Hmm. Is there any existing philosophy on whether players should be allowed to go back and redo parts of the game? I suppose it’ll also probably be hard to accidentally become an intelligent aquatic species or a plant.
Personally, I’d be happy with a sandbox where you could switch species at any time. I quite often ignore the objectives in video games to explore anyway. But that’s just me.
I’m sure non-fixed body plans is a solvable problem, I’m just not totally sure how. What do you think designing a tree in the editor with metaballs would look like?
Players could just load a save game from a previous stage and make different decisions.
Yes. That is a good solution. That would be like Crusader Kings 2. I often switch characters in that game if my character is unsatisfactory. This does mean that the species the player spent so long making is abandoned, but it is far better than a game over. Besides, taking something the AI made and trying to be successful with it can be fun, as the AI is usually quite stupid.
I think designing a tree would be the same as designing anything else. What I was suggesting was variation in attributes, not body plan. I expect in Thrive all plants of the same species will have identical body plans, like in Spore. From what I understand, the metaball system would not allow for nonspecific or amorphous body plans (so no amoebas).
2 Likes
Zahyyy
(the one who spent 4 hours writing a single post)
23
One way to approach this is when designing amorphous bodies you would be just setting the variables such as the branch frequency, distribution, angle, the frequency of leaves and the concentration and stuff like that. Then, the editor would just put some of the branches and so on down according to what you specified. If done well, this would grant us with a variety of plants of the same species and would be easy to control. Also, the auto-evo would make more realistic plants with this system, in my opinion.
If my explanation is unclear, I can give some examples.
Edit: You wouldn’t even have to set those values up, the game would just look at the plant you made using meatballs and find out the values from there.
That sounds nice, but I see two large problems. First, there will presumably be hundreds or even thousands of “plants” (or whatever analogue) around the player. We must have a system that an average PC could handle. Your idea seems like something that would require far too much calculation from the computer and affect performance. The game would presumably be calculating other things based on the plants’ shapes and sizes (e.g. gas or energy production), and these cascading calculations might result in huge lag spikes anytime the player is near plants.
Second, that sounds very difficult to implement. I know almost nothing of computer programming, but I bet that would be a ton of work to merely have variation among the foliage. Spore looked fine with cloned plants, and I think Thrive would be fine with them too.
1 Like
Zahyyy
(the one who spent 4 hours writing a single post)
25
I think that the lag spikes would be no greater than when around other animals. The calculations would be there only when “designing” the plant, but after it being placed, there would be no additional calculations.
And I don’t think it would be hard to implement (I cannot be sure, though, my knowledge of programming is also not great), it would basically be just like particles, or hair if you ever used Blender. And their placement would be determined by the variables. The Blender example describes it the best, I think.
Also, I think it would go nicely with NickTheNick’s auto-evo algorithm proposal, as it would create some neat and realistic trees based on stuff like “leaf” surface and the resources needed for the “trunk” and so on.
Last, but not least, I would not say that any part of Spore looked fine by today’s standards. What more, if you had something in the vegetation style as, let’s say, a birch forest, I think that the “copied and pasted” trees would be really off-putting and uncanny.
If there were no additional calculations after generation of the plants, then I suppose there would be long loading screens. Perhaps it could be like Minecraft or Dwarf Fortress, wherein the trees vary in size but still follow the same pattern. The algorithm would produce the base structure, and the variation would be caused by randomness after that. But then the game would need to know what parts can be random (like tree branches) and what parts cannot be random (like arms of bilaterally symmetrical animals). Again, that doesn’t sound feasible if the idea is to have the same editor for all organisms. I disagree with you about how Spore looked. I think its creature stage still looks good, albeit cartoony. I do not find the cloned trees uncanny. In real forests, the trees often all look the same from a distance. Besides, if they are functionally identical, why bother making the variation? Even if the game would be prettier with your idea, I want the developers’ efforts to be invested in meaningful gameplay, not beauty.
I quite liked the example image in the auto-evo thread of a creature standing in a rocky area. All the plant-things in that image are the same, yet it looks good. I think we are somewhat off-topic, as I do not think this is pertinent to plant gameplay.
How would being a sessile organism be fun? Earlier in this thread there was an idea of making it like Plague Inc. Even so, would the player be allowed to progress to later stages? If I cover every continent with my super-grass, what is left to do? Either there needs to be some kind of CK2-esque species-switching system, or there need to be non-LAWK options to allow intelligent plants (maybe cordyceps-esque hijacking of organisms with nervous systems?).
2 Likes
Zahyyy
(the one who spent 4 hours writing a single post)
27
That’s why I gave the example of Blender particles. Look up how Blender particles work, I think it might be usable. You have your base trunk and across the trunk, some points are scattered based on the particle system. Then, there are branches from these points. Do the same for the leaves and voilà, you’ve got yourself a plant.
I think this wouldn’t be incredibly hard to code, wouldn’t cause long loading screens and would make the game overall better-looking.
Well they wouldn’t. The placement of leaves and such would be important for the energy input/output, but it wouldn’t have to be calculated on the run, just as the game doesn’t calculate it for every microbe right now.
In my opinion both are very important (Take a look at the game Ancestors, the graphics are kinda off-putting). The gameplay can be amazing, but when it doesn’t look good, it’s just not that great of an experience (I’ve actually recently written a post about the importance of the atmosphere the game has and landscapes and such). And vice versa, if you have a game that looks great, but lacks things in gameplay, it’s not going to be well-received either. Also, this is not just about the looks. As I said, it would make the auto-evo more responsive. How would auto-evo work in your scenario? And how would a computer even generate the plants? I cannot imagine how that would work.
True. I think that playing as a sessile organism could have one part of the gameplay being the editor and the second one sort of a strategy gameplay - not exactly like Plague, but still closer to Plague than to a “mobile”(?) gameplay.
Edit: Also, I don’t think that plant gameplay should be possible in the later stages. If you chose to be a plant, you are a plant. I’m not a fan of the idea of sapient plants, and I think that’s the general opinion on the forums.
I started this before Zahyyy’s reply, and I’m going to go ahead and post it. I don’t have anything to add on except my agreement that intelligent plants don’t really make sense. The only way of experiencing the later stages of the game as a plant should be if you’re a crop that your local intelligent species grows.
I know some stuff about programming and about how computers work. At its simplest, I wouldn’t expect making procedural trees to be very calculation intensive. You more or less will need to pick a random length for each branch, and there might be 1000 branches in a tree. Big deal, a modern CPU can do billions of calculations a second. Making roots that explore and curve in a random way would be more intense, as would rendering the tree. Memory could also be an issue if your storing a forest of random trees.
The dev team will have to decide what tradeoffs to make here based on the situation whenever they get to this, and Solitarian is right that we’re slightly going off on a tangent.
For an individual plant, growing is the game. I don’t think having a pre-defined shape would be an option if that’s what you’re playing. In editor you’d have to do some combination of pre-defining parts like leaves, and specifying when and where these parts can grow. How that would look if it was part of the same editor as is used for animals is where it gets fuzzy for me.
Right now the editor concept pretty focused on drawing one body plan and using it right off the bat and until you re-enter the editor. You can’t grow a new segment or go into a cocoon and metamorphosize. The other extreme would be simulating signaling proteins and genes turning on and off and all that other stuff. That would make the game far too complicated for anyone who’s not an embryologist, besides being impractical. What’s really needed is some level of abstraction between the two. Do we have any actual biologists in the community who might be able to point us in a direction?
If playing a single plant turns out to be too easy, excitement could be added by playing a group of them. Going further and playing the whole species would also be a possibility, and would resolve the problem of needing non-fixed body plans. The game, then, would be finding a nice niche. For example, one could play a weedy plant that produces lots of seeds and exploits disturbed environments, or a carnivorous plant that can grow in nutrient-poor soils. The hard part of that approach would be simulating all of the ecology.
I’m not an actual biologist… obviously, but I do know a fair bit about some stuff. What you may be looking for is HOX genes. HOX genes are short for homeobox genes which fundamentally decide where organs go which may be helpful in plant gameplay (more specifically evolution) or any other kind of gameplay.
Also I have the humble opinion that players should be able to create their own organs. I don’t have any reason or rhyme to this I just think it would give players more freedom.
I disagree, Zahyyy. An ugly game with great gameplay is much better. Dwarf Fortress and Cataclysm: Dark Days Ahead have ASCII graphics (if you don’t use a tileset), yet they are amazing games. The auto-evo system would produce plants through the same natural selection process that produces animals.
If you say so, Pentaphon, then perhaps I will stop being so concerned about lag from plants. I agree that using the same editor for all organisms (which is a concept I support) seems to make this difference between plants and animals impossible to create in terms of gameplay. If the “body” of the organism in gameplay is the same as in the editor, then there can be no random growth nor non-specific body plans. Again, I think my idea of having things being random sizes between some maximum and minimum is a good solution.
Yes, playing a group of plants could be interesting. The problem I noted remains, however. What is left to do after the player has made a successful plant species? The player would hit a gameplay wall and just have to restart. This is like in Crusader Kings 2, where if the player creates a stable empire, the game becomes boring because there are no further goals to achieve other than painting the map. I don’t like the idea of the player being stuck with no ability to advance to the later stages. This would mean that playing as a plant would effectively softlock the game.
Again, I think giving the player optional non-LAWK options would be a good solution. I agree that intelligent plants don’t make sense, which is why the player does not have to use these options if he does not want to. However, I want there to be some way to progress to avoid this problem of softlocking yourself.
The problem is not generating the shapes for all the plants as that can be done during a loading screen when exiting the editor. The likely performance issue is that, when there are a lot of things that need to be drawn, instancing is used where the same model is duplicated to different positions. This is the approach games use to draw forests by having just maybe 5-10 different trees and then placing instances of them around. This would mean that each individual plant of a species could not be afforded to have a unique shape.
The advantage that blender has here that the properties of the parts that are based on some formula are known. We would need a more complicated system that could generate good placings for player designed branches.
That’s how modern graphics works, you “cheat” as much as you can. Otherwise the performance is really bad.
There is a similar set of genes for plants called MADS box, though. Animals also have MADS box genes but they aren’t used for the same thing (and vice-versa). Here’s a relevant blog post I keep being reminded of in this thread. I wonder if there’s a similar set of genes for every one of the many evolutions of multicellularity.
Here’s a browser game where you play as a plant. I think it’s the one I’m thinking of, but I don’t actually have flash installed right now to be sure. Sorry for the double post, but I think this deserves it.
Thrive as a plant could be similar, but in real time. Smoothly integrating a game about growth with the main game which is about movement continues to be a sticky wicket.
zenzonegaming
(Steve Buscemi Fish Clown Detective Cult Leader)
35
I hope someone has not already said this but what about Treant gameplay?
They are basically walking plants, right? That has been discussed a lot in the past. In summary: they aren’t realistic as they can’t get enough energy from photosynthesis to move around.
Agreed. If such plants were to exist however they would either need to find an alternate method of locomotion (that requires less energy), or find another source of energy.
So you people talk about plants not being able to get to space but i was wondering if this games species are supposed to be carbon based lifeforms?And if so does it rule out hypothetical carbon based life?
I can think of a few scenarios where carbon based life could evolve into space without the need of an space faring civilization.
1.Life forms on planetary rings.
Planetary rings are dense enough to allow migrations between rocky bodies in it (kilometers between for spreading spres or seeds) and could have all the elements for carbon based life(some theories argue that water on earth is something from space).And the star could provide ample energy for life creating conditions.Thus life could evolve from the start to be space adapted.So you could play locked in the plant stage but start from the beginning adapted to space.
2.Planetary features that reach into space ex.Very tall mountains or plateaus that reach into very low pressure or space environments .
This approach could see plants evolving from being adapted to typical life conditions on the planet to being adapted to vacuum,just because those biomes are accessible to critters on the planets surface.(Plants can spread from the base of the mountain all teh way to the airless top through natural selection).This way you can have a whole ecology adapted to vacuum.
I can see the second situation being viabile for a space stage if the planet you evolved on had low gravity,to enable the spreading of spores to other rocky bodies.
All those ideas allow you to play with pretty much any stage into space(microbes and such).
Let me know of your opinions.
I saw people here say that playing asa plant you will be stuck on planets?So no space gameplay.
I proposed some ideas to show that nonsentient space gameplay is plausible.
Like plant gameplay in space.
in wack. I apologize for my inexperience. However I do think you understand what I am trying to get at.