0.5.7 is out now. You probably know the drill by now, post any feedback you have about the new release in this thread if your feedback doesn’t warrant being its own thread.
Perhaps if there is a way to turn off (not sure what the correct word would be) the Chemoreceptor either manually when the cell fills up a specified % of the compound store or manually via an assigned hotkey, this allows cells with multiple chemoreceptors to turn off ones for a compounds when it can no longer store any of that compound due to the storage being full.
Way ahead of you, there’s an open issue already:
Times are busy for me right now, but I got around to playing Thrive and had a good time. Here’s some notes I jotted down.
- Very well-done and informative timeline. Icons are informative and it helps us understand exactly what is going on in a way that is very pleasing to the eye. It would help if the dates are more clarified in terms of chronology, as that confused me at first. We’re used to thinking about prehistory in terms of how long ago an event was from the moment we currently live in (we say the Cretaceous Extinction happened “66 million years ago”, not 3,000,000,000 billion years “after Earth’s formation”). But other than that, amazing iteration. Can’t wait to see it fleshed out more.
- This isn’t necessarily relevant to the current update, but the layout of organelle descriptions is sweet and that was a detail I wanted to point out. A nice division between background info, immediately important info, and processes outlined for each part.
- Modify function for chemoreceptor is great. Really intuitive design that I immediately grasped. I wonder why a player would prefer a smaller search radius over a bigger one, but that’s something organelle upgrades and the such will deal with. Slap some MP on it, but great introduction of a fundamental concept for Thrive. Excited for the future.
- I like the random tips in the help menu and the enthusiasm behind them. More “easter eggs” or fun bits of the game would be appreciated to demonstrate how much the developers love what they’re doing.
- Chemoreceptors were an incredible add, as they help make living in remote patches much more feasible. A whole new world between the vents and the surface was opened up, as you can now purposefully live in every patch. You no longer are swimming aimlessly hoping for luck, so living in patches like the open ocean, ocean floor, cave, etc. is not engaging. Hypes me up for the future to see what chemoreceptors allows auto-evo and players to do.
- Great job with the graphs update, new icons and magnification is tremendous.
- Also, even more praise for the chemoreceptors – it’s made becoming a non-photosynthesizing eukaryote much more feasible. Now I can reliably track down important resources and can focus on using size as a benefit instead, so now players are much more incentivized to use size as a genuine predation strategy.
- One again, great variety shown by auto-evo.
- I wish the timeline could have the ability to toggle between specific patches instead of just between global and local. Global should still be an option, but it’s useful to know what is going on in other patches as well.
- Auto-Evo info still difficult to look through (previously discussed).
- Chemoreceptor lines are pretty difficult to distinguish because of how thin they are, and can be difficult for colorblind individuals to use (the entire game can use better accessibility actually). Making the lines thicker and having more customization – for example, dashes or dots and such composing the lines – can add a fun layer of customization and can help accessibility.
- The timeline rarely says that a species disappears from a patch even when it doesn’t.
- Species list can use more division besides living and extinct in the graph-filter options. “Species in Current Patch” as an example, as I forget which species I am competing with when I tinker with the graphs.
- One time, I placed the pilus on what should be the front of my cell but was spawned in with a pilus on my rear next to my flagella.
- Lag is still significant in the late game to the point of really detracting from understanding what exactly is going on.
- Does the Auto-Evo prediction tell you what your cell population would be for your next trip to the editor or what your cell population is expected to rest around in a given patch given your morphology. I think the latter would be a more indicative piece of information regarding your cell’s population if not so. This came to my mind because even though cytoplasm is recommended in the tutorial for the first trip to the editor, adding a cytoplasm resulted in the auto-evo prediction moving downward, which might spook some new players.
- Along the lines of the above, a tutorial pop-up saying something along the lines of “Don’t be afraid if an adaptation lowers your predicted auto-evo population baseline even if you know said adaptation is beneficial. Although a large population is generally better than a small population, the species with the highest population is not necessarily more evolutionary successful than the species with the second highest population in the long-term. For example, larger organisms tend to have smaller populations due to added reproductive costs, but make up this loss of population with the benefits of size”.
- Perhaps have preset colors for the chemoreceptors that are the colors of specific environmental compounds.
- Having Pause Game (I don’t mean bring up the menu, I mean freeze the game in place) be linked to a hotkey would be very helpful, especially when we are more able to inspect other cells in our environment in the future.
- Now that they are in the game, looking into ways to help balance chemoreceptors would be a good idea. It works amazingly, but basically provides 1000 benefits and almost no cost right now – or atleast, you are not restricted in upgrading the function of chemoreceptors. Once again, organelle upgrades will help with this.
Thanks for the feedback.
Here’s replies to some of the points:
You actually can. The up and down arrows in a box next to the patch name on the report tab actually allows you to change the patch that is shown in the graphs and timeline. I realize there really is no tutorial pointing this out, but then again I’m not sure this is important enough to bother the player.
Anyone have any ideas how this could be made visually more apparent to the player?
The line thickness is a technical limitation. I’ve already opened an issue about writing our own line drawing algorithm to get around that:
I think this is because it might not take the external effects into account. For example it will claim the player goes extinct each editor cycle if auto-evo thinks the player should have 0 population.
Isn’t that the division currently? There’s species that are currently in the patch and ones that are not (extinct). I think at least that’s how it is. @KasterisK probably knows this.
Perhaps you ended up with the pilus being at 0,0 coordinates which uses a predetermined angle for it, which because the feature was originally made for the flagellum, points it backwards.
I can’t remember if we discussed at some point that the angle the player specifically used, could bias that selection here or not. Does anyone remember? I think opening an issue for this would be worth it.
Did you press F3 to check how much FPS were you getting? We’ve kind of tweaked it so that around 30 FPS should be about the minimum any player should usually get. I know that isn’t a ton, but I haven’t wanted to set the cell limit ridiculously low because I fear that would reduce the vibrancy of the game too much. There are a few lag spike and general entity performance improvements I’ve made for 0.5.8, which made us confident enough to increase the entity limit a bit again, but once again the target of having “playable” FPS for most people is hopefully reached (we got some testing done).
Our auto-evo algorithm doesn’t (usually) result in stable populations. Most work I’ve seen around this doesn’t result in stable populations, but ones that oscillate with multiple species interacting. So how long should we run auto-evo for to get an average?
Also it would be a bit of a lie to just take the average from multiple steps as in normal gameplay the other cells adapt to the player.
Other players seem quite ahead of you already in ignoring the auto-evo prediction. I did write in one tutorial that it is not the end of the world if you get a population drop from it.
I suppose it would. I’ll open an easy issue about this as we have plenty of key handling code examples etc. already so it should be doable for a new person:
Edit: I’ve opened an issue about the pilus angle:
I’m late, but better late than never.
The Timeline should be the default panel instead of the Autoevo, as it’s shown in a way that it’s easier to understand.
I still think that after reproducing, the cell should spawn to another place in the patch.
Is the auto-evo prediction going to be on by default? I’m uneasy about the player knowing how the AI will penalize the species. It feels like cheating.
Maybe when I click on a species in the Species Population graph, the timeline should highlight it instead of the player species.
Otherwise a good idea but it doesn’t actually show you all the info…
Feedback from other players seems to be positive about this, it’s like now you are no longer playing blindfolded.
It’s always possible to make some game element optional but I don’t think it’s worth even for me to spend the time to open a github issue about this, but I’ll accept a well made PR adding the option to hide the auto-evo prediction.
I think chloroplasts and mitochondria need to be more balanced, since they are in a perpetual loop, and you just need some oxygen and carbon to start the process, so it doesn’t make sense for them to be affected by the oxygen and co2 levels as much as they are (when they are implemented)
Bit late to be giving feedback on 0.5.7 maybe, with 0.5.8 just around the corner, but I’m gonna do it anyway. I feel like as one of the more ‘meta aware’ people on the forums I still have useful things to add.
I am pleased to say that - for the first time since I started playing (0.3.2) - the game’s meta now has two equally viable competing strategies.
- The Fast Way: Stay as one hex. Change to something like a chemosynthesising proteins if you’re gonna be alive long enough for glucose despawning to matter. If you’re gonna speedrun just stay as cytoplasm.
- The Easy Way: Get enough hexes to support a chemoreceptor and use that to get enough resources to support your many hexes. For the exact type of support you can try chemosynthesis, or (imo the best) move to the surface to get photosynthesis. I haven’t tried doing it with iron yet.
I think this can be attributed to the very clever design decision to make chemoreceptors, by far the most OP organelle, too energy-intensive for a small bacterium to support.
Interestingly, my Auto-Evo algorithm seems to agree that chemoreception strat can compete with 1-hex. That’s weird, because the ingame auto-evo clearly doesn’t; 1-hex is still the only way to get a high population by the ingame auto-evo. Maybe I should start work on my auto-evo again…
Other than that I unfortunately find that predation is still not a useful strategy, and neither is anything that requires a lot of size (eukaryotes, for instance, are bad both in game and in auto-evo). I would recommend:
Find a way for predators to get more energy directly from their kill. In order to kill a resisting prey in Thrive you have to expend a tremendous amount of energy chasing him and producing toxin. Not only that, but once the chase is over you’ll find yourself 10 miles from the closes compound cloud. This makes predation suicidal. It’s simply not worth it to end up with a net loss of energy, all for some ammonia and phosphates you could have easily gotten from a cloud.
Put species in the list of things you can detect with a chemoreceptor. Chemoreceptors have finally made small multihex designs viable for autotrophs, but predators still have to rely on pure luck to find their prey.
Make Health scale with the size of the microbe. The larger a microbe is, the slower it is, the more food it needs, and the longer it takes to reproduce. What does it get in return? Nothing as far as I can tell. Making a 10-hex microbe just as easy to kill as a 1-hex one is unrealistic and really just adds insult to injury. In a game where players can design any creature they want, it’s important to put thought into the more abstract dimensions of design, instead of just basing everything on all-or-nothing divisons like prokaryote vs eukaryote. The rules of the game should make an equal amount of sense at 1-hex, 100 hexes, or a million hexes.
Multiplying Health by something like: 0.5 * sqrt(hexes), would be a good start.
Then some non-meta related stuff:
I’ve been thinking for a while about how Thrive’s environment doesn’t really seem to feel like a real space. I know that in theory I am in a coherent space, that I should be able to go back to clouds I’ve already visited, that other microbes should continue to exist when off-screen, etc.
But in practice I have a very hard time spatially orienting myself. It feels more like a mini-game (think flappy bird, or agar.io), where things are just sort of randomly put on your path. I think this comes down to 3 factors:
Clouds continue to spawn even in areas that are already explored. While object permanence is technically a thing in Thrive, it doesn’t feel that way: everytime you come back to an area there are new clouds there that couldn’t have logically gotten there. There is always this sense that some invisible hand is putting down the clouds just outside your vision.
The difference between a newly spawned cloud and one that has been there for a while is clearly visible. Newly spawned clouds are like separate circular droplets, while old clouds all blend together into this compound soup.
This prevents the world from feeling like a real lived-in thing. When I swim around in Thrive, I don’t feel like I’m exploring a world that already existed before I got there: I know what a Thrive world looks like when it’s existed for a while, and this is not it. Swimming around isn’t exploring, it’s just a little ritual you do to make more clouds spawn.
The environment is very uniform. I know that realistically, these creatures are very small, but I still feel like there’s a distinct lack of larger scale structures. It would help a lot if you had to actually remember different places and where they were.
For example, maybe there could be a stone in the water, where lots of iron chunks spawn. Then somewhere else there’s, idunno, a droplet of dogpiss with lots of ammonia. Then if you wanted to get ammonia, instead of saying: “I know, I’ll swim in circles for 15 minutes straight until the game decides to spawn an ammonia cloud.”, you’d go: “I know , right now I’m by the stone, but a 2 minute swim to the south of here there’s that place with all the ammonia. Let’s go there!”
Anyway, those are just some thoughts for things that could be helpful to work on next. Overall I’d say this update has been one of the best for the meta of the game, but we still have a long way to go to make it feel like actual microbes in an ecosystem instead of just agar.io on steroids.
Maybe you could have the auto evo level with the player, without eliminating less complex organisms. You could do this using a complexity counter. Basically the auto evo will try to have the average complexity of organisms be around your complexity. (this can be adjusted higher or lower with a wider or narrower allowance depending on your needs) Allowing more complex life to form in a way so that the player isn’t frustrated at how hard it is to survive. Another way you could do it is you could have the auto evo naturally trend toward more complex organisms, increasing the realism since your species is not the first to evolve to higher complexity, and if you are, then the other species are not far behind. You could also have species already in the game (like in different paches and stages) that will evolve around you when you get to certain stages or pach maps, assuming you’ve already reached a certain complexity level by then.