With 0.7.1 out now, here’s the thread to post feedback about the new release.
Patch notes are here:
Our devblog will be posted later today.
I would like to start out with what I like, then what could be improved;
What I really like:
- I like the the new bloom effect. It really does make the compounds pop and stand out more against the background. I think it even makes the organisms slightly brighter.
- Starting out with an anoxic world plays differently than previous versions. Normally, I would start out by adding Metabolosomes immediately, but now I cannot do that. I have not tried yet tired to play this version without organelle unlocks, but it seems that the game is irreversibly different in terms of progression. With the oxygen dependent organelles now locked until oxygen begins to appear in the world, the game is now playing out more like how life started on Earth BYA.
- Since I enabled experimental features, I have noticed that most compounds decrease in the world, especially Ammonia and Phosphate. I am not sure if the rate of decrease is faster than without experimental features turned on, so I am unable to comment on that aspect. Also, since I am playing on large worlds, I do not stay in one patch beyond one turn. Perhaps I will see a better impact of dynamic compounds on future playthroughs.
Possible Bugs:
- I have noticed that organisms containing Rusticyanin, with experimental features turned on, they tend to use the Sidreopore in the same way as toxins. They fire at me and other non-self organisms using the Sidreopore, and I noticed this as well in the previous version. I am not sure if this is intended behavior.
- Even though I play with Chitin membrane, which means I cannot be digested, Prokaryotic organisms still want to eat me. Correct me if I am wrong, but I thought the AI will not go after organisms that have non-digestible membranes. I normally play as the smallest organism possible, so I wonder if there is still some conflict with Predation Behavior and Membrane type in the game.
- After the 13th turn, when the game was saving for the fourteenth turn, my game bugged out. While saving on the 14th generation, the game automatically opened the Thriveopedia for no reason. I had to close the Thriveopedia, and then the autosaving continued. I will send you the log and my save later.
What could be improved/added to the game:
- It would be nice to have a way to change the organelle reproduction order. I know that is a planned feature for 0.7.X, which would be useful in the game. I cannot tell you how many times my organisms in previous versions, despite being ATP positive before duplication, start dying due to ATP-dependent Eukaryotic organelles replicating before the ATP producing organelles.
- More organelles in the game besides the planned Bioluminescence and Radiotrophs. It would be nice to make ATP from light, allowing for other ways to utilize light without being a “Plant”. I am not even the first person to suggest adding a version of Rhodosphin: https://community.revolutionarygamesstudio.com/t/ideas-for-the-microbe-stage-put-your-ideas-in-this-thread/1885/81
However, I also understand how much more work it would be for developers to add every suggested organelle, assuming it is even feasible. - Will there be a patch for 0.7.1.0? I noticed that there was no patch for 0.7.0.0. I am not surprised, since 0.7.0.0 was very polished. I just want to make sure I am not missing any update. Everyone appreciates the time the developers take to improve the game.
2 Likes
Oxygen and CO2 both stay so low so that rusticyanin, photosynthesis, and metabolosomes all became unviable and I’m forced to became a predator with hydrogenase
The 3d membranes also look nice
4 Likes
Looks great! UI update looks good, bloom is gorgeous. I feel perhaps some ability to see where things are going in terms of oxygen could help (and uhh, pushing the player to pay attention to them. I kinda ignored the graphs and looked at the numbers of if I was on balance able to move around and stuff). I wasn’t photosynthesizing, but when the two glucose-to-ATP converters swapped effectiveness it felt weird having my balanced build suddenly not have enough ATP. I just leaned on iron. I’d never done a full iron play-through, and even though I don’t consider myself done yet, it’s been pretty fun! The oxygen changes happening to me without any input felt cool though, and I had some good AI experiences! Had some normal lag and a crash, posted the logs already, but pretty standard stability all things considered. One other thing: I noticed cells I was digesting showed themselves inside me, which is great and all, I think that was in previous versions, but I was a rather skinny cell, so when I ate very round or sideways cells they sometimes clipped outside my cell. Not game breaking but a bit weird.
1 Like
I think we need to include nitrogen stats (or all type of gasses in general) for the game and not just oxygen. the present its a bit of a stack because its present the total amount of the gasses and how much its have.
imagen the same stats of % oxygen on lower presser (like lets say on mars - 0.2 ATM) - we have in total much less on anything then on earth.
i think we need to address that issue because of balance for the future
2 Likes
Actually, ammonia or phosphates do not change at all without experimental features enabled.
One of the reasons why siderophore is not enabled by default is that the AI needs a lot of improvement.
This is correct. Can the AI actually digest you? They can if they have the right lysosomes. It should be impossible for AI to turn on engulf mode and try to eat something they cannot digest, but it could be that the AI just happened to try to eat something and run into you.
Does it happen every time? From the log:
ERROR: Scene instance is missing.
ERROR: at: (scene/resources/packed_scene.cpp:244)
ERROR: Failed to load Thriveopedia page Museum due to scene instantiate failure
It seems like another variant of:
or
Most likely not. At least so far no major bugs have been reported that could be quickly fixed (as the major problems again seem to be the engine scene loading failures).
Tell that to the people who have managed to get to 80% oxygen… (and broke the display in the process by causing the percentages to add to more than 100). But yeah, there’s more tweaking needed to the dynamic compounds system.
4 Likes
No. They cannot digest me. All of the AI cells are prokaryotes, as no Eukaryotes evolved in this playthrough. Being the smallest cell, all of the other AI cells specifically go after me, and then try to engulf me; if they do engulf me, when I come out from those engulfing cells due to being indigestible, they try to engulf me again.
So far, it only happened to me once. This is the first time I have encountered this bug. I will let you know if this reoccurs on other playthroughs.
1 Like
The AI sometimes decides to try to eat something if it is useless:
I always forget that point when someone tries to make a bug report about this. It would be in my opinion unrealistic if a cell didn’t sometimes try to eat something it shouldn’t.
So basically this is just a random error that isn’t related to the save or the game state. So I’m 99% sure this is the random engine load issue that I don’t think we can do anything whatsoever about it (other than there needs to be an open bug report to the engine about this).
2 Likes
Dynamic oxygen:
I love it! No major complaints found yet, but probably still needs some balancing.
It is not actually mentioned in the patch notes, but CO2 is also dynamic. Sometimes it suddenly spikes upward though, without clear explanation (no metabolosomes yet).
Graphs/reports:
Having images for everything instead of just names is great! Much more recognizable.
The added food chain screen is also something I have been waiting for. It does expose something weird though: Each food source can only have one organism connected to it? Quite often I am surviving on for example light just fine (population doing fine too), but I am not connected to any food source on the food chain screen, just being predated on.
Graphical improvements:
Looking great! Game is really coming alive now.
- The bloom makes things pop better. Was it toned down from the release candidate? It seems less distracting now.
- The distortion effect is the step in the right direction, but: right now it mainly seems to turn mostly static dots and circles into very thin squiggly lines. Perhaps the distortion pattern could be a bit more dynamic?
Miscellaneous:
- I understand why the oxygen requirement on Iron oxidation was removed (though I think it could still exist as a variant), but was the CO2 requirement always on there? This reaction does exist IRL, but it is light-dependent, and is an autotrophic process, “produces glucose”. For what you’re going for here, I recommend just cutting out the CO2 requirement. We can just assume it uses the common metabolism that oxidises iron using nitrate (which you don’t simulate).
Experimental features dynamic compounds:
I realize this is heavily work-in-progress, so while I am going to list everything that is currently not yet working properly, please don’t take this as harsh criticism.
To be frank, in it’s current state I don’t think it is properly playable. The typical pattern (using Hydrogen Sulfide as an example) is: There is a lot of H2S → An organism arrives/evolves to consume it → organism blooms to enormous population in 1/2 cycles → H2S plummets straight to 0 in one cycle → next cycle the organism goes (nearly) extinct → H2S stays at 0
Now the speed is not necessarily the problem here. The great oxidation event probably lasted ~200-400 myrs, so it’s pretty reasonable if you are planning to stick with the (rather large) 100 myr editor cycles.
- I believe part of the problem is simply that the game only takes a “static” starting compound amount as exists in the non-experimental game, and then modifies it by biological processes over time. I think it is in this case quite obvious what needs to be added: There’s a constant non-biological production of H2S . Adding that in as a factor should ensure that H2S does not crash to zero, and would also result in a steady accumulation in the patch (and possibly adjacent ones?) when nothing is there to consume it.
- For CO2 the story is a bit more complex: you almost have a neat equilibrium going on between photosynthesis and aerobic respiration, but there are “leaks” that consume CO2 without producing O2 that eventually still lead to a collapse to 0: H2S chemosynthesis and Iron oxidation. The chemosynthesis does I think lead to glucose increase, but anaerobic respiration makes the carbon disappear and never return to CO2. Iron oxidation simply annihilates CO2 for no apparent reason. The decrease to 0 of CO2 everywhere eventually essentially ends the game because all that is left is anaerobic glucose fermentation.
- Non-carbon-fixing iron oxidation that consumes CO2 is an inaccuracy that should not be in the game.
- Consider returning the carbon from fermentation back to CO2 “behind the scenes”
- IRL, the CO2 concentration is heavily buffered by geological processes on long timescales, which is how oxygen could rise from nearly 0 to 21 % of the atmosphere over time while CO2 stayed in between 0.1 and 0.01 % (even though photosynthesis would have been consuming CO2 at the same rate that it was producing O2!). For Thrive purposes, I would just put constraints on how high/low the CO2 concentration can go, preferably in an “elastic” way, with added consumption/production based on how high/low the current concentration is.
- Come to think of it, CO2 is also dynamic when experimental mode is disabled, but it does not typically drop to 0. Why is that?
- It seems like with experimental mode on, the normal glucose reduction over time is replaced by including them in the dynamic system, which is nice! Especially good to see that carbon fixation results in more free glucose in the environment as well. It does otherwise feel strangely stable over time. Some remaining glucose for scavengers (if produced by autotrophs) is fine, but right now it seems to hardly drop, even when it is the only remaining energy source being used.
- Iron seems to be essentially not effected by the dynamics system, I am not sure why. You will eventually probably want to move to a system where Iron is initially abundant everywhere, starts rapidly disappearing when oxygen arrives (keeping oxygen levels low a little bit longer) and eventually only existing in a few oxygen-poor patch types/patches where reduced iron actually still gets produced a bit.
- Regarding atmospheric gas compounds and surface patches: you probably want to simulate an actual “atmosphere” that exchanges gas compounds with the surface patches. This will both equalize the surface patches a bit (which feels more realistic) and helps act as a buffer to changes that feel to fast.
- You can see if you want to add some H2S across the world initially. I don’t think it’s entirely accurate, but Thrive does not simulate the H2, that definitely was available for chemosynthesis (still is, in hydrothermal vents too)
Overall, the choice to often represent within the game only a single compound that is actually part of a cycle might be biting you in the ass here, and you will have to do some creative accounting to approximate outcomes believable close to IRL life history.
3 Likes
(warning: poorly organized text)
Finally played it, and feels generally pretty fresh. the 3D membranes look nice (haven’t tested yet the other graphic effects), and for some reason, the oxygenation thing actually plan my cell more than just add metabolossomes everywhere, and the food chain is a HUGE thing that keeps me motivated, like “heck yeah i’m the apex predator, and i just make another species extinct! muahahahaha!!” however, i noticied an pretty big and annoying bug after going to another patch, that made me unable to see any information at all in the editor, and progress the game.
https://imgur.com/a/gfoeTbI
fortunely, reverting to a recent autosave i had reverted it, but maybe one day i won’t have an autosave saving me again.
5 Likes
Many other processes produce CO2 as well. And there’s also environmental CO2 being constantly generated by volcanism (this effect is pretty strong to support later game photosynthesisers) so if there aren’t a lot of cells to consume it this can spike the CO2 pretty quickly.
Volcanism.
That’s how miches work. Only a single species can occupy a miche at once.
This is actually one of the ways the game is made easier for the player. If a species doesn’t manage to occupy any miches, it will instantly die due to not getting any energy. Except if it is the player species, the player species is allowed to hang around and be not extinct even if the player doesn’t manage to get any miche energy sources.
I’m almost 100% sure it was not changed at all.
The speed of the distortion was adjusted to be low so that it isn’t visually too distracting, because for game readability it is terrible to have a bunch of fast movement constantly that has no gameplay purpose whatsoever (causing just extra visual noise that might look good for a little but then gets extremely tiring and making actual playing annoying).
I’m almost entirely certain that in 0.7.0 the iron process used both oxygen and CO2. I just removed the oxygen requirement to keep iron viable in 0.7.1.
I assume all of the rest of your comments are with the experimental mode on, which is a totally different approach and we’ve already gotten enough feedback on it that I think it should be basically just scrapped and a proper dynamic compounds mode made based on the photosynthesis production code (which is the mode enabled by default without experimental features). Please clarify if any of the rest of your post also applies to non-experimental mode.
Iron is contained in chunks, which is fundamentally different from cloud or environmental compounds, so handling it needs to be coded separately.
You got hit the random Godot Engine bug where game scenes stop loading correctly:
This pretty consistently gets reported for each release multiple times (so I had already posted that link in this thread).
5 Likes
Yes, I see. I was assuming the experimental dynamics system was just the same algorithm/principles applied to all of the compounds instead of just oxygen (/CO2). That’s why I was wondering why CO2 went extinct in experimental mode, but not otherwise.
I am aware of the Miche system, and that makes sense.
I do call into question that one food source = one miche. By that logic trees, moss and grass cannot exist within the same patch. Those miches should probably be split up by different requirements. I just looked it up and Iron consumption is apparently already split up into big/small chunk consumption, so I am assuming this is already in the plan.
Alright, in that case the core of my feedback remains the same: Non-autotrophic (not glucose producing) Iron oxidation that depends on CO2 is non-LAWK and biochemically unlikely.
Oh.
The experimental mode was announced on the blog, and the game settings call out to please provide feedback on the experimental features:
So I did…
I did want to make sure it was seperated from the non-experimental mode, hence the divider. So all my observations below that point are indeed solely for experimental mode. For example, as you pointed out as well, CO2 does not drop to 0 in non-experimental mode.
The remaining are then just some suggestions for when the dynamics system does include the other compounds.
5 Likes
If only Thim was here to scream about how lawns exist…
That’s a reference to:
(sorry for no timestamp but I have no clue at what point of the stream that discussion happened).
This is purely a technical thing as they are different chunk types so two miches are generated to eat them: one for each chunk type.
That’s fine, but we’ve gotten a lot of mixed feedback recently where people didn’t disclose that they were using experimental mode. So that’s a bit of a problem as you are one of the few people who actually knew about using experimental mode and provided feedback correctly. I’m just about to switch that text to be red and much more heavily worded to say like “don’t use unless you are very experienced with Thrive”.
5 Likes
Ah yes, lawns, the well-known feature of pristine, natural environments /s
4 Likes
Haha, don’t worry, I remember what you’re referencing from one of his own videos.
Well, regardless of the technical reason, it does sometimes result in two different species being linked to Iron in the food chain.
Splitting the other food sources into multiple niches by other criteria would probably make for a more accurate simulation.
If you wanted to hide it more, you could also add a setting in options menu, that enables you to see the option in the game start menu hahaha
And another comment on here: It’s probably a good idea to have this failsafe, as right now it’s often difficult to explain why another species is considered better suited for a miche. And even when that’s clear, the algorithm may disagree with what a player would likely find optimal/what is optimal in “gameplay” vs auto-evo. For example, adding more chloroplasts than actually necessary to survive. In addition, you can’t see if your current changes in the editor will place you in the miche, or lose it to another.
This dovetails into another point of weirdness: The foodchain screen appears to show newly appeared species, and species that have already mutated (as they will appear after you exit the editor) and your own species, before any modifications you make in the editor (it does not get updated). Technically, this isn’t accurate to the situation before or after the editor session. This situation is only used to make this display, and is not used for any actual calculations on species numbers, right?
Edit:
Hmmmm no, it does appear like you are using next generation AI cells and last-generation player cells to calculate current populations:
Edit: On a completely different topic:
I just added a rusticyanin in place of my cytoplasm and immediately started dying:
Seems like I wasn’t given an initial supply of iron initially, but I was after I died.
2 Likes
Swinging back to this project after a bit of time and have to reiterate - fantastic work! A few remarks on my end…
- Game looks gorgeous and very alive.
- Auto-evo feels like it’s in a really good place overall to me. Not only do the designs it comes up with make complete sense, but I’ve actually had situations where I get complacent in my current strategy, only to find that it is no longer as efficient next turn as i.e. one of the prokaryotes I preyed on developed chitin/toxins/pili.
- The oxygenation system adds such a great sense of progression to the game, and patch/resource dynamicity is easily the feature I’m the most excited to see expanded upon.
Now, some critiques:
- For some reason, I often see prokaryotes completely disappear from shallow-water patches and get replaced by eukaryotes. This is honestly probably my biggest gripe with the current build - it’s quite immersion-breaking, and atmospherically one of the things that makes the world feel alive to me is seeing those little swarms of prokaryotes swimming around the comparatively huge eukaryotes.
- Mitochondrial endosymbiosis needs a serious look-over IMO. Half the time I don’t have any candidates to perform endosymbiosis with, and even then 5 turns (and yes, I’m aware that it can be lower, but I’ve only ever seen it be 5 turns) is brutal for what is such a fundamentally important organelle to eukaryotes. Furthermore, since in practice it usually takes even longer for them to develop, what I usually see is eukaryotes radiating and becoming well-diversified by the time I manage to acquire mitochondria, so like half the eukaryotes in the world end up relying on metabolosomes. This doesn’t really check out with LAWK on Earth, where by all accounts the mitochondrial endosymbiosis event happened on very early on in eukaryote evolution (and hence why there’s basically no eukaryotes that lack them, and even those that do have only lost them later on). This also partly plays into the problem where early eukaryote gameplay is easily the toughest part of the game right now. Don’t get me wrong, it should be challenging, but right now it’s always so challenging due to the lack of mitochondria for such a long time that it feels like it breaks the principle of natural selection.
- The reporting system needs work. Right now it just haphazzardly dumps a ton of info on you that really isn’t that interesting, while making it very hard to get bits of info that would actually be interesting to know. The kind of info that I’d like to be notified and recorded, for the moment, would be stuff like atmospheric oxygen reaching milestones or endosymbiosis events, things that really only happen a few times in the entire campaign. Of course all the current info should still be recorded, but in a separate, lower-prio list.
- Pls add a way to filter/sort species in the global list by stuff like total population and cell size.
- One of the sole major aesthetic complaints I have about the game currently is eukaryotes and their nuclei. Namely, seeing that identical white ball everywhere around me not only genuinely hurts my eyes, but makes eukaryotes paradoxically feel a lot more sameish to each other than prokaryotes. It’s not necessarily that the model itself bad, it’s just how prominent it is yet identical everywhere. IMO it’s actually a major missed opportunity when it comes to visually distinguishing cells and making them feel aesthetically unique. At its simplest, it would be fantastic if the nucleus could be colored a-la the membrane; even better if you could do aesthetic modifications like the shape of the golgi apparatus/surface texture, etc. Would go a massive way to making eukaryotes more visually distinct from each other.
- It doesn’t feel like there’s a lot of size or honestly even morphological diversity when it comes to eukaryotes, and this is probably one of the bigger gameplay weaknesses of the late game. There’s probably several reasons for this, but chief among them (and probably the most easily solvable) is that MP costs remain static throughout the microbe stage. I really feel like getting the cell nucleus should significantly discount the cost for most organelles to give eukaryotes a faster pace of evolution and allow them to actually diversify in a reasonable pace. On the size front, maybe there should be soft caps to how big you can grow depending on the size of your cell nucleus? Bigger shouldn’t necessarily always be better, but realistically there does need to be room for size diversity with eukaryotes, which is right now basically nonexistent from what I’ve seen.
- The most minor point for last, a small suggestion - tint the background color of epipelagic biomes depending on factors like the presence of iron or oxygen. Would be a tiny, but great detail and would add to the sense of progression of going from the green-tinted, iron-rich waters of the prokaryote stage/Archean analogue to the more bluish water of later stages.
- EDIT: Forgot this one, but I’m regularly seeing several eukaryotic lineages develop in parallel. While this should not be outright impossible, it should really be rare especially if LAWK is turned on. I was honestly a bit shocked when I noticed that there were, by my count, 5 different independent eukaryote lineages in my latest game, and that I’m the only extant genus of my own lineage…
6 Likes
When auto-evo is run before entering the editor, it will store various information for referencing later. This includes obviously the population numbers which are needed to update the game state, but also all miche information is saved. So barring any bugs the food chains and population numbers are exactly what was just calculated and applied to the game world when entering the microbe editor.
You don’t get free resources for exiting editor (other than the free glucose cloud if enabled in difficulty options).
This is an interesting piece of feedback, because most of the time people complain about there being way too few eukaryotes or other big cells.
Also welcome back.
Due to the missing starting oxygen maybe it is different now but it used to be the case that by the time you got a nucleus, you easily had used metabolosomes long enough to basically almost instantly unlock the mitochondria.
AI doesn’t know how to do that, so I think it is pretty clear right now when it has happened as the player needs to intentionally trigger it. But yes it would be nice to add it to the event timeline, though I’m not too optimistic as that’s a really neglected feature with no one wanting to work on it for the last couple of years.
Is it always just white? Unless there’s a new bug the nucleus should take on the colour of the cell it is in.
You must be very lucky then as (well at least before 0.7.1) it was super rare to see even a single AI evolved species with the nucleus…
5 Likes
Thank you! Okay, will attach some details of my current world in case on if this info is useful to you. World is 3.1 billion years old. My current patch - no prokaryotes, and the recently extinct species are all eukaryotes as well:
Neighboring coastline patch; also no prokaryotes:
I’ve analyzed all visible patches and noticed that most actually either have solely prokaryotes or solely eukaryotes. In some cases it actually makes sense (anoxic hydrothermal vents with only extremophile prokaryotes), but in a lot of them it doesn’t (completely comfy epipelagic or coastal patch being entirely bereft of eukaryotes or prokaryotes, respectively). In the below image, red circles are prokaryote-only patches, while white squares are eukaryote-only:
Huh, maybe! I though that that countdown only starts after you acquire a nucleus, not before. I’m probably wrong, though. Still, it might be worth knocking down the metabolosome-fallback countdown down a bit (as well as the endosymbiosis countdown) - my experience is that by the time you get 5 metabolosomes you’re probably way less than 7 turns away from being able to flip to a eukaryote.
Here’s how it looks like now, compared to what I’m suggesting below:
So yes, currently the nucleus is technically very slightly tinted the color of the surrounding membrane, but the effect is so subtle that I honestly did not even notice it until now. Compare to the lower image, which IMO looks a lot more vibrant and the cells starkly different.
Possibly! I just went through my phylogenetic tree, and recorded all independent eukaryote lineages (6 of em).
My own:
As you can see, I’m literally the only living genus left in my entire eukaryote lineage. My closest living relative is a prokaryote that’s separate from me by over 1 billion years of evolution. Tiny eukaryote phylogenetic trees (well, bamboo trees) isn’t a problem only I’m having; here’s 3 more eukaryote lineages that barely diversified at all:
There’s 2 more eukaryote lineages that I didn’t image in order to not bloat this post too much, but they’re reasonably sized at least.
Hope this was helpful!
3 Likes
11 posts were split to a new topic: Reworking when auto-evo is applied in the gameplay loop