|
Post by Immortal_Dragon on May 27, 2016 20:43:56 GMT
Pardon my piping up, and if this isn't appropriate for this topic I'll remove this post and possibly put it in a thread of its own if need be, but wouldn't a plant/animal hybrid type organism best be made starting in this stage? As in, an animal-type cell evolves the ability to bond, but instead of attaching a copy of itself, it bonds with a plant-type cell and they start to evolve the ability to share nutrients until they're basically the same type of species. It'd technically be cheaper than evolving your own cell to contain chloroplasts as you only need to evolve sticky membranes/collagen/whatever substance will allow it, then find a plant cell to stick to.
Just a random thought I'd come up with from reading what's been going on in this thread.
|
|
|
Post by mx on Jun 2, 2016 1:58:34 GMT
Just an idea from someone with very little knowledge of programming So once this conversion happens you're back to a generic palette? Like there's muscle and bone and nerves etc which are the same for all players? Because if we're going with generic palette what's the purpose of the intermediate phase where you specialise the cells? I guess it smoothes out the transition but a shift to a generic palette is always a shock as it forgets the past completely. perhaps moving from 2D to 3D you can look at a few bigger more generic features of the colony and have 7 or 8 big features of the 3D model demonstrate these features for example - the speed of your colony is higher than X thus your creature will have X fins (legs?) added to the generic preset 3D body - You have generated X energy through chloroplasts and X energy through absorbing other cells therefore the predatory parts of the generic preset body will be minimized/taken out and leaves (or the aquatic version of leaves) will be added or vice versa + Above can also have added what method was used to kill the cell- You have X mitochondria in the averaged cell of the colony which transcribes to X strength in the 3D model With just these three ideas and the two ways of killing other cells currently in game (absorption and toxin) you could get creatures like - a fairly strong creature with two legs that photosynthesis, weak creature with eight legs that spits a toxin, Strong creature that has four legs that eats weaker animals alive This way it doesn't completely nullify the work you put into the colony stage as you will still get a variation in what you get coming out but it limits the variation for code and players wont get abominations because the bone structure hasn't been generated nicely
|
|
|
Post by tjwhale on Jun 2, 2016 9:14:19 GMT
Immortal_Dragon I guess one difficulty of being a species with two cell types bonded together is reproduction. When you reproduce you'd need both cell types to split at the same time and grow in a coordinated way. Though not impossible that is probably quite a big challenge, might just be easier to have one base cell that can become both. mx Interesting. Well I guess the 2D -> 3D transition is the next one to worry about after the single -> multi cell. Not sure exactly how it'll work.
|
|
The_Wayward_Admiral
Spacefaring
The_Real_Slim_Shady
Atrox drew this awesome image of the Keldori!
Posts: 1,011
|
Post by The_Wayward_Admiral on Jun 2, 2016 11:27:22 GMT
tjwhale intercellular coordination happens on massive scales within multicellular organisms. Gap junctions basically allow two cells to share a cytosol, so if a two-celled organism has one cell producing proteins associated with mitosis, it could diffuse those to the other cell via gap junction. By this process, the two could divide together BUT keeping the dividided cells properly organised would be difficult without an ECM or the like. Edit: I suppose there would be three strategies that might help, and I'll do a little investigating once I'm done with work today. The first that I'm thinking of would be to rely on the close proximity of the cells, and assume that they'll start latching onto each other, each looking for specific protein markers in their partner cell and only be satisfied once the find it. The second would be for both cells in the parent organism to divide but not to wholly detach from their progeny until the progeny connect and send signals back to the parental cells. The third strategy would be for the parental cells to shut down all hormone production as they divide for a set period of time determined by protein decay. As soon as detachment occurs on the part of the child cells, they would start producing communication hormones that attract their partner cell for connection. Edit for the Curious: In large multicellular organisms, gap junctions aren't efficient at transmitting information long distance, so instead hormone messengers are sent throughout the body to relay information about division patterns and epigenetic controls. Edit 3: Perhaps for the 2D/3D transition, a message could pop up once you have 100 cells, saying, "Could you do me a favor? Use the scroll wheel to check on the underside of your organism."
|
|
The_Wayward_Admiral
Spacefaring
The_Real_Slim_Shady
Atrox drew this awesome image of the Keldori!
Posts: 1,011
|
Post by The_Wayward_Admiral on Jun 2, 2016 13:31:59 GMT
Stop the presses! I had an idea about the transition from differentiated cells in multicellular to functioning tissue in aware. If we assume that we use the Atrox-Admiral method of multicellular organism creation (joint patent pending), then there is a way to determine muscular movement from player defined muscle groups. If we assume that the computer recognises tissues by their ability to carry out actions, then it can decide when a cell is capable of enacting locomotion. It will then run several calculations once the player constructs their organism. Firstly, it will determine how much energy the muscle mass on the organism consumes when resting and when active. Secondly, it will determine which bones the muscle tissue is articulated to. This can be as simple as "every muscle touching this bone until I hit something that isn't muscle" (which necessitates connective tissues like cartilage and fibroblasts for larger organisms). It can also be something like "Every muscle cell within range X of this bone cell, and if a muscle belongs to two bones then it works to draw them together or apart." Thirdly, it will analyse the structure of the bones/cartilage/hydro skeleton/exoskeleton to which the muscle adhere. If a muscle is attached to a ball and socket joint, or to a lever joint, or to a sliding joint, etc. it will perform the action indicated by the joint type of the bone it connects with (perhaps the computer will ignore bones altogether for muscle movement and focus wholly on joints). Larger muscles can perform this action with greater force. If no bones are articulated with the muscle, then it can act in an undulating motion in any desired direction (or in the case of cardiac muscle, expand and contract to pump blood). Calculations and attributions similar to these can take place for every tissue type, determining if a nerve network functions, which cells have adequate blood supply, and so forth. Let me know what you think Edit: and I used the term muscle cell solely for the purpose of simplicity, to the computer it would just be "cell with enough actin, myosin, and mitochondria to move other cells"
|
|
|
Post by timetraveler22 on Jun 2, 2016 14:31:56 GMT
All of these ideas are great and all, but, I must mention that to make a load of algorithms for an even bigger load of cells in a 3D view, I'd imagine that the processing power to power this game could kill a high end gaming computer. I'm no computer guy, but, doesn't the processing that will run this game should be considered?
|
|
The_Wayward_Admiral
Spacefaring
The_Real_Slim_Shady
Atrox drew this awesome image of the Keldori!
Posts: 1,011
|
Post by The_Wayward_Admiral on Jun 2, 2016 14:39:09 GMT
My first post was not intended to reflect game mechanics, but merely a scientific excuse for the asexual reproduction of a small multicellular organism. My computer would definitely explode even thinking about that sort of detail (and take me with it, because lithium batteries are unforgiving).
And you have a point in reference to the second post, so past 100 cells, it needs to be on a tissue by tissue basis, rather than cell by cell. Perhaps once the game view switches from 2D to 3D the editor can lose all individual cells and only allow players to mold tissues into organs to a minimum allowable resolution (perhaps something that can be set in the options menu, so that every computer can be set at what it can handle).
Additionally, the "by the numbers approach" could conceivably be kept as low as 1600 lines of code, which isn't bad for an enormous-scale game (that is however, the most back-of-the-envelope estimation I've ever performed, so take it with several grains of salt).
|
|
|
Post by timetraveler22 on Jun 2, 2016 14:49:20 GMT
My first post was not intended to reflect game mechanics, but merely a scientific excuse for the asexual reproduction of a small multicellular organism. My computer would definitely explode even thinking about that sort of detail (and take me with it, because lithium batteries are unforgiving). And you have a point in reference to the second post, so past 100 cells, it needs to be on a tissue by tissue basis, rather than cell by cell. Perhaps once the game view switches from 2D to 3D the editor can lose all individual cells and only allow players to mold tissues into organs to a minimum allowable resolution (perhaps something that can be set in the options menu, so that every computer can be set at what it can handle). have you ever played Kerbel Space Program? When Building crafts in the editor, you get things like struts and fuel tubes that you can connect to other tanks and fuselages to either stabilize or drain fluids to equalize balance and stuff. We can do the the same but with muscles and tendons. Where you place the origin of the muscle is the base and where you connect it will be the "constricting" end(I don't know anatomy really well). Think of a hydraulic machine on a crane. We could use something like that for muscle.
|
|
The_Wayward_Admiral
Spacefaring
The_Real_Slim_Shady
Atrox drew this awesome image of the Keldori!
Posts: 1,011
|
Post by The_Wayward_Admiral on Jun 3, 2016 19:26:23 GMT
1) I love KSP even though I'm really bad at it (RIP Jebediahs 1 through 91)
2) What if, instead of the computer saying "okay, this cell is capable of doing this, this and this (and not of using Oxford Commas)", the cellular behavior editor stays on in Multicellular stage. Using the behavior editor, the player could assign a behavior for each individual cell type.
exempli gratia: when stimulated with electricity, this cell contracts quickly.
|
|
|
Post by rebeldroideka on Jun 10, 2016 5:42:15 GMT
How will different cell types interact within the colony if placed adjacent to one another, and will cells of the same type get a benefit from being directly adjacent?
|
|
|
Post by lavathor on Aug 10, 2016 3:47:36 GMT
So read through the first page, but skimmed the second. So sorry if it's already been mentioned.
I like the percentage function idea, where everything you do to a cell effects it. IE 47% bone, 12% nerve, so it become a crappy bone. I do agree that it becomes a game of "make the best cells" but is that really bad?
You could further this in making multiple differnt types. So that you could have dual function cells, or even a trade off so lower functional costs less, while higher costs more. Spent your pints on a few really good cells, or get a bunch of okay cells.
|
|