When you move into a patch, you’re listed as extinct in the food web because your pop is 0. That is, if you appear in the food web at all; it often misses half the species that exist, with the remainder given bizarre and nonsensical arrangements. Complex food webs don’t appear; generally, you see half a dozen species with no food source given. Rarely, there’ll be a predator; it eats everything in the ecosystem. (I saw that once, at least.) Also, 25-50% of the food web will be extinct at any given time, mainly due to…
The continuing issue of extreme species turnover. Species go extinct at a very rapid clip, which is probably the cause of the very simple food webs seen. If you remain in one patch for more than a few generations, you may witness every other species die off, forcing you to move to a different patch to continue hunting. This is also one of the root causes of the problems with endosymbiosis, along with endosymbiosis being self-defeating, in that eating your endosymbiont will reduce their population.
The food web seems to drive bizarre decisions by the AI. While still in prokaryote level, I witnessed an attack on me by long cells, which did not have the size to engulf me nor any alternative means of attack. They rammed into me harmlessly until I grew large enough to engulf them, and continued ramming me afterwards. I observed this behavior even when I was a eukaryote ten times their size (given that they hadn’t developed much in size themselves). I also observed spike-wielding cells attempt to kill me, using the spikes on their butts while they ran into me with their fronts. Completely harmless. They would have been threatening if auto-evo had correctly placed their spike, but auto-evo doesn’t place spikes intelligently - a spike is a spike, and it never places two.
The cell death noise and animation still plays when you digest a cell. If a cloud of cells rams into you and you engulf them all, you may receive a speaker-straining noise later.
Auto-evo still has the strange tendency of always building downwards, leading to unnatural patterns in the shapes of cells. This is very prominent with small cells, where you can often see the original cytoplasm hex and that all new parts are below it. (It has been explained to me that this is because auto-evo always randomly picks the bottom of the central hex as the first position to attempt a part placement.) There is also the returning bug of Giant Strings of Hydrogenase, which often compounds two or three times to produce cells with a staggering excess of hydrogenase. I saw one of these cells, with 6 hydrogenase, listed as a rustiphage on the food web due to possessing 1 rusticyanin. (It also had 1 thermosynthase. This and the glucose-eating weren’t listed.)
Atmospheric gases remain wildly unbalanced. Oxygen on my world went up and up and up, to the point of 40% oxygen or more. Supposedly, the gas chaos has been reduced, but I’m not seeing it.
Large cells aren’t handled well. The player, being incentivised to constantly grow their cell (as there aren’t many other ways to spend MP to improve your cell, and they are finite, like membranes), will soon end up with a very large one. But auto-evo stabilises cells at a certain size and doesn’t allow them to grow larger. As a result, a large hunter has no prey. This may be a cause of the lack of AI eukarya, since you need to already be large to fit the ATP cost in. Additionally, when you’re a large cell, speciation is very minor. It’s the same scale as a small species’ speciation - ie, one or two parts added. Large cells thus have less capacity to evolve. This is especially noticeable when it’s rare to see eukaryotes that independently developed nuclei. As a patch job, I suggest that species diverging from the player should get a temporary buff to their auto-evo powers to define them better. Not sure if this should apply at all times; if the buff increases over time, then those divergent species would be extremely divergent later on, so maybe the buff only applies to divergences within a certain period after the evolution of a nucleus?
As an extension of the above, auto-evo is not nearly as aggressive as the player. Depending on what parts they add and on the MP cost, the player could grow by 2-6 hexes, while auto-evo tends not to do so unless it’s making gigantic hydrogenase strings. This may be a cause of me rapidly outpacing other cells in size, although really the actual cause is just ‘auto-evo doesn’t like cells that are any bigger than 3 hexes’. As a result of this and auto-evo’s other failures in analysing the player, I don’t see counter-evolution when I become more powerful. (And it is ‘more powerful’, not ‘better-adapted’. Most parts are one-size-fits-all, although ATP production can take a big hit when you move into an environment with lower oxygen - like the underwater cave your endosymbiont is in.)
Organelle suggestion seems to be faulty, according to some reports I’ve seen already. In my case, it was very obviously glitchy, as it recommended myofibril before I had become multicellular. In fact, I may not have been eukaryotic. I don’t recall.
When you’re in the multicellular stage, maturing a new cell while you’re engulfing causes the engulf graphic not to apply to the new cell.
I added a flagellum to my cell and it reduced ATP consumption, by a significant amount. I was in the multicellular editor, my cell was round, I had four mitochondria, oxygen was around 40% so they were very strong, and I didn’t have any parts other than mitochondria, cilia, flagella, and the mandatory two for a multicell.
As a 3-cell, I had 15 storage of the main three compound, but 174 ingested matter storage. I assume that compound quantities were rebalanced so you didn’t get absurdly large amounts of compound to manage. If this is the case, a spot was missed.
Issues with parts
Spikes still cause abrupt deaths. I once ran into a cell with a spike at high speed and went from 45 to 0 health instantaneously. (This was as a prokaryote.) Later, as a eukaryote, I observed an NPC player cell get obliterated by vortexing spike cells. The latter isn’t an issue; spikes are the obvious counter to vortex cilia, which are…
Way too powerful. Vortex cilia are only useful for eating very small prey, but luckily, those are the only prey sources you’re likely to see. Enemy eukaryotes are very rare. Spikes are the natural defense against vortex cilia, but I doubt they’re implemented into auto-evo. Spike-wielding species also only ever develop one spike, and at a random position where it may be of little help. Removing vortex cilia wouldn’t help, because predators are weak without them given how small most prey is. If they were implemented into auto-evo, this problem wouldn’t exist.
Flagella modify your membrane shape as if they’re based in cytoplasm. As a result, if you have a column of flagella, they’ll create a weird spike on your cell. They render as a single flagellum, though, if you don’t have them horizontally spaced.
Specialised vacuoles don’t seem to be much help. As I understand it, the +8 to storage applies individually to everything, which means that specialising your vacuoles causes an objective drop in total storage capacity, which is magnified the more resources you are managing. (You also can’t specialise vacuoles for ingested matter, which is rather irritating, but hunters don’t need buffs right now. I’d like to see structures for containing ingested matter in Early Multicellular.) If storage doesn’t work this way, then the tutorials may have a problem, or I may have failed to read something.
Problems and suggestions to fix them
The cladogram needs a search bar, desperately. It rapidly develops into a very wide and chaotic list of incomprehensible randomly-generated clusters of half-syllables, with genuses only consisting of 1-4 species. This makes it difficult to find the species you just clicked on to look at in the cladogram, since you aren’t sent directly to the species you clicked on. I’d also recommend adding an option to make the cladogram horizontal instead of vertical. It also doesn’t look much like a constantly-dividing cladogram; there are many long arms where a species persists for a long time despite constantly changing.
The chemoreceptor description is outdated, as it only refers to compound detection and not species detection. Additionally, I’d suggest adding a priority slider that ranges from ‘distance’ on one end to ‘quantity’ on the other. At max quantity, the chemoreceptor would direct you to the largest cloud in range. At max distance, it would direct you to the closest. In between, some sort of strange math would be performed to balance the two.
A simple adjustment to compound intake would make gas clouds much more interesting and aesthetically pleasing. Right now, when you enter a cloud, you instantly eat everything in range, until you’re full. This reveals the pixelated edges of your intake radius, the fact that the intake radius goes far beyond the bounds of your cell if you’re long, and also cuts a swath through the cloud. I suggest simply making it so you only eat part of the compounds in the cloud when you enter. This makes more logical sense (the mechanisms of diffusion that are presumably enabling your compound intake would be slower when less of the diffusing compound is present) and would make the effect much visually cleaner, by having the cloud slowly thin as you are present. It also makes you stay in a cloud for longer, and makes you judge whether it’s worth remaining in a thinner cloud or looking for a thicker one.
The drivers, in my experience, of player eukaryosis are “wow, there’s so much oxygen! My ATP balance is insane!” (experienced recently) and, more commonly, “man, I’m rotating so slow!” Becoming a eukaryote is a large up-front investment, and most of the benefits are ‘Normal part, but more powerful’. There are a few special benefits, like multicellularity and cilia, the latter being absolutely necessary for any large predator. There aren’t otherwise many incentives to make the leap, other than ‘nowhere else to go now’. I suppose this is accurate, but it’s still odd.
Divide order is a good addition, but it’s undercooked. The UI doesn’t feel user-friendly; the mouse-based sliders don’t work well, often failing inputs if you didn’t get your mouse in a narrow column, and it’s hard to visually grasp the block of text. The coordinates don’t seem like they’d work well in a hex grid; a three-hex system would work well. I’d also recommend a block-mode, where instead of micromanaging each and every part (which gets tedious, especially when you’re a large prokaryote), you move parts by type. So, for instance, instead of delicately arranging your thylakoids, metabolosomes, and flagella in a 20-part order, you instance have a 3-part order. Thylakoids, metabolosomes, and flagella, instead of thylakoids, metabolosomes, metabolosomes, thylakoids, thylakoids, flagellum, metabolosomes, thylakoids, thylakoids, thylakoids, metabolosomes, metabolosomes, metabolosomes, thylakoids, flagellum, flagellum, thylakoids. (If that was hard to read, you see the problem with the divide order.)
At some point, there’s going to need to be a level-of-detail implemented. Already, at this stage of the game, a big eukaryote and small prokaryote will have an absolutely vast size disparity. When the multicellular stage enters proper development, that’s going to be compounded. Right now, you could still see that small cell, but at the cost of not being able to see your own cell due to its extreme size. The zoom needs to be scaled to your cell size, and the natural addition to that would be to make it so that cells that have too big of a disparity with your organism’s total size vanish into abstraction. Their predators still predate them, but instead of simulating that predation, the predator just receives resources when simulated. (This would need to be expanded somewhat in the Aware stage - with the size disparity involved when you’re comparing ants and elephants, at some point you’d need to have to abstract things to just ‘flying insects’ and ‘tree insects’ and ‘ant-like insects’.)
The addition of divide order is good, but I’d also like a system for vestigialising things. In this case, you’d be able to set cell processes or parts to be automatically off when you begin the round; you can later make them vestigial, making them no longer functional, but also making them cost little resources to maintain. Later, you could get rid of vestigial parts for a lesser price, or reenable them for less price than replacing the part entirely.
Misc
Someone suggested merging MP cost and AI mutation rate into an ‘Evolutionary Pressure’ slider, which I thought was an interesting idea. I’d recommend implementing it with an option that allows you to individually tweak the two components, so your slider has sub-sliders. This is mostly due to the conceptual appeal of having so much configurability that your configurations have configuration. A more sane implementation would be having an Evolutionary Pressure heading with these two sliders under it.
I’m not sure what the coding for floating toxins is, but they seem to have been given the kludge job of a massive internal size to prevent engulfing them, because when I ram them while engulfing, I’m informed that my cell is too small. Whatever the reason, this message should be suppressed or its cause fixed.
The tutorial should tell you about the impact of cell shape on rotation and speed. Long cells increase speed, round cells increase rotation. Wide cells are bad in both respects. I had to find out the second of those through spontaneous experimentation and the third by simple hypothesis.
You can see the seams at the edge of the patch, presumably where it loops. You can’t take resources on one side of the seam from the other, which is noticeable, especially as a thin cell, and the flow of clouds behaves weirdly around seams.
Vortex Cilia plus Injectisome Pilus will let you pull cells your size into your piluses and kill them to gobble what they drop. I find this more effective then engulfing, though engulfing is required to use this strategy.
3 Likes
aah31415
(The maker of SitF, Radiostrocity, The Lifenote and TGBing; The Second Ascended...; And just maybe a security warning come alive...?)
3
Injectosomes in general seem more powerful than regular piluses to me, even tho they should deal the same amount of damage to non-resistance-modifying membranes
I think the “toxic” part means they should do toxic damage in addition to the regular damage (unless you are tolerant of the toxin type). Also, I have been killed by regular Piluses a few times, so the regular damage isn’t that bad.
3 Likes
aah31415
(The maker of SitF, Radiostrocity, The Lifenote and TGBing; The Second Ascended...; And just maybe a security warning come alive...?)
5
Oh, I thought it was supposed to replace regular damage, not add extra damage upon that…
3 Likes
AnthropocenianAge
(AnthropocenianAge Arthropleura wants to give you a hug!)
6
Out of curiosity, what conditions do you use for playing, @thriveuser1454? When I use max mutation for AI, which is often for my playthroughs, I see a lot of AI Eukaryotes being generated. Also, hilariously, I found and fossilized a 20-size (40 hex) AI Prokaryote in my most recent playthrough.
1 Like
aah31415
(The maker of SitF, Radiostrocity, The Lifenote and TGBing; The Second Ascended...; And just maybe a security warning come alive...?)
7
Those are rare nowadays, but prior to the miche system they weren’t that hard to find
Really? When I was playing, it looked like vortex cilia didn’t apply to large cells. That’s why I mentioned that they weren’t useful for hunting similar-size cells.
I believe I was playing with 0.8 MP cost and AI mutation rate at x2. I don’t recall exactly how I modified the stats. I think it’s usually those two on top of the Normal preset. I’ll try a run with maxed-out AI mutation and come back.
1 Like
aah31415
(The maker of SitF, Radiostrocity, The Lifenote and TGBing; The Second Ascended...; And just maybe a security warning come alive...?)
13
Pretty sure they can still somewhat slow down a similar-size cell if it’s running away from you
I was away for a bit from working on Thrive when this was posted, so I’ll try to reply a bit now.
This has so many issues interwoven that it’s very hard to unpack.
The first thing is an effect of the player species not yet existing in the patch you move to, but the food web always being forced to show the player species (even with 0 population) as otherwise the situation is often that the player is totally missing from the food chain. Which was reported as a bug and was fixed like that.
Food chain not having the correct species list is a mystery bug, which has an open issue:
Are they extinct at least? Nothing except the player should be able to exist in the foodchain without a food source.
Species are added to the foodchain that are predated on to show that relationship. This means that extinct species get pulled in so that data about other species can be correctly shown.
I can’t say anything else except someone needs to improve the auto-evo. I’m not an expert in the system so I’m highly unlikely to be able to do anything about this.
Though for endosymbiosis:
The food web actually has 0 impact on the gameplay AI. It is another system that I’m not specializing in and instead will just wait years for someone willing to improve the AI to come along.
But did you see gases go above 100% in total? Or oxygen above 70%? If not then clearly the system is better than before.
I agree the second point is valid, but we try to tell the players in various places that they don’t always need to just get bigger and add more stuff. Hopefully this problem gets lesser if we add more ways to spend MP (some of which are on the microbe stage roadmap).
Are you sure? Because this should have been a thing that was bugfixed ages ago.
Please show a before-after screenshot.
Ingested matter storage is dependent on your size in hexes. And if you were a cell colony (As a 3-cell)? Then what the GUI shows you is the sum of engulfing capacity, not your individual capacity, because otherwise you couldn’t see what the engulf situation is like in your other colony cells.
If you don’t want to get instakilled you need to invest in enough health that it is above the max single pilus hit damage (there is a cap somewhere in the range of 30-40 but I’m too lazy to look in the code to see the exact value). That max pilus hit is balanced so that it cannot kill a default cell at full health.
Spikes as defence are implemented so this is a nice emergent gameplay interaction.
Not doing this would require reworking the entire game to have hex-based and non-hex based organelles. I cannot imagine the nightmare that changing this would cause.
Which way do you mean? Specialized vacuole offers just double the storage it would normally for one compound. It doesn’t offer more as we’ve been careful balancing it to not make it like so that just adding one is super powerful and encourage the player to always need to add one. It is more meant like a special situation where you wish you had more storage for one compound type that is key to your strategy.
I assume you mean the evolutionary tree. This has been suggested in the past but I’m not sure how useful a search bar there would be, as I don’t think the average player remembers what names they’d want to search for.
I take it this is the core of the issue. It’s actually temporary that clicking on a species takes you to the evolutionary tree. In the future once there are Thriveopedia pages for each species, clicking on a species will directly take you there and no longer to the evolutionary tree at all.
Good point, I’ll fix that to add or microbes
That seems very unnecessary change as I expect to be more confusing for new players and more experienced players will inquire about it and will want maybe more configuration.
This has the effect of either needing the player to stop moving or move very slowly when reaching clouds or moving in a circular manner in clouds. Which I’m not sure would feel that good change and would slow down the cloud collecting gameplay considerably.
Which part are you talking about?
Thrive actually uses polar coordinates so the coordinates are exactly what is used internally to distinguish the organelles.
This is unlikely to be done as this would need quite a lot of special tweaking, which I don’t foresee having a need for in the body plan editor. This organelle reordering was mostly added as many people have asked for it and the feature is basically what will be needed in the multicellular body plan editor anyway (and there shouldn’t be the same kind of need for grouping).
I think this has been suggested in the past. And it is a doable thing but would be quite a lot to program so it is quite unlikely this feature will make it into the game unless someone very enthusiastic about this specific feature comes along and programs it.
I’m pretty sure our GUI controls don’t support that so that would be quite insane number of hours required to make a custom slider component. Much easier to just keep the separate sliders as that’s actually doable with the GUI framework. Though there could be a checkbox to tie the slider values together.
I suppose this is actually possible now to fix as the chunk system supports chunks that aren’t marked as engulfable. I’ll fix it (mark their engulf size as 0 so they don’t get the entity component that marks them as engulfable).
This would be nice to include somewhere in the game, but I don’t think our short tutorial should have this added in as it currently is focused to only tell the critical information. There’s a bunch of gameplay information that I’d argue would be more deserving of being in the tutorial before adding this information.
You mean in the compound clouds? That’s a known rare bug that I have no clue what we’d need to do to fix:
Actually injectisome always deals a fixed toxin damage amount but the base pilus deals damage based on physics impact strength.
3 Likes
aah31415
(The maker of SitF, Radiostrocity, The Lifenote and TGBing; The Second Ascended...; And just maybe a security warning come alive...?)
15
That could be why they seem so superior when your speed is not much greater than that of the chasen microbe.
I see a lot of extinct prey species, but now that you mention it, I’m unsure if I saw any living species with no listed food source. I’d like to say yes, but I can’t be sure without testing, which I don’t have the time for right now.
I saw a lot of patches with atmospheric holes. I’m not sure if I saw any with oversized atmospheres. Oxygen was going up steadily, so if it could have gone above 70%, then I didn’t stick around long enough to observe.
Yes, I’m sure. I saw it happen.
I don’t have one because I don’t compulsorily screenshot in-game. I probably didn’t have a save because it was multicellular, and like I’ve mentioned above, I don’t have the time to dig through my old saves to try to replicate the bug. ATP cost reduction was slightly below the ATP cost of a fully-extended flagellum.
I was a multicellular species with three cells. I’m not sure why engulfing capacity would be summed but regular compound capacity not.
A. I don’t have a hard time remembering very specific names that I’ve just clicked on to look at.
B. You saying that names are hard to remember implies a serious deficiency in the naming system.
You have to drag the icon to the column of icons, and I don’t think it always works. It’s somewhat difficult to tell initially if your drag has worked, there’s very little visual distinction between the different parts you’re moving around.
Because compounds are distributed so that each cell has the same average amount so non-summed bars still work fine, but for ingested matter that doesn’t make sense to do the same display.
This is not true? You can grab every part (of the organelle entry in the growth order menu) to drag things around except the up / down arrows. Is the darkened colour not visible enough to tell what’s being dragged?
2 Likes
aah31415
(The maker of SitF, Radiostrocity, The Lifenote and TGBing; The Second Ascended...; And just maybe a security warning come alive...?)
18
To be fair, a large chunk of real species’ names are hard to remember aswell.
