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

[GS] Rethinking Lumber Mills

Discussion in 'Civ6 - General Discussions' started by Lily_Lancer, Oct 15, 2019.

  1. leandrombraz

    leandrombraz King

    Joined:
    Sep 19, 2014
    Messages:
    979
    That's the problem with reducing how you play to a mathematical formula. You need to know how to react to each situation that is presented. If you're just blindly following a formula, you might get more total production here and there but you gonna be losing something, somewhere else for making the wrong decision at that time. Chopping might give more production on standard but it doesn't mean chopping is always the right decision. It was before because Lumber Mill was really weak unless on a river, now it's viable.

    Something to consider is flexibility. Sacrificing total production to get a city with better production per turn, gives the player more flexibility to react to whatever is happening on his game. If I need to produce something at a specific time for whatever reason, I can. Chop is powerful but once you use it, it's gone. OP reduces the matter to which one will give more production in X turns and he doesn't see that production per turn have its own advantages that can't be reduced to a general formula..

    I never said it was a record, just that it was pre-modern. You're reading it that way and giving importance to the turn I won because you seem to be incapable of looking at anything without your "quick wins master" glasses. It made zero sense to focus on the turn I won there, it make even less sense to bring that up here. What that has to do with chop vs lumber mills? Does chopping reduce the congress timer? Would I get a congress earlier if I chopped more?

    You're the one suggesting I should always chop without even considering any other variables. I'm not the one following a religion here. I chop when I need to, I go for lumber mills when I think it will be better for that specific scenario.


    You're not Newton and we're not talking about physical laws.
     
    Nefelia likes this.
  2. Lily_Lancer

    Lily_Lancer Emperor

    Joined:
    May 25, 2017
    Messages:
    1,731
    Location:
    Berkeley,CA
    I don't see what I'm really losing. The "casual" playstyle of not able to appreciate T219 standard speed as quick victory? Huh.
     
  3. Sostratus

    Sostratus Emperor

    Joined:
    Jul 31, 2017
    Messages:
    1,460
    Gender:
    Male
    Location:
    Minnesota, USA
    Sadly I am traveling and my laptop battery has gone kaput - but I don’t want to wait the few days in this thread before I’m back to my hole computers. Mobile formatting, hooray!

    Anyways, there was a thread (though a long while ago, in early 2018 at this point) about how to value chopping. I did an estimate based on economics, if anyone wants to dig it up, I think it was a thread called Choppity Chop.

    The core question of the harvest mechanic is
    “Which is more valuable: The instant yield or the per turn yield?”

    this is almost impossible to comprehensively answer in the bounds of civ6 except in what is arguably the case that people actually wonder if it’s worth it: just chopping woods to boost normal city production.

    As some posters like leandrombraz or Victoria have mentioned, there are substantial intangibles around what exactly you are chopping for-
    Namely competing for Positional rewards, like finishing a wonder first. That would probably be best modeled as something like an option that expires, but realistically player intuition is likely the best approach.
    For the case of clearing to place a district, then the long term value of the feature becomes 0 and its best approximated by whether its worth the builder charge to remove in time.

    However, if we remove those two types of scenarios, then we are left with whether it’s better for the city to chop the woods or build a lumber mill; in general this applies to all kinds of bonus resources and features, but it’s easiest with production. This sort of thing is best valued with a NPV or an annuity model. I used an annuity model in that thread way back. (The actual reality is slightly more complex because technically if you leave the woods, you can always harvest it later, for the unknown cost of a builder charge. But like no one is gonna run around their empire and harvest everything 5 turns before they win.)

    What lily is doing in fundamentally correct if you constrain to that scenario- chopping for generic production- and yes, choosing the end turn is quite important, because the game does end and this puts a sunset on production. I used a similar approach.
    But it’s not going to capture the fact that Production in civ6 does not have a linear utility function; that is, I only need so much of it to build out my core infrastructure and military and beyond that it’s much less valuable. I recall threads I’d make about Hansas where Victoria would ask “what on earth are you doing with all that production?!” Precisely because eventually you cannot actually build useful things anymore and the yield return on city projects is fairly poor. And with GS you can’t spam additional military units either.
    That said if you look at that 2018 thread, when you factor in improvements etc, it is useful for players to know that harvesting stone now to get that campus up is a net win or loss over having a quarry- and loosely when that transition point occurs. (I also plotted the value of a chop as an annuity so you could dial in things like, adjacency for a district as well, which would add +1 or something deoending on cards etc.)
     
    Meluhhan, Victoria, AsH2 and 3 others like this.
  4. Tabarnak

    Tabarnak Pô Chi Min

    Joined:
    Sep 17, 2010
    Messages:
    5,885
    Location:
    Québec
    I play online speed multiplayer and games are even more slower than singleplayer, making lumbermills even more important. I like to go for mills right after apprenticeship and time it with feudalism.

    Personnally i usually chop forests on hills and keep flat ones, unless i have a ton of them and use some to chop early in the game.
     
  5. Lily_Lancer

    Lily_Lancer Emperor

    Joined:
    May 25, 2017
    Messages:
    1,731
    Location:
    Berkeley,CA
    My calculation shows that the best way may be chop those flat ones and keep those on hills. Since the condition for lumber mill to be useful is that "even if the tile is chopped, you still don't have a better place for your citizen".

    However if your chopping provide easier defense then there's no problem. Being alive is always the highest priored.

    BTW mills unlock at construction so in my games I usually go for it directly after I researched ancient techs. Also you gain the entertainment complex eureka by researching construction early.
     
  6. Tabarnak

    Tabarnak Pô Chi Min

    Joined:
    Sep 17, 2010
    Messages:
    5,885
    Location:
    Québec
    I had think about going for mills before apprenticeship but never tried it yet. You convinced me. I don't remember perfectly but if construction tech remains in class. era it's a good thing to delay apprenticeship to get little more district discounts possibilities.

    Do you put mills on forest hills as much as possible? Do you have a kind of rule set for that part?
     
  7. Ownsya

    Ownsya Warlord

    Joined:
    Sep 25, 2018
    Messages:
    215
    Gender:
    Male
    I think that the inflation rate is what will make the biggest difference in the analysis in the end, and I don't think this way of picking the inflation rate is particularly good. Do you mean your total yield from all your cities? Or for a particular city?
    Depending on the city, I'm sure inflation rate will vary significantly. Not to say that an average can't be done, but it shouldn't be based on total empire yield per turn. That will vary too much if you are still in the expansion phase for example as you settle more cities. A more accurate average can be done if you look at yield per turn for several individual cities as a function of the age of the city.

    Which brings me to my next point. This inflation rate will also vary depending on the player and how fast the player is. Of course this is a matter of efficiency, and one could set a standard somewhere. Good players likely have figured out that the best way to play the game is doing x, y, and z and so will probably have similar progression rates in the game. That could be a standard. But for a slow player who perhaps haven't figured out the optimal ways to play, a generalized answer may not be appropriate for that player then. This is evident also by @Sostratus' Choppity chop post where he estimated inflation at 0.77% using a method based more on the design of the game rather than the player's progression rate.

    That said, I do realize that even overly-simplified models can often be powerfully predictive, but it must be shown that this is so, not assumed. It is a useful analysis nonetheless to say that from a pure production point of view on standard speed, chopping is generally better than a lumber mill. Then leave it up to the player to factor in specific conditions that may lead to a different conclusion for the scenario.

    So would a good way to understand this be: chopping should be the default action, but certain conditions can make a lumber mill the more sensible choice?
     
  8. Sostratus

    Sostratus Emperor

    Joined:
    Jul 31, 2017
    Messages:
    1,460
    Gender:
    Male
    Location:
    Minnesota, USA
    My Choppity chop (A+ thread title @acluewithout ) post was based on the fact that production costs scale from 1 to 10 between the start of the tech tree and the end. This is a true inflation rate, in that things straight up cost more and that eats at the value of production in the future. A dollar today is a dime in 5 years, type thing.

    However, there’s another compounding factor that I left out, which is almost impossible to estimate well, and that is the fact that production spent in the present can generate a real return of additional production in the future. In simple terms this is the return on your infrastructure. Building an IZ, for example, or a settler, or a builder. At certain stages of the game, you can spend very little production to get HUGE returns, like your first builder.

    because of this, it’s very hard to measure the actual value of a decision like: should I chop to push out this settler? That’s analogous to taking a bank loan to start a business: the return from the business needs to exceed the interest rate from the bank.

    but, in my long post, I was really assuming that yield was yield for players, and what they do with it is up to them; they only want to address getting the most yield over the game. So from that standpoint it’s a post about maximizing your lifetime income and says nothing about how to invest your money.

    In general these analyses are super sensitive to the inflation rate you pick and the number of turns yours looking at, especially over short periods. There’s really no right answer for that rate and you could totally justify a high rate if you’re a good player looking at the rapid expansion of the early game.
     
    leandrombraz and Ownsya like this.
  9. Alexadamz

    Alexadamz Warlord

    Joined:
    Dec 13, 2018
    Messages:
    230
    Gender:
    Male
    Location:
    São Paulo, Brazil
    It is a good and fair point of debate here. Doing the math for chopping now versus saving features for improvements' production in the long run depends on a lot of factors - and many of them are not very economically rational. I understand the OP's point, for maximization of efficiency, i.e., doing things to finish game faster. That is a point. But the game goes beyond that. Most players play the game for roleplay or such, and for that the economics model is not a good way to see the things. I think about Daniel Kahneman's book on thinking fast and slow: sure, i can do the math every turn for every action and maximize profit, but that can take a lot of the fun in playing the game.

    In the end, all depends on your focus as a player, do you want to finish the game faster and faster and beat your own record, or do you want to roleplay and do a beautiful empire, even if that takes longer? You can always achieve a middle ground anyway.
     
  10. Builderphile

    Builderphile Chieftain

    Joined:
    Dec 5, 2017
    Messages:
    51
    Gender:
    Male
    Location:
    Canada
    I originally posted because I thought your first post in response to Lily was unecessarily insulting. However, I have now read some of the other threads in this forum and realize that you and lily have a bit of a feud going so I can see where that crudeness came from.

    However, from a discussion standpoint, this "chopping vs. lumber mills" has been a constant source of second-guessing for me personally. I too have been leaning towards lumber mills over chopping in recent weeks, but it is certainly not "obvious" that it is the correct move (from a "playing optimally" standpoint). I don't think you should be discouraging attempts by posters on here to figure out the underlying math to uncover what the best strategy is. Improving my strategic game is one of the main reasons I am on this forum, and, like him or not, Lily is one of the best at bringing forward strategically important information about the game, and offering advice about how to win quickly.
     
  11. Sostratus

    Sostratus Emperor

    Joined:
    Jul 31, 2017
    Messages:
    1,460
    Gender:
    Male
    Location:
    Minnesota, USA
    I forget if it’s Kahneman or another author, but someone remarks that thinking fast is a vastly better idea if you have a good intuition about a problem over someone who has no expertise.

    One convenient side effect of performing quantitative analysis on Civ type games is that you can understand ahead of time what the dynamics are, so that during the game your brain can quickly “fill in the gap” with a good answer to a problem like this. I strongly encourage everyone to read my linked post from Choppity Chop; specifically because the bar chart demonstrates the kind of useful knowledge that can be widely applied in game. (It shows the value of a harvest action as a per turn yield for the remainder of the game.)

    If you loosely know that somewhere in the medieval is when harvesting a feature just for the production generally starts to overtake the value of just keeping it there, you can make an informed choice. This extends to questions like what if that woods is giving adjacency (when does the bar chart cross 2 yield, or 3, etc.)

    In a similar vein, laying out Infrastructure for several IZs can be a mentally time consuming task, but that’s why I advocate learning some basic templates so your brain can just intuit where to drop a template and fill in the details. (My signature has this covered in depth.)

    I also contend that a lot of 2018 posts on khevsurs really helped FXS see the need to buff them. I mean they were literally worse than swords, but you might not know that unless you did the math. What I’m saying is this is CivFanatics and some of us are overzealous...
     
  12. NukeAJS

    NukeAJS King

    Joined:
    Jul 28, 2008
    Messages:
    776
    Lumber mills are at a good place right now. It still makes sense to chop in some circumstances but it also makes sense to keep your forests and lumber mill them in others. To me, there are really just two situations where I chop -- I'm going to place a district or wonder (so I might as well chop) or I need something sooner rather than later (like wonders or a settler for a place that I'm worried is gonna get settled before I can complete it normally). Early-late game, I'll also chop once I get conservation and I have the builder charges to replace and place a lumber mill (although 3 charges for a single tile is pretty steep).
     
    Nefelia, lotrmith and Builderphile like this.
  13. Alexadamz

    Alexadamz Warlord

    Joined:
    Dec 13, 2018
    Messages:
    230
    Gender:
    Male
    Location:
    São Paulo, Brazil
    Yes, preciselly. That leads to the problem of changing your way of playing if you are biased towards crap strategies without realising the math. The best way is to learn some basic templates to help in most situations. Ofc you can do the math and think everytime to maximize everything, but that is not fun all the time for most of us. Learning the basic templates also helps roleplay, as you can build the empire the way you want without sacrificing too much on the efficiency.
     
  14. lotrmith

    lotrmith Prince

    Joined:
    Jul 28, 2009
    Messages:
    328
    For me it's more intuition over math because there are a lot of things you just can't quantify. What exactly are you chopping? Does you city have adequate production post-chop to be at all useful? Does what you're producing now provide a tangible benefit to going down immediately thanks to a chop (ie an early district in a new city) or is it something that can wait, saving the chop for something planned later? There are also the completely intangibles... did a chop now leave you without any resources to chop later that you may end up needing?
     
    leandrombraz and Nefelia like this.
  15. Tabarnak

    Tabarnak Pô Chi Min

    Joined:
    Sep 17, 2010
    Messages:
    5,885
    Location:
    Québec
    We can all agree that Lily is showing that chopping is always superior at standard speed only in a matter of finishing the game as fast as possible. It's a good indicator of what is best or not. Any discussions related to roleplay, personnal preferences, etc. have unfortunately no value in this thread.

    More you play at lower speed more the chop part is important. Since online doesn't reflect the 200% speed from standard it has to be corrected and explained.

    The question would have been surely different if it had a real 200% impact and the discussions would go around the units moves saved(like chopping settlers) more than anything else(i guess), and probably that chopping would have a higher value at online speed, but still not as much as standard.
     
  16. Aurelesk

    Aurelesk Warlord

    Joined:
    Oct 26, 2017
    Messages:
    133
    Gender:
    Male
    Chopping is an interesting mechanic when you know you are going to win in a short amount of turn. So, it is like playing a speedrun.

    For people that do not aim for short turn victory or don't even know they are going to win, or don't even know what victory they will aiming for, saving the woods are more valuable. Waiting a key moment to chop them, like make room for a district or wonder, secure a wonder, produce units in war time... feel more natural and you react with the flow.
    You may say that playing without knowing what victory type you are going to aim since 20th turn is kind of problem, and I agree.

    Basicly, the last part work with Lily Lancer argumentation: since you know what you are doing and aiming for, you basicly chop all the time because it is a key moment all the time.
     
  17. lotrmith

    lotrmith Prince

    Joined:
    Jul 28, 2009
    Messages:
    328
    Lily's way presupposes that the game will play out as planned without even a slight deviation. One surprise war, one bad barb spawn, one disaster and, without reverting to a prior save, it's Restart or bust. Not very fun and ultimately not a useful strategy.
     
  18. Lily_Lancer

    Lily_Lancer Emperor

    Joined:
    May 25, 2017
    Messages:
    1,731
    Location:
    Berkeley,CA
    You can be totally unaware of what you're going to aim even at T100, that's no problem.

    However you must have a clear goal of what you'll be doing in the next few turns. Also plan for events like surprise war or flood or barbarian infantries. (That won't really change the game however that may interrupt you by 1 turn or 2, making you re-do the planning)

    The idea is that under the inflation rate of 7%, what is 100 turns far away only worth 1/1000 than what is useful now. So you actually need to think more about present. That is, you don't plan far-away (I myself don't do that too), but you have to plan for now. You must find out what is the most urgent to do now. This requires a sense of timing. For example, you only have courser and your neighbor has 50 city defense. You may miss yourself but in fact you may be only 10 Turns to 62-strength Cavalry if you plan carefully. Your gameplay level relies on the fact of capture timing--Figuring that out "Oh I can get Cavalry 10 turns later, if I have 10 Cavalries, I can take my neighbor's whole empire!" And you start work on for that goal.

    When I'm saying T=50 or 75 that's only an estimation. Of course that number varies but if you can remove and chop in late game why not? So there must be a number.

    I play MPs, I'm always the one who develops most quickly around T50~60(when everyone else quits the game hearing I've got muskets or even cuirassiers) so I believe I have a more robust strategy. I only capture what is important, which option is strictly better than another(like chopping vs mill on standard speed), and don't revert or restart. Such playstyle may delay my victory from T160s to T170s but making things more robust.
     
    Last edited by a moderator: Oct 17, 2019
  19. Victoria

    Victoria Regina Supporter

    Joined:
    Apr 11, 2011
    Messages:
    10,223
    It was a very useful thread for the initial post but now has turned out to be a great thread. Great discussions and points.

    That additional value now rather than then is hard to add (+2 science +GS earlier) but is extra in the chops favour so any borderline decisions aren’t really so close. Naturally the cost of the builder also comes into it and hence the value of pyramids and why people wait until Feudalism which really is only part of the real answer which is the time it takes to get a builder and even the value of chopping in a builder which is a no brainier late game and early. It is also interesting to think how expected size of empire may (should?) alter the view of the value of every chop. The reality is we get builders rarely at the start because we are so busy building everything else which does beg the question of when is it more important to get a builder rather than a settler. I also wonder about new cities and the value of a mine over a chop or nothing. A new city has such poor production that a mine could be a serious increase but I guess Urban Planning takes the edge off that a bit.

    The workings highlight the value of the inflation and hopefully highlights one of the values of the logistics card which becomes of more value if you do not have a golden age (should be rare though in fast games).

    Could you imagine what would happen if we had unlimited charge builders like in V?

    It is weird, last night I played a game where I was efficient and at the end I noticed some uncut trees and crabs and it annoyed me. Other times I will be playing a longer building game and a large bare patch from chopping has to be covered with new growth purely for aesthetic reasons. It goes back to what @Sostratus said, that being informed is useful in most cases. As long as you understand rather than just know (“any fool can know something” - Einstein) you can make judgements around it.

    Final thought is drought, introduced as a chop negative but has no real effect on someone that does not improve their lands.
     
    acluewithout likes this.
  20. TheMeInTeam

    TheMeInTeam Top Logic

    Joined:
    Jan 26, 2008
    Messages:
    25,181
    The longer the game goes, the better having a lumber mill then chopping later looks. But chopping can also help end the game sooner.

    Presumably Lilly_Lancer is is making turn assumptions based on played games/win dates. Even if most wins aren't t120, the decision point for chopping in most cases will be at least a few dozen turns into the game too (some chops are earlier, but on average).
     

Share This Page