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. Nefelia

    Nefelia Prince

    Joined:
    Aug 12, 2005
    Messages:
    458
    The length of a game vetoes greatly by difficulty level, map size, victory conditions chosen, the skill of the player, and whether the player wants a rushed game in the first place. Making definitive statements based on one player’s experiences simply means that the statement will be false for any player who plays the game differently.
     
  2. Victoria

    Victoria Regina Supporter

    Joined:
    Apr 11, 2011
    Messages:
    10,463
    Often they are not on one players experiences but a group of players who play the fast game where fast means efficient. Now you can play less efficiently (in your case do not chop and have cities large spaces apart) but this then means you will not play such a fast pace and increase the chance of being crushed by an ally on Emperor at T250.

    People will say “I am getting crushed on king, can I have some advice” and the advice is based on the fast game but you can chose to use it or not, but do not criticise what is in effect people trying to help other people. This thread is really an advanced thread very useful to those that need the finer points once they can win on deity and like to push themselves.

    I think what is most incorrect in what you are saying is the word will, just because someone plays differently does not mean something will be false for them, and how differently is that player playing. The advice from the deity level players is often sound for most types of play. The Onus is on you whether you want to follow it or not.

    Also be careful of any thread started by the OP, it will be a high efficiency thread and he is one of the very few that can make you feel mocked on this forum, generally it is a lot more civilized but these threads can get hairy.
     
    Last edited: Oct 19, 2019
    acluewithout and Sostratus like this.
  3. knighterrant81

    knighterrant81 Warlord

    Joined:
    Aug 9, 2011
    Messages:
    196
    When reading Lily's threads, remember the goal is to find the most efficient way to play - defined as what will result in the shortest time to win possible. That may not always be the most fun way for you. But if you are struggling at higher difficulties or restarting tons of maps, you might want to think about how to improve your play. Lily's threads will cut out your inefficient assumptions and bad habits. It won't be fun, but improving isn't always fun or easy.

    I added numbers to your post for ease of discussion.

    1 - this is a pretty special consideration. Lily is talking at a pretty general level. If you decide you really need a Wonder (there is no such thing as a required Wonder, by the way) then it is ok to keep, say, ONE woods for the Mahabodi Temple, for instance.
    2 - Population (after size 4 or so, and value starts really dropping after size 10) and most unique improvements are inefficient. Besides, you usually have to chop first to put down a unique improvement anyway. Efficiency is a brutal razor.
    3 - In an efficient game (say, a Science Victory speedrun) the only reason you build a new city is for the additional Campus. Once you've built the Campus, it doesn't really matter how much production that city has available (everything else can be bought with gold or faith if necessary). You've also built your cities near efficient natural sources of production like Hills and sources of Chop, haven't you?
    4 - Appeal applies less in efficient games (again, which are mostly science-based, but even a Cultural speedrun is likely more concerned with Theater Squares and early Wonders - which, again, benefit from Chop). For most Civs, appeal doesn't really apply until very late, and you've probably already won by then.
    5 - Generally production is better than all other yields, and the timing (while difficult to math out because there are a lot of variables) can be assumed to put Chop over the top of any adjacency bonus. 60 production NOW is better than, say 60 faith or 60 science built up over 60 turns - when 60 turns represents a third of your game time.
    6 - This is pretty irrelevant. The AI is better at defending itself now, but it is still pretty bad at mounting an effective attack.

    Here's the thing. I don't particularly like to play this way. It tends to bypass a lot of fun (but inefficient) mechanics and options. But I do always open Lily's threads and try to learn from them. What I've learned today is that the buffs to Lumber Mills and nerfs to Chop have, in general, failed to make Mills better than Chop in almost all situations. I'm not even sure I agree with Lily that you should consider Mills on Online speed - because 1 production doesn't seem to outweigh the admittedly hard-to-math "I build it NOW" benefits.

    This kind of thread may frustrate you. Understand that Lily has thought about all of these things and he's skipping over them because they've been discussed (and dismissed) years ago. Just because it frustrates you or you don't want to play this way doesn't mean that he's wrong. Instead of trying to prove him wrong, try to learn from what he's saying. I'd suggest asking why he's not considering certain exceptions.
     
    acluewithout and Victoria like this.
  4. Ownsya

    Ownsya Warlord

    Joined:
    Sep 25, 2018
    Messages:
    215
    Gender:
    Male
    By the way, if lumber mills gave +4 production instead of +3, then the present value based on these same parameters becomes 81 > X = 80. 77.9 from production per turn for 75 turns, 3 from the chop after 75 turns.

    This suggests that with Canada, snow/tundra lumber mills are better than chopping. I can't remember if any other leaders have a similar bonus, but any time lumber mill is +4, it's this same conclusion.
     
    Last edited: Oct 19, 2019
  5. acluewithout

    acluewithout Deity

    Joined:
    Dec 1, 2017
    Messages:
    2,729
    Agree. Best to watch these threads while sitting back and eating pop corn. Always informative. Usually entertaining. Don’t take things personally.

    These threads have great stuff if you want to improve your game and play more efficiently. They are also great if you don’t play efficiently, because you better understand what you’re side stepping and ignoring.

    Personally, I think the most important thing is to play well. That means avoiding drifting and or grinding. It doesn’t necessarily mean playing to maximum efficiency. But it does mean I guess playing as efficiently as possible and making as many meaningful decisions as you can within whatever are your parameters of what you want to achieve and what are acceptable strategies and or goals.

    I mean, even if your goal is just to roleplay, it’s important to understand which governments or cards are inefficient, so at a minimum you can factor that into your roleplay and your expectations about chances of actually winning and or turn times.

    I mean, I usually start with a scout to maximise exploration, because I think that leads to more fun games. But I have no illusions about it being the most efficient option (usually it isn’t, although I realise there’s some debate about this).

    Basically, everyone on this forum is super tolerant and polite. Lily seems to get some additional leeway. I’m okay with that, because it’s often quite amusing, and he makes some really great points if you can just roll with the mindset and writing style.
     
    darkace77450 likes this.
  6. Ownsya

    Ownsya Warlord

    Joined:
    Sep 25, 2018
    Messages:
    215
    Gender:
    Male
    Recalculation taking into account the yields from the tiles after chopping and building a mine. The original analysis in OP assumed that if you chop now, production from that yield ends there, but really it doesn't, a plains still has +1 production per turn, and if it's on a hill even more. Also after chopping on a hill you can build a mine. So really the comparison should be lumber mill + chop later vs chop now + yield from tile (+ mine when appropriate).

    In this analysis, to calculate the scenario (Chop now + yield from terrain + mine when appropriate), I used the same number of turns and inflation values of course.

    Below are the results taking into account different terrain, with hills and without hills for both standard speed and online speed. Note that I've listed a yield of +3 under Lumber Mills because it is 1 from forest + 2 from mill. While if you're chopping now the +1 from forest is never applied. I have used the same parameters as in OP.


    I may have misunderstood what you mean here, but I think if you take into consideration that you can replace woods with a mine then you'll come to a different conclusion as shown above.

    EDIT: Did not consider builder cost for constructing a mine, misleading results have been removed. See discussion on builder cost below.
     
    Last edited: Oct 20, 2019
    acluewithout likes this.
  7. Lily_Lancer

    Lily_Lancer Emperor

    Joined:
    May 25, 2017
    Messages:
    1,753
    Location:
    Berkeley,CA
    If you build lumber mill on hills, you still get the hill+1, so chopping on hills still reduces your yield by 3p per turn. (from 2f4p to 2f1p)

    However, if you chop the woods on the flat plain, your citizen may have better places than the chopped 2f tile to work, so actually you suffer from 2f3p(milled) to 2f1p(work on another tile), you only lose 2p by chopping on flat tile.

    For convenience I just assume your citizen work at least on 2f1p tile. If your citizens can work on better tiles then the yield of lumber mill is reduced since even if you chop that tile your citizen can work elsewhere.

    Yes Canadian mills may be useful even on standard speed, but you need to be careful that you're assuming you still have to work on that tile even if it is chopped.
     
  8. Ownsya

    Ownsya Warlord

    Joined:
    Sep 25, 2018
    Messages:
    215
    Gender:
    Male
    This is where our analyses differ. I assumed the same tile will continue to be worked. Your assumption is probably more practical.

    But still, if you include building a mine (and work the same tile instead of moving to a 2f1p tile) then in that case you go from hills tile with 2f4p(milled) to 2f2p(mined). In this case, chop+mine gives better yield overall. If you will not build a mine soon after chopping (didn't calculate where the cutoff is), then yes milled hill is better than chop now.

    Although, I just realized I didn't consider the builder cost for the mine improvement. Maybe that will change the answer. Give me a value for mine improvement on builder cost and in that scenario I can subtract it from the X value and see.
     
  9. Lily_Lancer

    Lily_Lancer Emperor

    Joined:
    May 25, 2017
    Messages:
    1,753
    Location:
    Berkeley,CA
    You shall consider the extra builder charge cost. Chopping and building lumber mill basically uses the same number of builder charges so I don't take this into account, but if you wanna compare with mines then you shall take this. Builder charges are not cheap in Civ6.

    I guess a builder charge cost is ~25 at the early stage for standard speed, and ~12.5 for online speed? (Assuming feudalism, without feudalism this number is much higher)

    In the current patch Industrialization is sometimes prioritized and sometimes ignored until late game. I'm not sure. Assume mines to have +1p and +1p for apprenticeship so altogether +2p may be a better idea.

    So chop-and-mine:
    You gain 2f3p on that hills but pay an extra builder charge, which can be subtracted from your chop yield, so your chop yield under this situation shall be 55/27 instead of 80/40.

    Therefore you get : 55+2/0.05-2(since you have to wait another turn to build your mine up)=93 maybe on standard speed.
    27+2/0.07-2=53 maybe on online speed

    Yes this is the best. But if you don't have Apprenticeship and Feudalism (I usually do that in my games, beeline Printing or even Cavalry) then this becomes the worst situation since it becomes 73/39 on standard/online speed.

    You shall build mines if your citizens are still working on 2f1p tiles as long as you have Apprenticeship.

    Generally you shall build improvements, if it can gives you extra 2p (or other equivalent yields), but not build them if it only gives you extra 1p.
     
    Last edited: Oct 19, 2019
    acluewithout and Ownsya like this.
  10. Ownsya

    Ownsya Warlord

    Joined:
    Sep 25, 2018
    Messages:
    215
    Gender:
    Male
    Not sure if this is the best criteria since not all builder charges are of equal value. But it's impossible to make a generalized statement about how much the builder charge costs since it depends on what you use it for. One could model it after an opportunity cost, but that also depends on what you're after. Say we only care about production. Then the opportunity cost for building a mill is that we could've instead chopped the woods so in this case that's fine to say charge for mill equals that of chop.

    But then if we already chopped and now want to build a mine, well now the opportunity cost depends on what else is available (on other tiles) for the builder to spend a charge on. Suppose there is another woods, then instead of building the mine you can spend the charge on another chop, in this case the builder charge costs X (here, X=40/80 for online/standard). Alternatively if there is another woods available one could also consider building a mill instead of chopping, so the opportunity cost of building the mine equals the yield of the mill over time T which in this case is 41.4/77.9 for online/standard.

    So to really get the ideal action (including potentially building a mine), it seems one should consider opportunity costs based on what else can the builder charge be spent on (assume in the same city only, not another city). Then the builder charge cost of making a mine will vary according to if/then statements. I prefer assuming an early game scenario before apprenticeship and feudalism:

    If another woods exists, then builder charge cost for mine improvement = max[chop, mill+chop] for that tile;
    Else, builder charge cost = ??? not sure what. If there is no opportunity cost since no woods available then maybe 0, but the builder charge should never be 0 cost, some standardized value should be used, i don't know what.

    Regardless, it seems the builder charge cost for mine improvement really is very expensive and your intuition:
    sounds in line with all the above. Maybe even never build a mine (on just hills) if a woods is available (unless you don't want to chop/improve that woods for some reason like saving it to chop something in later)?

    EDIT: removed results from previous post since without the builder cost they are misleading.
     
    Last edited: Oct 20, 2019
  11. Victoria

    Victoria Regina Supporter

    Joined:
    Apr 11, 2011
    Messages:
    10,463
    Gaining a +3 adjacency. +50% in a city, hard to chop that away.
     
    Ownsya likes this.
  12. Nefelia

    Nefelia Prince

    Joined:
    Aug 12, 2005
    Messages:
    458
    I can be unimpressed with poor presentation and overly narrow conclusions without being frustrated.

    If the intention is to provide an analysis and conclusion that only applies to a very narrow playstyle (speedy victory on Deity), then by all means make that clear in the initial post. That way I'll know not to bother discussing the issue because I am rather indifferent to that style of play.

    Either way, I appreciate your effort to clarify.

    I went back to the original post to try to put a finger on just what bothered me so much about the equation and analysis. I think the following is a good place to start:

    Viewed another way, that value inflation of 5% means that the chop of 80 production translates into 3106.61 production 75 turns later. That is a staggeringly high number suggesting that chopping is basically an "I win" mechanism.

    A most extreme best-case scenario in which a chop completes a wonder that would otherwise be stolen by an AI rival, such a valuable chop makes sense. But in more mundane best-case scenarios, the chop would translate into completing a settler a handful of turns later, which would hardly translate into 3106.61 late-game hammers.

    In all other cases - producing units, districts, or buildings faster - a mid-game chop of 80 certainly does not translate into 3106.61 late-game hammers.
     
    Last edited by a moderator: Oct 21, 2019
    leandrombraz likes this.
  13. Lily_Lancer

    Lily_Lancer Emperor

    Joined:
    May 25, 2017
    Messages:
    1,753
    Location:
    Berkeley,CA
    Why does a mid-game 80 hammer not translate to 3100 hammers late game?
    Suppose you have one more horseman in mid game like T70~80, don't you think it is better than 2 GDRs when the game is about to finish? Those GDRs do not have any real use, however the mid game horseman does.

    On T80 maybe my whole empire is only producing less than 50 hammers per turn, however on T160 maybe the whole empire is producing 2,000+ hammers per turn. And, 1 turn's time always worth the same.
     
    Victoria likes this.
  14. Victoria

    Victoria Regina Supporter

    Joined:
    Apr 11, 2011
    Messages:
    10,463
    Irregardless of difficulty, because it is not just related to deity. The AI does not chop and therefore every chop you do provides advantage. On deity it is used often to catch up but you can win without chopping.
    On Prince there really is little point apart from perhaps to get some wonders faster.
    Chopping is the single most powerful ability in the game and pillaging is not far behind, You will find Lily has threads that say they both need nerfing and I agree.
    Lily has done a very good at building in the opportunity cost IMO
    You are on your second game. Give it a chance.
     
    acluewithout likes this.
  15. Nefelia

    Nefelia Prince

    Joined:
    Aug 12, 2005
    Messages:
    458
    I... am on my second game? Whatever gave you that idea?

    I have 1378 hours played on Civ VI. I may not be in the 2k club, but second game?

    I do have to thank you for the chuckle though.

    On another note, I’ll have to strongly disagree with your assertion that the AI does not chop. I see it happen regularly, and by late game the AI’s territory is nothing but endless farm land where once there were forests.
     
  16. Victoria

    Victoria Regina Supporter

    Joined:
    Apr 11, 2011
    Messages:
    10,463
    Aaah mixed you up with a post by someone else a couple of days ago. Apologies but glad it pleased.
    Ok, well I’ll take that late chop at face value but all I know is when you are chopping well and you take a city late game it has either chopped nothing or very little allowing you the pleasure.
    And I guess this is the crux of it, late game non chopping is longer than late game chopping.
     
  17. Ownsya

    Ownsya Warlord

    Joined:
    Sep 25, 2018
    Messages:
    215
    Gender:
    Male
    Lily estimated 5% inflation rate based on the increase of his empire's production per turn. This value is not necessarily correct value, it is an estimate. The true value doesn't matter as much as understanding what this all means (on standard speed):
    1. Chopping is more powerful than building a mill (unless you have +3 mills like Canada's UA)
    2. The value you get from mills / mines decreases over time since the game is nearing it's end and the number of turns available to get yields per turn is less. Therefore chopping makes even more sense later in the game. @Sostratus' comment above links to a previous analysis that demonstrates this clearly.
    So how should this inform/revise one's gameplay?
    1. You don't have to change anything with how you play if you're already doing fine and not interested in being more efficient/faster. Fun and enjoyment trump efficiency, it's a game.
    2. If one does want to be faster, plan carefully ahead what woods you wish to keep for say, district adjacency bonuses or for later wonders in the game, (or indeed the climate, trees help prevent drought and deforestation amplifies co2 effects). Everything else is chopping wood. Chop as soon as you can to boost your production and advance more quickly. Having something now is better than having it later.
    3. For trees that won't be chopped, the sooner you build mills on them the more value you will get out of them. In the late game if no chopping wood is left and you really wish to rush something then it could make sense to chop these tiles too even if it does need 2 builder charges to remove the mill and chop. But try to avoid this situation by planning ahead.

    I guess the final question I have is, do the same rules apply for both woods and rainforest? You only get half the production when chopping rain forests, although the ability to build mills on them (Mercantilism) comes relatively late, when mills have even less time left to be useful.
     
    Last edited: Oct 21, 2019
  18. Victoria

    Victoria Regina Supporter

    Joined:
    Apr 11, 2011
    Messages:
    10,463
    Jungle is a little more complex
    We still have adjacency but must consider pop,appeal and amenities. Chopping a city to 15-20 pop may or may not be a good thing depending on what you are doing but 10 pop is the target. It used to be a no brainier with goddess of the harvest and excluding adjacency concerns the general idea is just do it, but yes lumber mills are an alternative.
    Normally era points are not an issue but early chop of jungle to get you to 10 pop is an option.
     
    Ownsya likes this.
  19. acluewithout

    acluewithout Deity

    Joined:
    Dec 1, 2017
    Messages:
    2,729
    Good summation.

    I guess FXS moving lumbermills forward to construction was an attempt to address this issue.

    It’s also a bit of a pity chopping doesn’t have more of a negative - there’s meant to be an increased chance of environmental disaster, but it seems trivial at present.
     
    Victoria likes this.
  20. Lily_Lancer

    Lily_Lancer Emperor

    Joined:
    May 25, 2017
    Messages:
    1,753
    Location:
    Berkeley,CA

    Rainforest is a lot more complex.

    Let's take these (maybe incorrect) presumptions.
    1: For cities smaller than size 4, food is better than production.
    2: For cities between size 4 and size 10, food is as good as production, if you have researched Rationalism, otherwise food is not as good, but we can count them as 0.5:1.
    3: For cities larger than 10, food has no value unless for the Urbanization eureka or it has Governor Pingala. (Or Kumasi CS so you really wish the city to be 13 or 16 or 19 or even 22 to make full use of KumasI)

    We have some options:
    1: Chop right now at Feudalism (we take a builder charge cost being 35 )
    2: Delay chopping until Rationalism (since at that time food becomes more valuable, take builder charge cost being 45), and that being 30 turns later than 1
    3: Wait for Mercenaries to Build Lumbermill (take builder charge cost being 45), and that being 30 turns later than 1
    4: Wait for Late game to chop( take builder charge cost being 65), and that being 75 turns later than 1.
     
    Last edited: Oct 22, 2019
    acluewithout likes this.

Share This Page