Hey there friends! I missed the boat to support Steve 14 years ago, as a result I’ve been out of the loop all this time.
Are there any juicy tidbits that have been shared over the years that I’m unlikely to know about? does anyone still have a copy of any old builds, or better yet an archive of the old forums that I can examine for myself?
Ooh! Good god! I didn’t think the forum was working – it worked for me, but last night it didn’t work for a different me, if you see what I mean. But you’ve proven me wrong!
Nobody else even knows about it yet, but I’ll write a blog post later and tell everyone, now that I know it’s working 🙂
It seems that there may still be problems with the forum but I’m going to add some info here that I’ve discovered on my own through poking around in the code. At least people will have something to read when they arrive!
Full disclosure, I wouldn’t call myself a programmer, i was a very long time ago and at the moment I’m learning C sharp and Unity to better understand the code. That being said, I have been able to follow a lot of what is happening in the symbiosis simulator.
What’s the symbiosis simulator, other than great alliteration, you say? It seems to be where the bulk of the calculations within the world happen (once Steve has the time I’m sure that he’ll enlighten us further, but here’s my take on it.) Creatures and objects in the Frampton Gurney aren’t built out of cells and molecules as they would be in reality nor is the Unity game engine designed to work that way, in this sense the symbiosis code is the layer (to use Steve’s terminology) between the simulated chemistry Steve has built and the simulated physics of Unity.
This allows for creatures, objects and the player to interact in vastly complex and varied ways but there is even more to it than that, for example it establishes a vocabulary replete with nouns, verbs, and adverbs enabling us to one day directly communicate with our phantasimals. Lots of nifty Brain stuff happens here that is more akin to representing electrical potentials of various systems and then polling them to form opinions and desires than it is to chemistry.
A side note is that in reality all chemistry is driven by electrical charges, but I can’t begrudge a man for simplifying the system (valent bonding and electron orbitals always messed me up in chemistry.)
With that out of the way, one of the things i initially found most accesible and interesting is that we have three different crops which can flower, cross polinate and mutate from generation to generation. Spotted mushrooms, round cacti, and tubers. There is an Item called Feedmill which appears to be designed to manufacture food pellets using the crops as feedstock and it will imbue them with a similar nutritional/chemical payload to the crops it receives as input. These crop mushrooms are a domestic varietal not to be confused with a separate entity known as Mycopatch which is also fully able to mutate and evolve.
I’m not sure if this was an implemented feature or something Steve worked on quietly, but there was a Gloop adoption center! if I’m understanding it correctly, you needed to load a gloop into a crate, then use a forklift to load the crate onto the train which would whisk the critter off to the cloud where it could then be adopted by another user but with all of its memories and experiences intact. I imagine that this is the basis of our upcoming village system as well.
I found the text strings from the adoption agency quite amusing so I’m going to share them here; “Bred by our experts at VanSoc Central, these creatures are generally very young and inexperienced. Each one is unique, though, so maybe this is the AI that’s finally going to be smart enough to take over the world, as we’ve long been promised! But then again, maybe not.”, “We couldnt wait to get rid of this one, to be honest. Maybe it will behave itself better in a new, loving home.”, “There’s not much we can tell you about these factory-bred vanimals, because they’ve not been with us very long. No guarantees, although we’ve done our best not to mess them up too badly.”, “All of our in-house creatures are bred from a small, select gene pool. They shouldn’t be too much trouble, and they make good vanimals for beginners. Maybe you’ll put this one back up for adoption when it has learned a thing or two.”, “Jenny said this one bit her hand once, but we don’t really believe that. They’re not that hard to avoid.”, “A bit of an ugly duckling, even by gloop standards, but someone will probably love them.”
There’s a lot more going on under the hood and I’m unsure how much of this is of interest to other people, so I’m going to leave it at that for now.
Cheers!
By all means, post your datamining… if Steve will let you! I’m a bit interested in the genetic inheritance. I might have thought blue and red eyes would make purple for some reason for example, but I got green in my save.
(I’m sorry if I’ve misinterpreted anything in this post, but I’m always open to corrections from anyone that knows better😊)
As I said earlier, I’m not a programmer, and I should also specify that I’ve never taken a biology course, thus I wouldn’t consider myself qualified yet to interpret the specifics of the genome ( like which gene encodes blue, red, or green eyes) but I can tell you a tiny bit about dominance in genetics as I understand it.
Creatures had a monoploid genome, meaning their “cells” only carried one copy of each specific gene and they each only had a single “chromosome”. This means that they only had space in the genome to account for their own eye colour whereas haploid (two chromosome?) genomes incorporate dominant and recessive genes, including a variant of each which is inherited from each parent respectively.
This meant that norns would inherit either mom or Dad’s gene for eye colour and Grandma nor Grandpa actually had a direct influence on the offspring beyond the child they reared (mom or dad).
When it comes to humans, we are haploid creatures which means we all have two chromosomes, X and Y in the case of males and two Xs in the case of females. We inherit one copy from each parent, just as our parents inherited one from each of their parents. The consequence of this is that we all carry two nearly complete genomes within our bodies!
There’s a lot more to it than that, but for clarity I’m going to oversimplify (in fact, I already have).
Even with that, this is where it gets complicated.
Generally, for each gene, there is a dominant or recessive variant. This interplay determines our “phenotype”, which is a fancy word to describe which copy of a gene (mom’s or Dad’s) gets used in construction and maintenance of your body. For example, whether you have brown or blue eyes.
Going deeper, there is epigenetics which has, to a certain extent, to do with environmental stimulus altering gene expression but it’s outside the scope of this answer.
Mom’s blue eyes we will call ‘a’ and Dad’s brown eyes (substituting brown for red as humans don’t have red eyes excepting albinism and other melanin disorders) we will call ‘A’ because in humans blue is recessive (lowercase ‘a’) and brown is dominant (uppercase’ A’).
In this scheme the are four possible combinations of ‘a'(blue) and ‘A'(brown). These would be ‘aa'(blue,blue), ‘aA'(blue,brown), ‘Aa'(brown, blue), And ‘AA'(brown,brown).
Dominant(uppercase) genes will always be expressed rather than recessive(lowercase) genes when both are present. I.e. “AA, Aa, aA” all result in brown eyes while only “aa” would yield a child with blue eyes.
Still with me?
DNA is made up of base pairs which are represented by couplings of enzymes represented by the initials G, A, C, T, B. These five enzymes encode everything we ever could be. Ultimately epigenetics will decide how strongly each and every gene will activate across each chromosome, that is to say, it’s dominance.
So in biology, the permutations of dominant and recessive across twenty-five possible variants of base pairs allow for nearly unlimited variations.
Bolly’s seem to have six “enzymes” which encode their phenotype, A though F, though they have nine chromosomes. Lots to dig into.
I’d love to elaborate more but I’m still getting my head around syntax and operands in C#. Rest assured, I’ll continue learning and digging. I might be in over my head now, but that’s how I learn best.
Also, to help me dig in and understand better, do you think you could share your bolly’s .ser files? I’ve barely run the game myself but I’m fully invested already.
They can be found (under Windows) in c:\users\*your name*\appdata\locallow\grandroids\phantasia\states\
Mine are all named bolly_cacfah8haeec***.ser
They are about 4.5mb each.
Hey there friends! I missed the boat to support Steve 14 years ago, as a result I’ve been out of the loop all this time.
Are there any juicy tidbits that have been shared over the years that I’m unlikely to know about? does anyone still have a copy of any old builds, or better yet an archive of the old forums that I can examine for myself?
the pre-pre alpha build would be the most interesting one, since it git a hughe world attached (to big) with a few hidden details all over the place. I am not sure if i still got it. Even a working train to travel around. The slide diesn’t work but it is nice to explore.
anyway how should i send it, if i still have it? Is here a private message system in the forum?
@genesis hmm, I imagine that the forum would support smaller files like the states “.ser” files I mention earlier.
I’m not sure which apps or operating system you are using, but I’d recommend uploading to Google drive, one drive, drop box etc. Then just share the link
@foggygoofball gotthe build from october 22, compleatlypublic shaing is something i would not do. There is a reason steve took down old builds. Private to YOU is a different story
Makes sense, how big is the archive? Maybe I can set up a private cloud storage you can upload to.
@foggygoofball i just need an email and i send it to you – you can edit your postz here and remove it afterwards, right now i am checking the forum so i would see it
Thanks again Mabus! I got the old version up and running now.
I’ve already been able to answer a couple of my questions by comparing the two. The way that the genomes and debug logs are stored is actually very enlightening. 😎
@foggygoofball if you could make a video where you explain the differences between LLMs and wahetever steve is doing in a highly specific way, that would be very helpfull. With something that explains the “magic” behind the game i could strike deeper and more efficient into the AI communitys to recruit new friends! (just make sure to include a link in the description to phantasia.life)
@genesis with LLM enthusiasts you might as well just feed Steve’s blog posts/replies about AI into one and regurgitate whatever that implies… Moonshot Kimi is probably the most bang for your environmental impact at this time. Even more lightweight than Deepseek I hear.
@kurgan that’s what i already do. But videos can have a higher traction than comments. I hope for a bigger convertion rate into friends with that.
Besides most of those LLM enjoyers just don’t know what they are missing. I just want a tool to help them better understand the shortcommings of LLM
