General Balancing Thread

I wanted to create a thread on the forums since I notice balancing thoughts on Discord oftentimes get brought up, spur some conversation, and get forgotten. This is a thread for ideas which don’t require their own forums - an entire rework to the difficulty system for example should deserve a thread of its own. This is more meant to call discussion to a specific area, or store thoughts.

————

Storage

To start, I wanted to discuss storage balancing. Currently, I personally almost never find the need to use a vacuole, and very rarely see players create organisms with a vacuole in it on the showcase section of Discord (I saw maybe 2 creatures with them skimming it).

I definitely don’t think that vacuoles have to be a standard for every single cell, but I do think this points to storage being an exceedingly abundant resource for organisms. There are rarely times where I think “man, I wish I had more storage”. And it shouldn’t be so that my organism is constantly starved of resources, but players should be forced to consider means of increasing storage capacities; it creates deeper and more intentional thought towards fleshing out your organism.

I want to try two things:

  1. Reduce storage amounts for all parts and see how that goes
  2. Reduce storage amounts for eukaryotic parts only since storage for prokaryotes is less approachable

I think a fundamental part of this issue is the fact that storage starts getting doubled as your organs split for reproduction, which I do honestly think is problematic for balancing purposes as a whole. It makes it so that you have to treat values as double, which can make balancing for a single part excessive, and balancing for two parts punishing during the players time as the organism they actually created.

That’s a more dedicated revamp though, so I will be trying to json some storage.

3 Likes

Any luck with this? Obviously I could made that branch for you if you need me to, but it sounded like you got this.

3 Likes

Thanks for the check in! No need for the branch, but if this is something you personally want to see addressed soon and I’m being slow, I won’t be upset if you go for it. I’m going to try to attempt some balancing either this week or the next.

1 Like

I am going to try tinkering two different things soon:

Lifespan/Growth Rate - The community has previously given feedback of there being dull moments in gameplay which, generally, revolves around waiting too long for a reproductive cycle that is clearly successful. In other words, they consume enough phosphate/ammonia and resources to be sure of successfully reproducing, so it takes maybe 30 extra seconds of lingering around before entering the editor.

This isn’t that much of an issue in the later parts of the Microbe Stage since by then, some sort of competition tends to arrive and there are abilities to interact with. As simple prokaryotes however, where the environment can be empty and cell functions are pretty bare, this can be an issue.

As such, I will try to to reduce the starting reproductive time (how long it takes for LUCA to reproduce). If further steps are needed, I might also tweak the amount of ammonia and phosphate additional parts if reproduction time increases too slowly or rapidly.

Engulfment Rebalance - Currently, engulfment is really, really powerful. This is much better than how it was before, where predation was so unrewarding that you could barely fill up on energy from consumed cells. But an opposite problem is now present, where basically every organism the player sees can easily be consumed if the player bumps into them, and bumping into a crowd basically means a free trip to the editor.

Predation being rewarding isn’t the problem; I think it’s more how easy it is to transition to relying on heterotrophy. In my prior playtesting with Thrive, I’ve found that I inevitably started consuming organisms, even if I generally didn’t have that playstyle in mind. And, again, I found that just chasing organisms around turned out to be the quickest and most reliable way to get to the editor, despite my organism not being very finetuned towards adaptation. Ideally, it should take some design choices and specialization to become a predator, but, once adapted to it, predation should be rewarding.

Here are some balancing changes I am weighing for this balancing question…

a. Reducing Engulfment Storage - Currently, engulfment storage can be insane, allowing relatively innocent-looking player creations to voraciously devour an entire horde of cells for supper. Ideally, we should balance that in a way so that unadapted cells can consume maybe one small-enough organism, but require more adaptations and the placement of lysosomes to consume multiple small organisms or medium/larger organisms as well.
b. Adjusting Size-Discrepancy Needed for Engulfment - The size discrepancy between cells doesn’t need to be that different for engulfment to be successful. This could be looked at, but I am going to look at engulfment storage first before tackling this. The two in combination could be very restrictive and difficult to overcome. It could also complicate other aspects of the game, such as endosymbiosis.
c. Adjusting Digestion Speed/Efficiency Stats - I am not 100% familiar with how the lysosomes work exactly, but through personal experience and reading the developer wiki, I assume they tweak digestion speed and efficiency. Having baseline stats for each be lowered could incentivize players who want to become predators to invest in lysosomes more.

2 Likes

I think this is the wrong thing to do because there’s been no feedback lately that the early game is too slow to reproduce in. The only part of the game that has gotten that feedback recently is about trying to get to 20 colony members.

Additionally I think the early game cannot be faster because new players would be overwhelmed if they need to play it any faster. So I wouldn’t make at least the first 3-4 generations any faster than currently. I think it is plenty fast to get through the first few generations.

This could be true, and the info I’ve seen could be biased towards the people who want to chat with us, but from what I’ve seen is that everyone complaints about the AI being able to counter their engulfing with spikes. So I think that unless very carefully designed any change to make engulfing harder will just make it unrewarding again.

I was actually thinking recently that we would need some engulfment buffing for multicellular colonies as it is a pretty common thing to create a bug report when your colony lead cell is full on engulfed stuff but other cells are empty. For example if it was realistic we could add some kind of multicellular exclusive organelle for transferring ingested matter between cells. Anyway that isn’t related to what you said but I just wanted to share that.

I think this is a probably doable idea, but will require quite many changes everywhere to change how engulfing storage is calculated and how the check for what you can engulf works.

Technically this is much, much easier to change as it is just a single constant variable that needs tweaking. Though I guess there will be some in-game text that will be outdated with this change.

I think it would actually benefit at least the realism of endosymbiosis. I saw one screenshot where someone was big enough as a bacteria to endosymbiosise an eukaryote…

The baseline gain from engulfment has been buffed multiple times over the past few years. So if it is nerfed back to what it was, then all that tweaking was pointless and we are back to engulfing feeling “useless.”

1 Like

Some of the feedback I’ve seen is that the beginning of the game can feel somewhat slow. I don’t think any significant shortening is required, but even a roughly 10-15 second shortening can lead to much better balancing down the line. Regardless though, I’ll try both shortening the base time to reproduce and reducing the increase to reproduction time gained by placing additional parts exclusive of each other and see what feels right.

Huh, not sure if this is possible, but it is important to consider similar tweaks making it easier for cells to share material for predation to be worthwhile. I think I’ve heard it’s possible for simple multicellular organisms to break down whatever is digested and split those smaller components across different cells, but I could just be making this up to be fair.

Regarding the burdensome changes needed to engulfment; I think it is important, but it might be best to delay such a change to 0.9 and try other means first.

Fair

Were the tweaks an increase of the maximum amount of compounds given by an engulfed cell, or was that amount constant throughout these tweaks, but efficiency/speed of digestion was tweaked? I was assuming that engulfment was fixed by increasing the amount of phosphate and ammonia they could give.

Considering how rewarding predation currently is, I think it is fine if we require some specialization or part placement to fully take advantage of it. If the reward for adapting is good, then I think an additional burden to get to that point is fine; the satisfaction of configuring your organism to take advantage of that niche could be a nice feeling for players.

Regardless, that could be something to consider after adjusting the size discrepancy between the engulfer and engulfed, and adjusting storage.

Actually, I think only glucose has really been buffed. There’s a special function in the game that creates glucose out of thin air when you engulf something to make it more beneficial to engulf. And I think I remember the digestion efficiency base multipliers being increased to make engulfing more efficient. I think digestion speed hasn’t been changed (much), because that would nerf engulfing when you get less time with getting the resources.

I see a potential problem where if early engulfing is not rewarding, new players will try it and determine it to be a bad strategy and never commit to it to get the real benefits. I think quite a big potential pitfall in discussing changing the balancing of early game based on our experience and the fans who talk to us, is that the average new player’s experience is probably totally different. Reading recent Steam reviews I think we get more people complaining that they cannot understand the game than people complain that there isn’t enough depth or that the early game is too easy. So I think we are just about at a state where the early game cannot be made harder. And that there’s only marginal benefit to making the later microbe stage harder or more in-depth. Though that’s just my interpretation of the data.

I hardly ever use lysosomes with amylase (the default enzyme). I don’t really see the extra resources gained from digestion as being worth the extra cost for adding the part - though of course it’s better than just adding an extra hex of cytoplasm.

I use lysosomes with chitinase or cellulase when I keep trying to engulf cells with chitin or cellulose. Recently, though, I’ve found that I rarely need either, as a large proportion of cells I encounter are only recently split from my species.

Wanted to share a bit of my balancing “philosophy” since I’ve been making some tweaks to the game recently, and want to make a few more tweaks. Though my balancing tweaks are making Thrive a bit more challenging, and though I do think Thrive should be a bit more challenging, my changes are not “let’s just make the game harder”.

By sharing this, others can provide adequate feedback, point out inconsistencies in any proposal I make to my approach, and question my approach overall.

Difficulty

I think difficulty can be very broadly thought of as the margin of error a game provides a player. Easier difficulties have a higher margin of error, allowing less tight performance a chance to progress, and harder difficulties have a lower margin of error, requiring more strict and optimized performance to succeed. A well-balanced game finds a Goldilocks Zone for their margin of error (or otherwise provides their players the ability to finetune their own margins).

If we look at this incredible Microsoft Paint visual I drew below: the left end represents an extremely relaxed MOE (margin of error), and the right end represents an extremely tight/tense MOE. Red represents where I felt we were, and green represents the Goldilocks Area.

[image]

Because I felt that the MOE was too safe beforehand, I think players weren’t being driven to really experiment enough, or otherwise optimize their cells. Gameplay

This isn’t to say that the game wasn’t challenging at all - atleast for new players, learning the mechanics presents enough of a challenge already. But atleast in more experienced players, there was a repeated ethos of the game not encouraging them to really stick out their neck on a new strategy. I think that is partially why a decent number of players generally play when a release is out, and then wait until the next release to have another playthrough. Another part of that is also because we have some older members of the community who have tried this game for years at this point.

The Essence: Planning vs. “Stumbling Into” Playstyles

Explained in the simplest terms, my approach to balancing recently is a desire to make playstyles require more intentional strategy and adaptations as opposed to just “stumbling” into a playstyle. In other words - if there is a playstyle a player wants, they should alter their organism in pursuit of that playstyle.

I think very lenient balancing puts Thrive in a place where any playstyle is possible and easily accessible, but as a result, there is no meaningful difference in the effectiveness of different playstyles. And thus, no reason to choose a particular playstyle over the other, or minimized differences in part decisionmaking between different builds.

If balancing is too permissive, players have a really high margin of error. A high margin of error isn’t bad at all; the opposite extreme, a low margin of error, also stifles diversity in playstyles to a similar extent because only one or two metas exist. But still, a really high margin of error in Thrive’s case means that there are very few situations where a particular playstyle won’t work. And if the game is balanced to ensure that all playstyles work at all times, then there isn’t much of an incentive for a player to experiment besides their own whims.

I’m obviously broadly brushing over a really lenient Thrive experience - we do have mechanics, like an evolving atmosphere, which means that certain metabolisms won’t be effective at certain times. But overall, I do think it is true that certain balancing items were straying too far in the safe zone.

If we tighten things up to really nail our margin of error, it won’t always be a whim which prompts players to play a certain way - it will be hardnosed strategic decisions, incentivized by the circumstances of a particular save, which make players play a particular way. And if metabolisms are tightened up to have a tight enough MOE to require decisionmaking, but a loose enough MOE to allow creativity, I think we’re in a pretty good place with the Microbe Stage.

Constraints, or the Lack Of Them

I’ve talked about constraints before - in essence, a constraint is a dynamic of sorts which limits players to a specific bound, forcing them to solve a problem contained within that bound as opposed to being able to bring in other tools. I’ve also shared my opinion that I think Thrive is a bit too easy on its constraints: we offer a pretty diverse toolkit, but the problems we present are universal and as a result make these tools behave similarly.

Constraints can obviously be introduced by new mechanics - size-related costs and the environmental events being worked on by Patryk are solid - but even tweaking .jsons can provide constraints. If we tilt the balance towards presenting engineering design problems for the player to solve which vary between metabolisms for example, then I think we’ll end up with some very solid depth for our first stage. As opposed to how it is now, where different metabolisms are just a tool to solve the universal design constraint: ATP.

So in the most recent balancing changes: this is why I made oxygen accumulate more slowly. The constraint the player must face in the Microbe Stage is the timing of oxygen, as well as some slight inconsistency in how powerful it might be on the onset of oxygenation. This is also why I made iron burn quicker: the constraint there is iron’s lifespan, which necessitates a playstyle surrounded around chunks.

Now, going back to the MOE discussion - I don’t think every single thing needs to be tightened excessively. Iron chemolithotrophy has received feedback that the burn rate is a bit too high for example, so that’s a valve I will loosen. But I still think iron should be characterized by its shorter lifespan - thus, my next balance pass will increase iron’s lifespan, but I will try that in combination with altering the lifespan of iron chunks themselves, so that players are forced to be mindful of how close they are to a chunk at all times.

The Summary

So in essence - rather than it being just a thought that the game is “too easy”, my approach is this:

  1. Figure Out the Constraint. What design problem are we tasking the player to solve?
  2. Is the MOE too tight/too loose? If it’s too tight, loosen the constraint. If it’s too loose, tighten the constraint.

Next Steps

I will try to post something describing the next balance item I am trying to tackle since a big focus of the next few releases is polished balancing. Here are some items I’m going to be looking at.

I will clarify that:

  1. My first goal will be getting the thermosynthesis changes which I obliterated in.
  2. There are more balancing changes we can do, but my focus for 1.0 is on metabolisms since it would have the most implications on balancing.

Iron

I’m hesitant on responding to feedback immediately after a patch since players still are getting used to new strategies, but I have seen atleast three different players mentioning that iron is burning a bit too quick currently. So I think I’m going to try and reduce the burn rate a bit, and will alter chunk behavior in the way I described above. The constraint here we should try to go for is tying activity to chunks.

Photosynthesis

The constraint here is the day-and-night cycle. We have received a bit of feedback stating that creating mixotrophic organisms is rather easy. I’m not completely allergic mixotrophic organisms - they actually are a very common strategy in eukaryotes - but we atleast want the constraint for photosynthesis to be more effective here. I’m going to look at potentially nerfing glucose output here.

1 Like

We only get filtered feedback from the tiny fraction of players who decide to provide feedback on the game. But I think that despite having really good goals, what actually really matters is how the players perceive the changes. And also how new players will encounter it.

As it stands the game basically has no guidance on working game strategies other than “stay small, go to the surface to photosynthesise”. (or at least in terms of explicit tutorials)

If everything is nerfed then there won’t be any easy strategies left, and I absolutely think we must have at least one easy strategy that can be explained to any player posting on the Steam discussions or reddit how they can’t just survive in the game and what they should do.

If we remove all the easy strategies in order to make all gameplay styles require planning / or be interesting, then we are effectively cutting off all players who are not (yet) so involved in Thrive that they can come up with the more complex strategies.

2 Likes

Drawing inspiration from real life: I don’t think at the baseline surviving on an abundant resource is generally super difficult. (Barring extreme circumstances). Plant a tree in fenced-off fertile soil, and it’s not exactly going to be in a mortal struggle for reproduction. Same thing with dropping some bacteria on a sterile petridish with suitable (not even perfect) growth media.

Rather the difficulties real life experiences is that there is usually other life trying to use the same resources (and they may be better at doing so), and while you’re doing that something else is probably trying to eat you at the same time.

If you feel that you can just pursue your food source at leisure, then that consistent predatory pressure is something that might be a bit missing in the current version of Thrive? And I don’t see it mentioned here.

This loops me back around again to the idea that there should be a default difficulty setting that has easy outs like this, with a harder difficulty setting that really requires you to effectively use the tools the game gives you.

1 Like

That is a very good point, and I do agree with you that an element of competition is missing (and could be improved upon via auto-evo).

At the same time, I do think an element of the period the Microbe Stage is representing in life’s history is the question of life adapting to utilize the resources it has optimally. Once organisms are optimized and well-established, those familiar ecological processes begin to take place - though at the onset, the immediate onus is configuring the structures needed to create and utilize energy efficiently. Something that comes to mind is photosynthesis switching from utilizing sulfur, to utilizing water.

We don’t have the level of detail needed to represent, say, the creation of proteins and the customization of how compounds are utilized. But we can represent the “figuring out how things work” phase of early life by presenting some constraints to these forms of metabolism, and presenting design analogies to the player which asks them to deal with the design problems life faced early on. In the future stages, I think the onus is much more on external competition.

I agree with Rathalos here in that us having multiple difficulties helps us. Many games, like Civ or KSP, inherently have a difficult onboarding process, but remedy this by offering difficulty presets. It is common practice for games like this (I’m more familiar with Civ myself atleast) for there to be a big portion of the community that prefers easier difficulties, and a big portion of the community that prefers the harder difficulties. If things are balanced to be too difficult for new players, we can explicitly recommend Easy Mode.

For Civ, I think basically everyone gets introduced to the game on a very easy game mode, and then they take a leap towards harder difficulties. These easy game modes are balanced to be extremely easy to win - the challenge for new players here is instead learning how to manage all the information and menus.

1 Like

A big problem is that the Thrive gameplay or simulation is not setup to handle different tweak values based on the current difficulty level. So without major changes all over, Thrive can only have the features balanced around a “normal” difficulty level, which I think is much closer to “easy” than “hard” because otherwise it will be just a vertical wall on a learning curve for new players.

1 Like

Well, the way I’m looking at it (which might be a bit too simple):

  • What’s the ultimate point of all these metabolisms? ATP.
  • How does not producing enough ATP kill a player? Osmoregulation cost.
  • Do we already have a difficulty setting for osmoregulation cost? Yes.

So as far as I’m concerned, we balance the metabolism to be balanced relative to each other. Then, we adjust osmoregulation cost further down on easy difficulty (which should possibly become the default for new players/with the tutorial). This does not just mean you need less ATP-producing organelles, but you also deplete resources more slowly.

Hopefully that means there’s not just one easy gameplay strategy, they’re all easy.

The only truly difficult implementation I could see here is if osmoregulation cost is not enough and we need to extend it to movement cost.

Of course there’s obvious downsides here. It makes things like cell walls less useful at this difficulty for example. But I think it’s worth it in order to give a uniformly easier experience without making specific metabolisms into “the easy option”.

3 Likes