-
We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.
So I spent some time on my old planet simulation and made space-surface heat exchange and visualization for water flow and ground erosion. Going to do more - atmosphere (currently - none) and some textures. Then, probably, some gameplay, little more intensive than simearth.
You forgot the part where it's literally one of the worst search algorithms that exists and often gets stuck to local maxima. And evolution is far faaaaarrr from being among the most complex scientific-theories it's just a search algorithm, and not even a good one.
One example I always like to bring up in these sorts of discussions is the left recurrent laryngeal nerve. It's one of the nerves that controls your larynx, also known as the voice box. Instead of going directly from the base of the skull to the larynx, the shortest reasonable path, it instead travels down into your chest, loops around the most important artery in your body, and heads back up to your throat.
Why? Probably because that was the shortest path to the larynx in the ancient fish-like things which all life on earth call their ancestor, and still is the shortest path in fish today. As the length of the neck slowly increased over millennia, so too did the length of the nerve, but that isn't really a problem. It's just a minuscule increase in the amount of resources needed for any given organism to grow to adult size. Given that the nerve loops around the aorta, there's really no way for it to get fixed; it'd require completely redesigning that nerve's path, and evolution can only work through minuscule changes to existing structures.
Today, the left RLN in giraffes is approximately 4.6 meters long and it ends up mere centimetres from where it started. That's a horrendous engineering solution but a perfectly reasonable evolutionary solution.
- 1
I only know this stuff from the informatics/maths perspective so reading this almost makes me sad. It is, however, a great way to show how bad evolution actually is at doing it's job ( although humans are a nice design because we will be capable of transforming ourselves somewhere in 200 to 300 years, I guess )
They were a "modern" version of sim ant in the 2000's, I think "Ants Empire" in English? It came out of a series of French book. It was a bit clunky, but fun from memory.
Hey, you probably have a list of people already, but could I be one of the testers for your game?
Hi. I kinda dropped it or suspended. By several reasons - 1. it is not really a game, more like a model, more or less boring.
2. I should not have use CPU calculations instead of GPU calculations
3. I'm suck at 3d visualization.
I'm trying to force myself to do it simpler - as a water erosion sim on plane, not on sphere. Because there is no much sense in all those heat absorption calculations without fancy visualization of deserts or jungles or whatever.
Some screenshot - http://imgur.com/a/MjIcq
No, there is no list. And there is no product to test so far, sorry. Just some glitchy unfinished concept.
Last edited:
Well, keep on going at your own pace. It is better than anything I could ever come up with and I for one think it would be fun to mess around with it even if it is just a physics model in the end (see Universe Sandbox).
Good luck, and have some fun.
You're reducing it to a purely mathematical concept, but there are so many different aspects to evolution. For example, we can explain how sexual selection works with an algorithm, but how do we model the evolution of sexual selection from organisms which don't inherently have the property of sexual selection? Obvious a computer game wouldn't need to get that complex, but to be perfect it would need to. Otherwise, determinism arises in the model. There'd have to be the chance that sexual selection doesn't arise, and it's unknown is how unlikely a thing it is for primitive organisms to achieve.
Just like say, "the RLN was originally looking for the shortest distance between points", but really there shouldn't be any preset "what evolution is looking to achieve" present in primitive species. Every 'innovation' in evolution is an accident. Our ancestors were nonsensical blobs, who after thousands of generations of exposure to the environment and their own developing systems changed into fully functional, and sometimes increasingly complex, lifeforms.
If we look at two processes, monogamy and carnivorous behaviour, we see the links between their origins' complexities are completely different. We've got monogamy, where sexual selection is already well established, with many different processes leading up to monogamy - with the ultimate 'conclusion' that mates which have rigorously vetted each other are more likely to produce offspring more capable of mating (this is a great oversimplification, as the productiveness of a process or species has less to do with its suitability than we deterministically like to think). Then there's the origins of carnivorous behaviour, where there was no previous link between organisms only capable of consuming non-biological material, and the behaviour of consuming other forms of very similar biological life. The former might have taken millions of years to develop, with perhaps the latter taking hundreds of millions through incremental change. Then again, there might have been an environmental precedent for the latter to happen, and the process of change might have been much simpler than the lead-up to monogamy. It's these types of processes and all their underlying origins which would be hard to model, especially where the very attempt to model them in advance to resemble our own evolution is counter-productively deterministic.
Last edited:
https://www.technologyreview.com/s/414934/robots-evolve-the-ability-to-deceive/
How did they evolve the ability to deceive even though they were not capable of it at first? The answer is simple the search-algorithm does it's job it searches for new ways and improve old ones to achieve whatever the target is. Don't mistify evolution it is little more than a mathematical concept.
This has nothing to do with the origin of complex sexual selection. Mystify evolution? It's a highly layered series of processes built off of each other, and you want to explain it away with a single sentence that conforms to your specialized and unrelated education. I think you're moving the goalpost because you can't explain how to model the development of sexual selection.
And when I say complex selection, I mean the difference between mating with organism because it is there, and mating with organism because it is bigger. The step between those two points is enormous.
Last edited:
Perhaps I was not the only one, who enjoyed the concept of Spore, very much. I always found the game particularly fun, but I hated the game for some of the features it did not have, like more advanced diplomacy, more political features, and etc. The list goes on, and on for the things I always wanted in those games. Perhaps if a company, or somebody, wanted to develop such a concept, I would very well buy said game, but until then I shall always have Spore.
Get a grip, if you have no clue don't say X is unrelated. https://en.m.wikipedia.org/wiki/Mathematical_and_theoretical_biology
Yes, because that'd be huge and my interest area is machine-learning not second-class search-algorithms. Although the two are very very related. Read Kuijper if you wanna know how extremely important mathematics is for sexual-selection theory.
Mathematics is never unrelated, if you think it is yoj have no clue about mathematics.
Of course anything should be fundamentally explainable by mathematics, but you said yourself you have no idea about even the most basic anecdotes in biology. You're talking about your knowledge of computer science, and calling one little sub-discipline the end-all to a much more massive field of study. Evolution is not an algorithm, it is explained by algorithms (and most if not all models are meant to explain only factors of evolution). The two are only conceptually related. There is no set of rules which we know of which evolution must follow, especially in the most primitive lifeforms. It has no direction, no purpose. There's no reason to ascribe deterministic traits to a natural process, which is only a "process" in the most categorical overgeneralisation. The fact that you think we can ascribe rules to the murky darkness of lifeforms that went extinct billions of years ago, whose origins will always remain unknown until we observe the formulation of life ourselves, is stupefying.
Complex processes in evolution had a beginning out of non-processes, and that's what I've been talking about. Origins are hard to model when we only have a flimsy grasp on the details, and modeling the origins could be very boring when they lead to dead ends. You have to have some basic knowledge of biology to understand that.
It's clear to me that your objection to the idea that evolution is a complex topic is based merely on some sort of religious agenda. You seem to believe you're in competition to make something better than this natural process. Frankly, I'd be surprised if another person has EVER uttered the sentence "evolution is a simple search algorithm". You're probably taking a metaphor literally here. Who exactly set bacteria to search for.. what, exactly? Life does not 'look' to become more complex. Do you also believe a stalactite growing is a 'search algorithm'?
Last edited:
Hmm would quite like it if Paradox done a Sim Earth type game, especially if they did a Save Converter for Stellaris
Intriguing, idea, playing as your creature through the first couple of eras, then converting over to Stellaris, but I would say they would need to make a DLC, to have your character look like like your character from the last game.
Gods, we could really make a species of Ewoks or Wookies then. I can just imagine the horrors that would await in the depths of space once the save was done. On another note, I like your quotes.
Oh thank you, and imagine recreating the whole of Star Wars characters with such a thing.