is stupid. Civ 5 is a game - could it be a game, please, and not a research paper? To need to peel through this much information when instead the game could just plainly tell me is idiotic. If the only thing the AI loses is some advantages in Wonders, then we might even call it balanced on Immortal, because I'm quite confident it'll still be winning many Wonder races, anyways, with all its starting advantages and the others.

It's whatever you want it to be. I'm just telling you that I don't see this having a role in the CBP. You are, as it stands, one of few who want it, and it would be a sizable change to gameplay.

Also, 'have the AI run a check' is a maddening statement. It is comparable to saying 'just make the AI better.' It is extremely frustrating from a designer perspective to be told that, as if I can 'just do it.' What you are asking for is a pretty big shift in AI logic, and in gameplay, and it could be easily gamed by a human player. Really easily. Imagine if you knew that the AI would suddenly shift its production every time you queued a wonder. You could queue it, wait for the AI to start, and then unqueue it. What then? Does the AI stop? Does it continue? If the latter, you've gamed the AI into spending turns on something you may not want and it may not want, just because you can. You could divert the AI away from a wonder you actually want this way. Heck, either way you've made the AI stop its plans temporarily, and without any actual 'decision' being made by the AI.

So this means we need to make the AI understand how to discern when to do this, right? Well we can't use flavors, as wonders have crazy high flavors compared to normal buildings. We can't use diplomacy, otherwise it is too easy to game. So what then? We program every single wonder into the dll, manually?

These outlines don't even take into account whether or not the AI would stop building it if you did. Or how maddening that might become for humans, getting bested by the AI every time for attempting wonders on higher difficulty.

I'm talking through this because I want you to understand that I'm not being arbitrary here. I've thought this through, and I'm not one to simply do something to do it. There's just too much at stake with this, it's too seismic of a shift to be made in this context.

G

 

[Gidoza]

OK so the "running a check" thought was a bad idea. Nevertheless, I think you're blowing the "knowledge of wonder construction" out of proportion - this wouldn't be seismic, and in 90% of cases wouldn't change results at all. Where it did change results, the human would get the *occasional* extra Wonder completed, and save some production from bypassing impossible projects. You speak of this as though the human will magically complete every wonder somehow if we reveal the start time. We need more evidence than this, don't you think?

 

[Gazebo]

OK so the "running a check" thought was a bad idea. Nevertheless, I think you're blowing the "knowledge of wonder construction" out of proportion - this wouldn't be seismic, and in 90% of cases wouldn't change results at all. Where it did change results, the human would get the *occasional* extra Wonder completed, and save some production from bypassing impossible projects. You speak of this as though the human will magically complete every wonder somehow if we reveal the start time. We need more evidence than this, don't you think?

This is going in circles. I think I've made it clear what the implications of this concept are for the game, and I've also outlined, as clearly as possible, why I'm not pursuing this. I'm sorry that you aren't happy with this choice, but I feel that I've been as transparent as possible with my explanation. As such, I'm not going to contribute to this discussion any further. I do hope you understand. Feel free to pursue this by asking another modder to attempt it, or by pursuing it yourself. This would be a fairly trivial modmod to develop, but it doesn't fit the design of the CP or the CBP.

G

 

[Gidoza]

Well, to backtrack to the "production is cheesy" variation, 3 conditions would be enough to eliminate any such cheese from the game.

1. When a Wonder fails, you get back your production, with a certain % cut off.

2. The production is ALL immediately invested in a new project (no carry-over).

3. The investment cannot be in a World Wonder or in a Settler.


Thus, any possible cheese is eliminated.

1. Can't say that building a Wonder is always worthwhile, because the production of all things from the cut-off would be slower.

2. Can't "save up" production from one Wonder to shift to another.

3. Cannot bypass halted population growth when building a Settler.



Is this satisfactory?


As for the "design" of CP or CBP - my issue has nothing to do with design of CP or CBP in particular. This kind of risk in a game as it stands is a fundamental flaw that requires some kind of correction.



P.S. For what it's worth, I don't think we're going in circles: The conversation has helped me to clear up the notches in my suggestion that I didn't see, and the legitimate portions are starting to stand out better - I call this progress, and I couldn't do it without you. Thank you.

 

[Deadstarre]

This kind of risk in a game as it stands is a fundamental flaw that requires some kind of correction.

the production thing is flawed in too many ways to even cover- it's just too enormous a buff/advantage to the player- you can't invent rules to make it work, and if you could you're getting into a ridiculous amount of work as was pointed out. try not to over-complicate the issue with a huge overhaul- most people dont even think there is an issue right now.

it's important to remember the current wonder risk model was -designed- in the game from the get-go, its really hard to argue that its a flaw.

the only feasible way to change the model at all is just a flat initial notification- with no change in the AI behavior at all owing to it. that would unquestionably be an advantage to the player that benefits ONLY us; however it's also an advantage we currently enjoy over the AI to begin with; if they stumble on a wonder skeleton they're not going to react to it. If they know 6 people have researched theology i seriously doubt thats going to stop them from attempting the Hagia Sophia. im willing to bet the only factor they care about right now is whether or not it fits into their game plan- thats it

As was pointed out in another recent thread; this game has gotten a lot harder from vanilla- for everybody, on every difficulty level. the AI is smarter than ever before, many things that were poorly designed and thus caused grief have been overhauled. the only question worth asking here to me was "would extending the player an advantage that they already have (to a much smaller degree) make any sense now, given the current state of the mod?" but he gave the answer already- no. better to let it go at this point