How jungles can make for terrible starts

Kevun1

Chieftain
Joined
Aug 18, 2021
Messages
13
I wanted to hear people's thoughts on the balance of jungles, especially when it comes to jungle starts. It seems to me that jungle starts are extremely rough, especially on higher difficulties. Imo, a jungle heavy start is probably the second worst start possible, next to a full tundra start. But even for tundra, there are many mitigating factors that can shore up its weaknesses (which I go into more detail below), while sometimes, you can just get completely screwed with a jungle heavy start.

For example:

Spoiler example start :

Ignore the fact that I'm Maya in this instance which can benefit from jungles.
jungle.png


Jungle starts can be incredibly :c5production: production starved

This is the biggest problem with a jungle heavy starts like the one shown above. In this example, I can literally count the number of hammers seen in the entire screen on one hand. Now, this is a bit extreme of an example, as it's situated on a peninsula, but in general, I have frequently found myself in a similar situation. The main causes for this are

  • Jungle hills do not provide any :c5production: production.
  • If you have jungle, you cannot have a forest, so you might not have any 1:c5food: 2 :c5production: tiles.
  • Luxuries resources that give :c5production:, like copper don't tend to appear near jungles.
  • To get any :c5production: production from your tiles, you need to build mines. Chopping jungle is available at Calendar, so you need both Mining and Calendar to build a mine on a jungle hill. Furthermore, building mines competes with connecting other luxuries/resources, so it comes with a big opportunity cost early.
  • It takes longer to connect luxuries/build improvements on jungle, since you need to chop. So to summarize, without any bonuses, it takes 12 turns to build a mine on a jungle hill (which is your only option to get production from tiles in many cases), you need both calendar and mining to do it, AND that competes with connecting your other luxuries, which also often take longer to do anyway because of the same issue of needing to chop.
  • To make things worse, it takes 2 turns longer to chop jungle than forests. This makes sense thematically, but it can be very punishing.
Finally, (I know this has been previously brought up), logging camps are usually worse than lumber mills. 1 :c5gold: gold is typically less valuable than 1 :c5production:, especially when you lack :c5production: to begin with. Logging camps are available a full era later than lumber mills, and to add on top of that, they take longer to build. I think this is a separate discussion, since these improvements come relatively late and I wanted to focus on starts, but I think this discrepancy definitely contributes to jungles being bad tiles overall. The one possible benefit is the 1:c5science: science from universities is often better than the extra :c5production: production from workshops.

As opposed to other unfavorable terrain, there are few mitigating factors

Tundra tiles might be technically worse than jungle tiles, but their lack of :c5food: Food can often be managed.

  • starting in full tundra is relatively rare (rarer than jungle heavy), and you can usually move out of the tundra with your settler, since tundra doesn't restrict movement like jungle does, and because tundra only appears at the edge of the map. This is technically mapscript dependent, but its true of the most common, standard, maps
  • You often spawn near the coast, which provides a source of :c5food: Food (from the coast and also from potential fish) that tundra otherwise wouldn't.
  • Tundra has a lot of deer, which again provide :c5food: Food. Again, mapscript dependent but the standard maps have this.
  • Probably the biggest factor: the tundra pantheons, Goddess of the Hunt and God of the Stars and Sky, provide :c5food: Food.
Desert starts can also be situationally bad, but deserts have
  • Flood plains and oases, which both provide a lot of :c5food: food.
  • Spirit of the desert can provide a lot of :c5production: production and :c5gold: gold.
  • Allows you to build Petra.
and thus can be frequently better than other standard starts.

Meanwhile, if you get lots of jungle, you just have to hope that there are also some forest plains nearby so that you can get some 1 :c5food: food 2 :c5production: production tiles, or hope that some nearby hills are not covered by jungle. Herbalists also give 1 :c5production: production to plantations, and jungles have a lot of plantation resources, but again, jungle resources take longer to connect than other resources. Also, you have to spend your precious hammers to build a herbalist, which you can often delay otherwise. Speaking of plantations, plantations also often give :c5gold: gold, which kind of mitigates the :c5production: production issue, but I find it is not nearly enough, especially factoring in the extra construction time in jungles and the opportunity cost of not building mines.

The main issue in my opinion is that the pantheons that are supposedly suited for the jungle, do not provide any :c5production: production:
  • Goddess of Renewal - in theory this would be great in a jungle heavy start, but again, it provides no hammers, and mockingly gives :c5food: food on Markets, which is the exact opposite of what you are missing in a jungle heavy start. This pantheon treats forests and jungles equally, but there is a fundamental asymmetry between the two that makes jungles far worse than forests. Perhaps this pantheon should give different yields for forests and jungles?
  • Goddess of Springtime - not specifically related to jungles, but plantations are mostly found in jungle. The :c5gold: gold is helpful, but again, we see :c5food: food, exactly the opposite of what we need in a jungle. Again, the time and opportunity cost of plantations factors in here too.
Thus, in practice, these pantheons are not pickable often, since picking Tutelary Gods is often essential. This also has big consequences for your chance of founding.

Furthermore, what about picking Authority as your policy, which gives lots of :c5production:? The problem is that jungles (and other rough terrain) make authority worse, as your units are slower to move around. Thus, you are slower to kill barbarian camps and slower to tribute city states. You also have very little base :c5production: production to begin with, which means less for units.

Personally, I've found the best strategy to pick Progress and go for Organization, which gives +3:c5production:, but generally you want to go right side first with Progress. Tradition is also pickable, but only if you can settle the rest of your cities in relatively high :c5production: production areas.


To summarize, jungle heavy starts without access to other types of tiles are often just terrible, and it is incredibly easy to fall behind on higher difficulties given the sheer importance of early :c5production: production. I'm not sure what is the best solution here, but my best suggestion would be to modify the "jungle pantheons" to somehow mitigate the weakness of jungle starts (or even add a new pantheon if that is feasible at this point).
 
You can chop trees with just Mining (or right at start? I don't remember). Your best bet here is to research Mining, buy a worker and immediately build mines on those jungle hills.
 
You need Calendar to chop jungle. Mining only allows you to chop forests.

This is part of the problem I was talking about with jungles. They block both the production from hills, AND the ability to build mines on them until you get both mining and calendar.
 
The :c5production: is from the plains underneath the jungle. Jungles always give 1 :c5food: to the terrain under it (and override hills).

I always thought jungle chopping was on a tier 1 tech. Maybe I'll change AssignStartingPlots:MeasureSinglePlot to count jungle hills as food instead of prod. This kind of starts will have a lower score then.
 
Last edited:
He's on Communitu or some other map script, those are not default jungle yields (and jungle always spawns over plains in vanilla/vp)
 
Part of the problem of jungles, especially on high difficulty, is also tactical. Jungles make your units blind and slow. It's harder to explore, clear camps, and move your army to engage barbarians. Meanwhile the barbarians have perfect sight everywhere and often have the jungle mover promotion. It may not seem like much on the surface, but the cumulative military effects of a large patch of starting jungle for the first 50-100 turns can be devastating. This is in fact one reason why I think Aztecs are very good, their Jaguars make them consistent such that they can start in jungle and not be instantly toast.

As for Tundra... Stars and Sky is busted good, an all Tundra start is amazing if you know how to use it and push a moderately fast pantheon to secure it (unless your starting food is REALLY awful maybe)
 
I do consider jungles one of that hardest starts, often a half - to a full bump in difficulty. Not impossible but definately tricky
 
He's on Communitu or some other map script, those are not default jungle yields (and jungle always spawns over plains in vanilla/vp)

I've played on Communitu for so long I didn't realize grassland hills never have jungle by default (maybe this is a good general practice for all mapscripts?). I did get unlucky here since all the hills were grasslands, but even with plains, you won't have any tiles with 2 :c5production: or more in a jungle start, which is pretty rough.
 
I've played on Communitu for so long I didn't realize grassland hills never have jungle by default (maybe this is a good general practice for all mapscripts?). I did get unlucky here since all the hills were grasslands, but even with plains, you won't have any tiles with 2 :c5production: or more in a jungle start, which is pretty rough.

Yes grassland jungle def make things worse, in theory it could be okish because 3food tiles but .... theres just no hammers.
I rarely work jungle tiles unless I saved some and went Rationalism - Scientific Revolution (which is somewhat late game), play Maya or Iroquois.
Its better with resources but even then its a lot of worker turns to improve.
This case with Tobacco as luxury really doesnt make the situation any better and starts like this is insta reroll for me.
 
I think the map script is making a big difference here. On the normal maps you get 2/1 on jungle and it will often generate bonus resources in the gaps giving you fine production. Jungle starts are sometimes a bit slower but more powerful later on when you get bonus science on all those tiles.
 
Normal maps still generate jungles over grassland. They just change the terrain to plains afterwards, creating ugly patches of mixed terrain land when you eventually remove them. I think changing the start placement is the way to go.
 
Yea I agree Jungle starts are terrible. For me the most frustrating part is that you will also have to chop down your jungles to connect your luxuries or any bananas that you do get, so you can't even get the bonuses from renewal or universities/rationalism later on.

I think it would be better from gameplay perspective if plantations didn't chop down jungles or maybe at least for bananas there was a special plantation which didn't.
 
Yea I agree Jungle starts are terrible. For me the most frustrating part is that you will also have to chop down your jungles to connect your luxuries or any bananas that you do get, so you can't even get the bonuses from renewal or universities/rationalism later on.

I think it would be better from gameplay perspective if plantations didn't chop down jungles or maybe at least for bananas there was a special plantation which didn't.

Yes plantations chopping forest/jungle makes sense from a realism point but its pretty bad gameplay wise.
That would also mean you didnt need techs for chopping and would make it a good early lux with the option to chop if you want.
It would probably put the flatland calendar lux like tobacco, perfume and incense even further behind.
 
I don't think anyone ever suggested jungle starts would be good. It's just a lot of extra worker-time chopping all of it and turning it into pretty horsehockey plains.
Anyways, some tiles are better than others, some resources are better than others, that's just part of the game and part of what gives it replay-value.
 
So, it seems there are four solutions :
  • Modify a pantheon to be better suited to jungle starts, or create a new one tailored for it (like there is one for tundra or desert) : something that allows production to be gained from Jungle and/or allows easier movement or vision in Jungle and in similar terrain (God of Night, Death God etc). I would actually prefer one pantheon to be modified to better fit jungle starts, and one to be create explicitely for these (so that you're not lost if someone else takes one).
  • Modify communitu so that all jungle tiles appear on plains
  • Divide the bonus production from chopping forests / jungles between Calendar and Bronze Working so that jungle starts aren't disadvantaged too much compared to forest start (that's at least how I interprete the problem about the tech divide between forest and wood chopping)
  • Make Plantations buildable on jungles (like Camps are buildable on forests). That however also means that Plantations will be buildable on marshes (since it's more that Camp don't require feature removal, and the resources associated with it don't spawn elsewhere than on forest or featureless tiles : it is not possible to make a resource improvement only remove some types of features).
 
Divide the bonus production from chopping forests / jungles between Calendar and Bronze Working so that jungle starts aren't disadvantaged too much compared to forest start (that's at least how I interprete the problem about the tech divide between forest and wood chopping)
No, the problem is you need Calendar to chop jungles at all. Maybe we can move that earlier.

The other solution is modify start scoring so civs are more unlikely to get that start at all.
 
The only problem with moving jungle chopping away from Calendar is that chopping jungle is required to build plantations in many cases. I believe it was originally changed to Calendar precisely because of this reason.

One possibility is to change it to a prerequisite of Calendar, like Wheel. It wouldn't make much sense thematically, and putting mines on jungle hills would still require two techs, but it would help a lot.

Again, there are other solutions as well. Personally, I just think it makes a lot of sense to have a pantheon suited to jungles that provides :c5production:, much like how tundra has stars and sky which provides :c5food:, but preventing really bad :c5production: starts is fine too.
 
I have never really understood why there are two separate prequisits for chopping forests and jungles, why can't they both be the same tech?

I assume IRL clearing jungle would take more work than clearing forest so having it take longer seems reasonable but as in both instances you are just chopping wood and clearing plants i have never seen the reason for it to need a separate technology to do so apart from to double down on jungle starts being harder.

Jungle starts generally need more tech, more workers and more time to get estabilish compared to none jungle starts and this wouldn't be so bad if the civs which have a jungle start bias actually made up for the issues around jungle starts but none really do...e.g;

Siam...any bonus to jungles comes quite late in the game and Siam actually suffers from jungle start as their unique mounted unit is actually slower than the equivilent so they have a slow unique unit with a bias toward terrain which slows them down even more.

Maya...does get the kuna which helps with science after the very early game but they do tend to take a while to get up and running due to movement penalties in jungle and comes after the roadblock to generally utilising jungle tiles so doesn't help with the core issue of jungle being a very difficult start.
Also tends to be less of a bonus as you tend to have much less production in jungle heavy starts so you often struggle to build or utilise the new shiney things you unlock from the added science.
Again generally suffers from jungle start bias in regard to it's unique unit which is ranged and is very hard to take advantage of in jungle terrain due to sight issues.

Brazil does get brazilwood camp but again this does not open up until calender anyway so no real benefit to actually starting in jungle and only adds gold (until much later) so doesn't help with the general production shortage that comes with jungle heavy starts even in base map scripts.

Aztecs at least have some advantage to a jungle start with their unique unit, jaguar, which is actually available at the start and does actually get some increased production from their unique building which makes them doable for jungle locations but far from advantageous especially as rough terrain is a two way street and jaguars suffer just as badly attacking into rough terrain as opposing units attacking them, especially as they are melee so you really do need a swarm of jaguars to make them really count which can be hard to pump out due to general lack of production a their peak time of use.
 
Last edited:
Top Bottom