Restart the Game Scene Every Spawn

Gameplay
1

I am continuing this quest of mine.

There are multiple gigantic reasons why having such direct continuity between generations presents us with problems.

  • Environmental Changes Not Actually Being Represented - Especially with compounds varying a lot more dramatically now, each new generation can bring a dramatically new experience with resource allocation. However, because pre-existing clouds are present, these compound changes really do not manifest at all.
  • Auto-Evo Changes Not Actually Being Represented - Species which should be extinct, or otherwise experienced significant changes in population, will still be present at their prior level. This can very easily lead to confusion, and adds an inherent level of inaccuracy to our GUI reports.
  • Improper Evaluation of Build Performance - If your organism finds a suitable spot (intense cloud saturation or a giant chunk as an example), there really is no need for your organism to prove itself as successful in a new world. Instead of the organism proving fitness in a new environment, conditions of an environment from 100,000,000 years before will still carry them forward. To the point of even having the same engulfed organisms being digested.
  • Logical Implications - Beyond gameplay implications, it really does make zero sense for us to be an evolution game which supposes that geological eras can pass with 0 change in the environment of your organism. Ancestors from millions of generations before should also not have such a direct impact on their great-great-great-great x100,000 descendants.

I will also say that the existing justification for having this continuity isn’t necessarily a strong one. Proponents of this direct continuity argue that it serves as a get-out-of-a-rut card for a player who might find themselves in a dangerous situation.

  • That’s a real evolutionary strategy. See mayflies and silk moths, who literally do not need to have mouth pieces as an adult because they reproduce so quickly.
  • Evaluation of build success should be contained to the gameplay loop - in other words, the test pass condition is just reproduction. Having it go beyond reproduction bleeds evaluation of an individual build’s success across multiple generations, which isn’t ideal.

There might be some loss of the feel of a living world if there is a fresh state every spawn since spawns technically restart - but isn’t there a way to quickly run through simulation steps right before the player spawns that isn’t very demanding?

4 Likes
2

The only way I’d ever personally accept this if we have a setting in the game where the player can select two options:

  • Keep immediate surroundings on generation change
  • Clear the world each time

And the second option has to also stay consistent in macroscopic and aware, i.e. anything like nests etc. the player has made has to be cleared for consistency as it makes no sense for such structures to remain if everything is to be cleared.

For that reason I’m extremely heavily in favour of making the default option to keep immediate surroundings.

3

I’d argue that having continuity in the macroscopic stages is an even more significant issue than continuity in the microscopic stages (except in the society stages, where it makes sense).

  • Instead of an abstract microscopic representation of a larger world, we are portraying that larger world. So geography and physical terrain are much more clearly defined. Therefore, players will be MUCH more aware that the world around them has not changed, which can be really confusing if we are arguing that each generation represents a significant step forward in time.
  • The biological macroscopic stages are a time where climates, environments, the world itself will change dramatically. Patches could switch from arid, sparse environments to humid, jungly terrain over multiple generations. If we are intending to simulate these changes in the world, a significant aspect of evolution, how could we possibly have continuity?

With cases like a colony or a nest (larger question of that representing a very basal “society stage”, but that’s besides the point), I would assume that the better solution is to have a player spawn into an existing colony/nest in a new world if they have certain adaptations dependent on those structures.

Besides that - I somewhat get this issue for things like ant societies, where they must spawn into an existing ant colony to really make their adaptations make sense (and migration zones for some nesting species). But most nesting structures are inherently more temporary in nature, not lasting in location or structure for more than a few years. It’s not like an organism inhabits the same nest of its forebears from 2 million years before; for many organisms, nest-building is a part of courtship and reproduction in their own individual life story.

Ultimately, I think macroscopic must have an inherent level of discontinuity - as opposed to the microscopic level, where there is some abstraction.

4 Likes
4

I have to (almost) completely agree with Deus. I don’t think the current situation is immersive or logical (Really, one “patch” millions of years apart should be considered just as separate as two adjacent patches ), and I don’t think it provides a better gameplay experience than the alternative.
An alternative that you can try out right now by just switching to a different patch every generation. I think the fact that switching back and forth between two identical patches is such a different experience is another illogical inconsistency caused by the current design. That’s also why this:

Is in my opinion a completely invalid argument: you can completely avoid unwanted continuity by just moving to a different patch. So the between-generations continuity is never a challenge and can only be a benefit once you figure this trick out.

Another inconsistency is that when initially spawning in, the patch does not get reset. But if you die, it does. So the game isn’t even consistent with “carrying things over”. It feels like a bug.

I want to emphasise that this is very bad for gameplay. A big topic in multicellular stage design is environmental changes. If iron disappears from a patch, an iron dependent player should be forced to find a place to live. Right now they can:

  • Just hang out with the already spawned iron chunk that lasts 5 generations.
  • Simultaneously be confused about why their population is dropping. (hint: it’s because patch persistence breaks the link between world simulation and the player environment)

It’s also handled inconsistently, because atmospheric compounds and environmental effects do change, while compound clouds and objects do not. I am convinced that this should be seen as a bug.

The reason I said I almost fully agree is that I am not sure if you are including carried over components in the “break in continuit”. That’s where I would draw a line. I think we can reasonably say “If you, the player, take the time to fill up your storage before reproducing, we can assume the parent of your next generation individual did the same.”

This preserves the gameplay of “preparing for the next generation” in a believable way, without preserving the environment.

Extending the mechanical logical from Microbe/Multicellular into Macroscopic, option 1 would regularly place you in a forest of trees that are supposed to be extinct, even if the area is supposed to have become a desert at this point. I personally find that to be a completely unacceptable scenario, that just exposes how much of a problem the current mechanics are. People just don’t notice it because the microscopic world is unfamiliar.

I also have to agree with Deus on the nests here: They should be cleared. If nest-building is a required part of the life cycle for this species, then they should be making a new nest every generation. If you’re talking about the “parental” nest, then we can just spawn one in because the parents would obviously have built one. If nest building is optional, then we could even apply the same logic as I mentioned for stored compounds: If the player built a nest for their offspring, the next generation parents are assumed to have done the same.

Option 2 is in my opinion the clearly correct option, so I would make it the default, if we even want to make it an option. (I would not).

I think strictly speaking it’s as much of a realism problem in the microscopic stages, players are just less familiar with the microscopic environment so they don’t notice as much.

And with that, I feel like I have to give the reminder: confusion about “old” species being around as well as compounds that should be gone is not something we never hear from the community. It’s a fairly common complaint among other things.

And just to re-iterate: patch continuity can be avoided at will for example by switching back and forth between two similar patches. The view that this mechanic makes players not use reproduction as an escape method is demonstrably incorrect. It’s an easily abusable mechanic that lets players circumvent environmental changes without creating any interesting challenges.

1 Like
5

Reading some of the discussion also on Discord and thinking on this for a while, I think I have realized something guide big.

Ever since joining Thrive like back in 2016 it has been the case that the player returns to the same environment after editing. Meaning that (at least for me) every single game design decision and line of code I’ve written for Thrive has been made with that assumption baked in.

And it seems that we have been working on a slightly different game all this time? Because to me it has never been a consideration, even in the past year, that any game design decision was made without the assumption of the continuous environment.

To even start to grasp the magnitude of the required change to swap to not keeping the immediate environment someone would have to go through every single last line of code in Thrive and categorize it either as not being impacted by this design decision or being impacted. And then to switch this design (that’s been in use since like the first ever Thrive versions) every single impacted area of the game would need to be adjusted.

This is such a huge amount of work that I cannot even fathom it. It would be only a bit easier than remaking all of the gameplay logic again (this is a slight exaggeration, but the amount of work would be within a magnitude the same).

So considering that the microbe stage was declared done, and I said I would allow other people to work on it if they did all of the work and didn’t break things so hard that I have to jump back in and fix it, this would require a separate task force to tackle this momentous task.

Like I said before, the only concession I’m willing to make here is that it could be made as a toggle that defaults to the old way. This is the only realistic way in terms of effort I see working: starting small with the option disabled and working for a few years on all the problems it causes when turned on, it might then some day be plausible to make it the default.

1 Like
6

I am sorry to point this out again, but when literally taken this is plainly not correct.

This has only been the case when not moving to a new patch. Which means your statement is only accurate for a sub-set of the time in any playthrough. Sometimes a large sub-set, but in other playthrough’s never.

Which is probably a factor in:

I very rarely stay in the same patch for more than a generation, so certainly it is possible for people to have been playing a different Thrive all this time.

If you have always been assuming a continuous environment, then the game you were imaging is not how Thrive actually works for everyone.

I find this very odd, because as said, the majority of my time in Thrive, I am not at all dealing with a continuous environment. So the presence of a continuous environment is not some stable foundational assumption, it is in fact a fracture line running right through the middle of the design, with gameplay being quite different in a binary way based on whether you stay in one patch or move to another patch (or stay in one patch but die and respawn. Do you see how inconsistent this is?).

I don’t think that’s a good foundation for the design. Every game mechanic has to work without the assumption of a continuous environment, because that’s what happens when you die or move patches. But because of the continuous environment when not moving patches (and before you die), everything needs to be balanced for that situation as well.

I have to admit that my first impression here is “doubt”, considering I can mimic what we are suggesting simply by moving to a different patch with the same conditions, or hitting the suicide button after entering.

I can’t speak to the code, but from a game design perspective, we’ve already been designing for a non-continuous environment because we don’t have a choice, that is how the game often functions.

This is fine, we cannot keep getting stuck on Microbe Stage mechanics. But we need to take a hard look at this for the Macroscopic/Aware environment. Because if we still try to implement a continuous environment there, I will predict for you now that that will produce obviously and hilariously broken outcomes like the trees in the desert I mentioned before.

3 Likes
7

Moving to different patches was a feature that was based on a quick look “only” added in 2019 (Devblog #21: The Patch Patch - Revolutionary Games Studio). So any reference to switching patches clearing the world can only be the case after 2019. I was unable to easily play 0.4.2 to confirm what happens in that version upon switching patches to the existing game objects.

I’m of course aware of the world resetting on patch change.

Okay, we are playing a different game then.

I only switch patches when I have to and once I find the one I’m looking for, I won’t switch patches again.

Then it’s a huge problem because during the past few years I’ve been doing most of the pull request and new feature testing (not to totally discount public test builds and releases), and I’ve absolutely not been testing to make sure all features work well with the assumption that the player is always switching patches. Or that switching patches is the main way to play the game.

I was not aware…

And as for respawning, to me it’s simply a feature to retry if you weren’t successful in your previous life.

I mean the world environment around the player would also clearly not be allowed to change so any existing trees would be in a “circle” of forest that would then be surrounded by dessert. Same as the microbe stage: the immediate player surroundings would be preserved so that there’s continuity in the gameplay and only by new stuff spawning / moving would the player see the new state of the world. Same as in microbe stage. Which is again why I think the same setting should apply to all stages.

8

I also tend to agree with Deus and Rathalos here. While my gameplay is less about switching patches continuously, it does make sense to restart every spawn. This especially makes sense in the macroscopic stages, where the environment will probably have to change due to climate and geology, more than only evolution. So persistence like nests or existing plants or other features would be completely broken. I think nobody stumbled into this issue particularly hard in the microbe stages, as the microscopic environment appears to be less volatile than how much a macroscopic environment would.

3 Likes
9

Ah, I was unaware of the timing, but this means the “ever since” statement ends whenever patch switching started, and/or whenever that started resetting the environment. From that point on, “continuous environment” was not a valid assumption.

I wasn’t thinking you were not aware, but from your statement it appeared you were not aware of how major of a change that is for basic game design assumptions. It means that a continuous environment is an invalid assumption.

Switching patches is not the “main way to play the game”, but it is a major part of the game. Even just moving from vents to surface can easily be 5 generations. Staying in one patch is also not the “main way to play the game”. Both happen, both need to be balanced for.

I mean, it is? But it also resets the environment to what the world simulation actually thinks it should be, which can be quite jarring if you’ve been playing in an environment that’s lagged significantly behind. (For example, iron does not exist anymore).

I see. I do think that will look extremely odd. But you’re probably right that it would be better to have it function the same across stages than having it be different in behaviour across stages.

2 Likes