Saturday, July 23, 2016

Portals And Bugs And Doors, Oh My

All right, this is a log I'm keeping in Notepad as I watch, pause to type, and unpause to watch again, in more or less 'real time'. I will post these words with illustrations to Addicted to CAOS later; if you're reading these words, obviously, it's later! :p

Dreameater visits Dusk in the C12DS house.

Since unpausing I have remained focused on Dreameater, since she's been in the game less time than Dusk and since Dusk still appears to be focused on that nonexisting door I mentioned last time. At first she seemed pretty reality-oriented, but for the first time her attention has begun to waver. She also noticed Dusk for the first time and made her way toward him, thinking 'pull Norn'. However, on her approach, she also, briefly, thought 'push grendel home' as shown by the CRC, while the HealthBar showed 'push self'. Aha! Her first hallucination that lasted long enough to be detected and recorded.

Unlike Dusk, who experienced a few different early hallucinations before becoming 'stuck' on one, Dreameater has been alternating between her desire to play with Dusk and her desire to check out that imaginary grendel home. She made her way to Dusk, tickled him, then began to make her way out into the garden while thinking 'push grendel home' again. But after going out a little ways she turned back toward the house, once more thinking 'pull Norn'.  Her thoughts have continued to vacillate now and then between her fellow Norn and that tempting yet nonexisting grendel home.

Now she's just hanging out near Dusk while thinking 'rest vehicle' on occasion... nope, now she's focusing strictly on rest vehicle, while lying right beside Dusk, who has turned sideways to face away from her but has not moved away.

Suddenly she stood up and began to retreat from Dusk, running out into the garden with a big grin on her face - this always seems to me like they want the other Norn to chase them, but I may be reading too much into that.

Dreameater thinks there is a flower here. There are flowers here, but the one she's focusing on is not. She is looking at a non-flower. A flower of unbeing. An unflower, if you will.

Oh, now this is odd. She lay down briefly in the garden thinking 'rest flower', and it would be easy to think she's looking at a real flower - but she isn't. The CRC and HealthBar show that she is in fact hallucinating a flower.

This is one of the interesting things about chemical 130 and the Norns' reactions to it. Sometimes the hallucinations are random and out of place, but other times they are oddly congruent with the Norn's expectations. I have even observed angry Norns who had never met an actual grendel hallucinate a grendel and then try to hit it, after other Norns suggested they 'hit grendel'! It's almost as if hearing the other Norns say 'hit Grendel' triggered her brain to produce an image of a Grendel, which in turn caused the angry Norn to react as he or she would to a real Grendel. I mean, this happened at least two or three times that I witnessed. It was really interesting. It's also common for them to hallucinate critters when hungry; fortunately the prey scent agent I made helps them not get too stuck on imaginary prey in favor of going after real prey.

I decided to leave Dreameater resting while meditating on her imaginary flower, and check on Dusk. He was still obsessing over the imaginary door, and I was getting a bit worried about him. So I decided to discourage this particular fixation by giving him a single slap on the rear, since verbal encouragement to try other things had no effect (he always just responds with 'Dusk push door') and tickling would just reinforce the fixation.

Notice that the green indicator arrow is visible above an actual item of food, but the CRC says '<nothing> food', meaning that Dusk is not actually physically using his 'look' neuron with the 'food' neuron - only the 'food' neuron is active in reality. But the HealthBar says 'look food'. The best I can make of this sort of thing is that it seems to mean that the Norn believes it is taking an action regarding a real object - in this case, looking at the food - when he actually isn't. Which, if I'm correct, is extremely surreal.

Thankfully, the spank snapped him out of it, at least with regard to the door. He briefly returned to reality, but then did that weird thing where he seemingly thought he was doing an action he wasn't really doing, so his grasp on reality was clearly tenuous. And within a few more seconds, he lay back down on the ground just inside the house and began obsessing over another hallucination.

Dusk meditates on a seed that only he can see while Dreamy plays with a hi-bounce ball in the garden.

Around the time Dusk lay down and began to meditate on the unreal seed, Dreameater suddenly got up, presumably abandoning her unreal flower, and began to play in the garden with one of those multicolored hi-bounce balls. Since Dusk is probably going to stay where he is for the moment, I'm going to switch to watching her again for a little while and see what she's up to. Dusk seems to experience pretty fixed and consistent illusions which draw his attention strongly for long periods, whereas Dreamy seems more impulsive and less prone to fixate on one thing (real or not) for as long as Dusk can. She does the sit-and-stare thing, too, but she seems to lose interest quicker than he does. I've never had to spank her to get her moving after several minutes of inactivity - at least, not yet!

So I unpaused and switched back to Dreameater. Dusk fell asleep shortly thereafter. At first, she was interested in playing still, heading toward and pushing the nearest hi-bounce ball. Then, intriguingly, she lay down and began to 'rest flower' again, only this time she was focusing on a real flower. She fell asleep within about ten to fifteen seconds of lying down; if she keeps doing this, she's going to develop a fondness for 'rest flower'; that is, she will prefer to lie down near flowers and/or think about flowers while resting, as a way of lulling herself to sleep.

This is a great example of how their hallucinations can sometimes influence the development of their personalities. This preference for 'rest flower' likely began because of that initial decision to react to that flower hallucination by resting near it. I don't think hallucinations give stims, so that in itself was probably not enough to teach her to like 'rest flower' - but the next time she felt like resting, there happened to be a flower nearby, and that made it more likely that she would decide to 'rest flower' again.

I'm not sure that got her any good stims even with the real flower, come to think of it, but the point is, the more often she does it and finds it pleasant or at least not UNpleasant, the more likely that she will continue to do it. And thus the habit of lying down beside and thinking about flowers while resting will take form. If that happens, she will eventually become a Norn that, when she hears another Norn complain of being tired, will suggest that the other Norn 'maybe rest flower'. In short, she will have developed a distinctive personality trait. This sort of thing can happen in normal Norns too, of course, but I find it interesting how often the 130 Norns are able to generalize things learned from interacting with hallucinations into their reality even if they seem to provide no meaningful benefit.

While Dreameater fell asleep and began to dream about potions and other things, I noticed Dusk had woken up and was talking a lot about 'push portal' and 'Dusk push portal'. There's no real portal present, so I'm pretty sure Dusk's early portal hallucination has returned. I'm about to unpause and switch back to him to confirm that, but it seems pretty likely from his bibbling at the moment.

Yup - he's fixated on an imaginary portal, and seems quite frustrated by it. He said 'push portal' three or four times before losing interest. Then he played with a real gadget and is now back to the 'rest norn home' hallucination. He fell asleep again, but only briefly. Now he's resting Norn, only he's not looking at Dreamwater... so he's hallucinating the presence of a third Norn. This is the sort of thing that intrigues me the most. And back to 'rest norn home' - not the real Norn Home he's right near, but an imaginary one.

He just seized and ate a dreameater bird, thankfully - still able to take care of himself despite his mental illness, for the most part, it would seem.

Meanwhile, Dreameater woke up and returned to the house area to be with Dusk. She's now lying down fixating on a hallucination of her own... let's go see what it is while Dusk naps again. (Oh! Just saw Dusk fly a little bit for the first time - he's become a child.)

Dusk figures out how to use his wings for the first time as he ages to the Child life stage! Notice that although he appears to be retreating he is in fact approaching a nonexisting portal in an attempt to push it - still! The ability to fly has only caused him to begin advancing on the "portal" rather than standing or lying still and staring at it while thinking 'push portal' longingly. 

While Dusk enjoys his first flight, floating cautiously through the house, Dreameater stared at, then began to pursue, an unreal bug of some sort. Dusk flew out of the house and into the music room area, following some hallucination of his own - it's going to be hard to track both of them now. But that's all right. Better that than having their results skewed from serious loneliness.

Poor Dreameater is really stuck on this invisible bug. I decide to help her by bringing a real bug or critter nearby for her to eat instead. Unfortunately she ignored the Balloon Pigeon I waved in her face. She's got her head on straight though, generally - she heard Dusk complain of boredom and told him to push toy.

She's still obsessing over that nonexisting bug. This is her first really strong fixation so far.

Meanwhile Dusk is chasing his unreal portal, hovering in the music room and complaining of boredom. Soon he'll get bored enough to try something other than 'push portal', but I've noticed that 130 Norns tend to be very persistent about their illusions when they concern things of great interest. And for my Dragon Norns, portals are fascinating objects - less compelling than food and toys, but not by a huge degree. So I can understand why a portal-related illusion would be really attractive for the poor guy.

For a while he seemed to get tired of it, switching to 'rest critter' while focusing on a real dreameater bird, but now he's back to the portal, although I notice that little green indicator arrow keeps winking in and out of existence over the bird, meaning he's alternating his attention between the real critter and the imaginary portal very quickly. I suspect that once Dusk gets hungry, that bird is going to end up in his tummy.

Meanwhile, it seems Dreameater has also become a child and learned to fly. (I took a screenshot of this but evidently forgot to save it... boo!) She's moved out into the garden and appears to be flying upward, though the game is paused as I type this. Let me unpause and see what she's thinking about. I unpause and it turns out she's chasing a butterfly - a genuine bug! She even said, "Dreameater eat bug"! Hooray! I'm glad she's finally been able to bring her attention onto real prey. She ate it, but that wasn't enough, because she complained of being hungry for starch still, so I suggested she eat a critter. At first she wanted to push critter instead, but soon enough she chased, seized, and ate a dreameater bird. Now she's 'resting vehicle' while looking at a nonexistent vehicle - oop, nevermind, she's up and flying and playing with a toy now. She's a quick mover, zippier than Dusk tends to be.

This is more or less where we're leaving them as I pause and close their world for the moment. They did not seem to notice the Grendel before I paused the game...  I can't speak for Dusk, but Dreamy was apparently too busy 'rest flower'ing again. 

Whew! I don't want this to get too long so I'm going to wrap up for now and close their world so I can rest myself and let one of the wolfling tester worlds run for a while. But we'll return to Dreameater and Dusk soon enough, and discover how they handle meeting a C1-type Grendel for the first time, among other things. I'm not used to 'nuture' style gameplay so I'm sorry if these posts aren't as interesting as those of people who routinely write and play this way, but hopefully I'll improve with practice. For now, though, we'll leave Dusk and Dreameater to rest.

Friday, July 22, 2016

So let me break you down on a molecular level

My two testers, Dusk and Dreameater. Dreameater has become interested in the clock. At least she's paying attention to reality; Dusk has been standing there obsessing over a nonexisting door for a few minutes now.

While the rest of the world is having fun with their cellphones and their Pokemon Go, my non-cellphone-owning self has had to find other things to do. One of those things is a somewhat more formal study of the 130 Dragons. For those who may not have read earlier posts, this is a version of my ongoing Dragon Norn genetic breed project that has been given a 'Pineal Gland' organ with a 'Psychosis Generator' emitter that produces chemical 130 at all times from the moment of birth. In short, they are constantly hallucinating and therefore 'insane', at least as much as a Norn can be said to be such.

I have done wolfling runs both of pure 130 Dragons and mixed normals with 130s, and both have been very interesting. I have written on the Creatures Caves forums and elsewhere, a few times, of more interesting incidents. But I decided I needed to do a somewhat more formal, or at least closer, study on these odd Norns and the effect that hallucinating their entire life has on their learning and personality.

Ideally, I would raise only one Norn at a time so I could closely watch their hallucinations every moment. But these Norns are more social than normal and would not thrive kept alone. Plus, social interaction is important even to normal Norns, given they can give each other advice, and the advice is an important way to gauge what they've learned about given subjects. So I decided, while keeping numbers minimal, that I'd have to have at least two of them. Given my need to watch them closely and probably nurture them some, I decided to hand name them, though I am using the Autovocab agent to keep from having to teach them language all the time.

I hatched the male first and named him Dusk for his colors. At first, he was fairly sensible, focusing on real things. Like most newborn Dragons, he quickly found a critter and ate it, then began looking for ways to alleviate his boredom. This is when his 130-ness began to show:

Dusk thinks there is a door here. There is no door here. Poor Dusk.

Here's a list of his first few hallucinations. These didn't all happen at once - there were periods of normal focus between them - but this is the sequence in which they occurred.

rest norn home
eat critter
push door
push portal
push door

At 'push door' he got stuck; he alternated between 'push door' and 'push portal' a couple times before deciding the imaginary door was more interesting than the imaginary portal. As far as I can tell he has fixated continuously on that nonexisting door without ceasing for at least a few minutes' worth of game time now.

As you can see from the screenshot above, I can tell when they're hallucinating by comparing the output of the CRC agent with that of the HealthBar agent. When HealthBar and CRC show different outputs, the Norn is responding to a hallucination, and the content of the hallucination will be shown by the CRC. 

I have also noticed times where the HealthBar will say 'action thing' while the CRC just says '<nothing> thing', which I'm not sure how to interpret; for example, HealthBar might say 'push Norn' while CRC says '<nothing> Norn'. In normal creatures, the CRC will display '<nothing> thing' when the creature is asleep and dreaming, or when it is awake and thinking about a given object but not considering any action in relation to that object. In short, they might think about food without thinking about eating it - just the concept of food in itself is crossing their mind. At least, that's how I interpret that in normal Norns. In normal Norns, when this happens, the HealthBar will typically display whatever the Norn is actually doing at that moment - if it's asleep and dreaming, for example, it'll say 'quiesce self' while the CRC displays part of the dream content in the form of '<nothing> object'. If the Norn is awake and thinking about an object but not an action the HealthBar will say 'quiesce object' while CRC says '<nothing> object'.

But in the hallucinating Norns I'm not sure how to interpret the feedback. My best guess is that the Norn is hallucinating taking an action it isn't really taking, even though it's focusing on an object that really does exist? But I'm not sure if that's possible. (Evol, if you ever read this, I'd love your opinion on that point!)

Well, once Dusk got himself fixated on the imaginary door, I decided it was time to hatch his female companion. Within moments of hatching, she seized and ate a dreameater bird critter, so her name became Dreameater. 

Dreameater has shown more interest in the real world than Dusk, and has not registered any but very fleeting hallucination awareness so far, though she may be experiencing hallucinations she's just not paying attention to (i.e., she's ignoring them in favor of paying attention to more interesting stuff which happens, so far, to be real stuff.) However, her methods for interacting with the real world are a bit unusual. Here she seems to be attempting to push the dodgeball against or into the refrigerator somehow? She kept reaching while 'pushing' at the same time. 

The screenshot above shows how much Dreameater likes to play. Although I know she is experiencing the same degree of hallucinatory symptoms as Dusk due to them both having identical genetics, already they are beginning to show personality differences. It did not take long for Dusk to become obsessed by a hallucination, while Dreameater has so far in her short life found the real world more engaging. She does like to do odd things, though, which I suspect is a subtle sign of her hallucinosis.

So far, the two Norns have yet to interact with each other directly. Autovocab hasn't kicked in since Dreameater was born yet so she's still not really talking, though she's starting to learn 'push' from my attempts to get Dusk to 'push toy' instead of staring at his imaginary wall while complaining of boredom. The game is paused as I type this, but I'm going to pause my typing and unpause the game for a little while. I'll be back with more on these two!

Wednesday, July 13, 2016

The Unique Scent of Prey

Icebound Hole, a Dragon Norn tester with a funny Protective Tub name, has caught a Balloon Pigeon. I've been testing the removal of the old 'hunting' stuff in their genome thanks to my new Prey Scent agent. I forget if his Gen1 ancestors are from the 'before' or 'after' batches, but it looks like he finds 'get critter' to be quite entertaining, so he probably still has the old 'chase' and 'get' instincts active.

Dragons are predators. Well before I began working with their sprites and a similar genetic concept, the original Draconian Norns were set up to seek out bugs and critters for food instead of the normal food, seed, and fruit class objects. I've changed the way my Dragon Norns' digestion and such works - they're a lot more interested in food type objects than the old Draconians were, for example. Also, they can survive entirely on just critters OR bugs, one or the other, and no other edibles, due to both triggering Ate an Animal and my setting up a chemical reaction that lets them get all three types of nourishment from that one stimulus. In a way they are more carnivorous than the originals, having no need to seek out a balanced diet.

However, being carnivores has always meant some sacrifices and constraints. They are not very good navigators compared to normal Norns. Part of this is due to the brain changes in CFF/TWBs, which make their thoughts a little more random and thus more creative/explorative but with less ability to stay focused on a long-term idea such as following a scent to its source. The Hunger Overwhelmsion stuff helps, but not enough. Some people, like Grendel Man, find CFF/TWBs stupider than average; I prefer to see it as a sort of Norn ADD. :p Having ADD myself, I can sympathize! This trait allows them to have more chance to develop personality and not be so overwhelmed by their instincts if their lifespan is long enough. Since dragons are very long-lived, I like these traits in them. But I can see why others would not, especially if working with much shorter lived breeds.

However, with that said, most CFF/TWB Norns with a normal diet will still be able to navigate in my experience, just with less focus compared to a typical base Norn. They can still smell food and still know that they should head toward food when hungry. They just aren't as able to keep focused on that idea when there are distractions around (meaning anything else they could possibly think of to do or try LOL) But they usually don't get hideously lost and starve to death in corners of the ship. It does happen from time to time, but it's not such a problem that I feel like I have to trap them in 'friendlier' metarooms to keep them from getting themselves lost.

Alas, this is not so with my Dragon Norns. I'm not QUITE as strict about that as I used to be, but it is pretty normal for me to essentially set up either one huge metaroom (Biodome or C12DS, say) or else two or three not-so-huge ones (Flooded Norn Terrarium and the Meso, say, or Deep Abyss and Meso plus Norngarden 1), release bugs and critters and place extra toys and otherwise get the places really set up nicely, then place the eggs and let them go. They'll have Offline Portals or something to travel between the rooms. But I'll lock the doors to any official metarooms being used, and sometimes use Door Hiders on both sides, too. That way they have a lot of room to roam and explore but no chance of getting into a place where they can get lost too far from food.

In addition to the CFF/TWB 'randomness factor', I have given my Dragon Norns a general tendency to wander when bored. Essentially, the "I am Travelling" stimulus has a boosted Boredom reduction compared to normal CFF/TWB, and I think I also tacked on a tiny, tiny bit of Reward, too. The result is a much-reduced tendency to clump up in one place. They still tend to do that, but the behavior is much more personality based now. Family/friend groups will tend to form in certain spots, like the C12DS garden area, but some Norns will spend their whole lives in such groups while other individuals will roam far and wide, visiting the groups but then taking off again on their own. I've also noticed a tendency for them to mate off and roam together in male-female pairs. Pair bonding is not something I was trying to create, and they're only slightly more monogamous than other Norns, but it's something I notice fairly commonly.

This wandering tendency means that they're very good at satisfying their needs and taking care of themselves to a certain degree. If bored, they will roam to entertain themselves, and the act of moving will eventually take them to a toy or something they can then play with, or other Norns to interact with. The problem though is that boredom isn't fatal. It's fine for them to wander aimlessly in boredom until they find a better way to relieve it than simple roaming.

However, this doesn't work as well for food. Even in a huge metaroom where the food spreads out a lot, the wandering will -usually- get them close enough to food to see it, and if they see food when hungry they will eat it. They are very good at taking care of themselves so long as the environment is stocked well enough to put food within casual wandering distance at all times. However, if they get out into the corridors of the Shee starship, for example, they can very easily get lost and thus starve to death before the random wandering takes them within sight range of food.

I have also, many times, witnessed flying or swimming Dragons 'pacing' back and forth with fish or birds nearby, but too far above their heads or below their feet for them to see. It was as if they could tell the prey was near, somehow, but couldn't pinpoint their location by sight, and so could not find them.

This was happening because critters and bugs do not have a unique scent in the default game. In fact I suspect most do not have any scent at all. Some critters, such as the Marine Terrarium fish, apparently give off Protein scent by default, and that should have helped, but it didn't, at least not enough to make a difference. Probably because Protein scent is primarily associated with fruit, which is not something the Dragons have any instinct to seek out when hungry. (Some of them learn to enjoy eating fruit regardless, but even they can't scent track it efficiently because the brain tracts related to scent tracking aren't activated by the concept of fruit since I changed the instincts and stimuli to relate to bugs and critters instead.)

Vivisection Identity is about to raid the nest. At least, she thinks she is. As far as I can tell those yellow birds aren't instinct friendly. I know that the Albian Carrot Beetles definitely aren't - they regularly cause major confusion in the poor Norns who like to hang out in the garden. I often have to take them away and put them in some improbable place so they'll stop getting in the way of the Norns' normal feeding. I'll have to fix that sometime. 

Now, when I say 'scent', I'm referring to CAs. CAs are not smells, per se. They are also used to represent such things as light and heat. Their most obvious application in usual gameplay, though, is the smells, since light and heat are invisible and most Norns, Grendels, and Ettins do not react much if at all to light and heat. (TWBs do react to heat, and it's a big deal for them, meaning it matters to Dragon Norns since they're modified TWBs, but I'm not sure it matters much to any other genetic groups - except perhaps the Magma and Fire Norns, who use a different system of heat detection if I understand correctly anyway.)

Normally, a hungry Norn will look around for something it instinctively knows is edible - fruit, food, or seeds. If it doesn't see any such things, its brain automatically checks to see if there are any relevant CA values in the room the Norn is currently in (not metaroom, but the 'rooms' making up the metaroom - you can't see them under normal circumstances but there are ways to make them visible). In a metaroom, each room has various  numbers assigned to it that reflect the presence of different CAs. Objects are set to EMIT (OH LOOK SOME CAOS IN MY BLOG took long enough *ahem*) a set amount of a single scent (it seems like a given thing can only emit one scent at a time, but more than one thing can be assigned to emit a given scent), ranging from .1 to .9 typically. In the room where the agent emitting the CA is emitted, the value for that CA is equal to the value emitted by the agent. In rooms immediately adjacent, the number is divided somehow (I don't know the exact math involved) to lower it a little. The higher the number, the more 'rooms' it takes for the number to drop to zero. In short, the stronger the smell, the farther it carries before it 'dissipates'.

So, if a Norn's brain detects that the CA associated with something it knows is edible (say, Protein CA for fruit) is at a value higher than zero in the current room, it knows that fruit is somewhere nearby. As it moves around it will avoid directions in which the CA number shrinks, and move in directions in which it gets larger. This leads to a very direct and purposeful march toward the location of the fruit (assuming the CA links are set up correctly, but that's a topic for another post).

There's a common procedure people use called an IQ test, but IMO it tests the Norn's scent tracking ability more than intelligence per se. I do not consider my Dragons to be stupid at all - I've seen them do some really interesting and complex learned behaviors. But they fail that IQ test every single time. I mean like, 100%. No exaggeration. It's because they can't scent navigate at all whatsoever in a normal world. This leaves them desperately trying to figure out how to interact with their environment to find food. They try plenty of solutions, in a way I consider quite intelligent. I'd do the same thing if I were trapped in an empty room and slowly starving to death, after all, and I'm a person with an IQ in the 130-150 range. I am not at all stupid. But when you're in a low-resource situation, alone, you have no idea what's going on, and you're panicking, you try everything you can think of even if it seems to make no sense, just in case it might work. And that's exactly what I see my Dragons doing. So they don't seem stupid to me at all. Just stuck in a situation they don't have the resources to deal with.

Of course, it doesn't help that I made their baby stage a lot more realistic. Most Norns die in 30 minutes or so without food. Dragons die in about four minutes. So they have a LOT less time to figure it out. But even if I give them a half hour, most of them won't make it. They'll get stuck trying to figure out how to make the teleporter work, typically. Again, reasonable behavior that makes sense given their genes (they instinctively want to use a teleporter or portal if excessively lonely, crowded, or bored). But not helpful at all in that situation. And nothing about the situation discourages them from trying alternate idea after alternate idea ("OK, the teleporter won't work no matter what I do to it... let's try hitting, eating, pushing, pulling, and staring at the elevator now.") until they starve to death.

Again... reminds me a lot of  having ADD! I'm bad at remembering to eat when I'm stuck on a problem too. :p That's that 'randomness' bit of the CFF/TWB genome at work, that part of it. But if they could scent navigate, they'd have a much better chance of making their way out of the Workshop and into the Meso, following those ever-increasing CA counts.

Recently I hit on the idea of giving bugs and critters their own scent. Unfortunately there was no totally unused scent available; number 9 was given to plants and flowers by the Butterfly Norn agent years ago. And somewhat more recently, 19 was given to toys by the New Toy Smell agent. But I decided that in my world it was more important for them to be able to find food by scent than toys, given what I mentioned earlier about the relative deadliness of boredom versus hunger. They rarely have much trouble entertaining themselves anyway - they love to interact with their environment in whatever way they can. If there are no toys around, they'll stand there gleefully hitting, pushing, and pulling whatever is available. Like, anything.

So I'm not worried about them being excessively bored. I made them prone to high boredom for a reason and I like the way it encourages them to develop their personalities and preferences over time.

But having to essentially baby them by locking them into at least moderately favorable environments annoyed me. I want them to be pretty tough and survivable, but if wandering too far out of range of food is generally deadly, that doesn't really match my design specs. So I began thinking about making an agent using CA 19. I'd have to stop using New Toy Smell, but like I said, as much as I like that agent, I'd rather have prey smell where my dragons are concerned.

So with the help of the awesome people on Creatures Caves' forums, I put together my first working agent! It's very simple, but I'm pretty proud of it. Right now it fires way more often than I intended and I need to fix that, but otherwise it seems to be working exactly as intended. I also had to edit a .cos file to make it work properly, but that wasn't too hard.

Here's how it works. I started with the injection script.

INST to install something new! I think that's what that means, anyway. That's how I read it at least.

Then two global commands. This is what I also put into that .cos file I edited, but I wanted to include them in the agent script too just in case the .cos edit didn't work, and I edited the .cos in case this bit of the agent didn't work. Thoroughly redundant and very amateur-coder-ish, I know, but sometimes anxiety makes me do weird things. Point is, I know at least one worked. XD

Anyway, the globals in question are CACL. This is the bit that makes the Norn brains expect to associate a given CA with a given item class by default. So in this case, it looks like this:

cacl 2 13 0 19
cacl 2 15 0 19

That says to the engine, "All bugs and critters should emit scent 19." In turn, all Norn brains will have that expectation from the moment of birth. Whether it MATTERS to them or not is another question - that's a genetic thing - but this is what allows me to make hunting and feeding-related genes involving CA 19 and expect them to work. Without that brain association, I could have the animal agents emitting lots of CA 19 and I could have given the Norns instincts and stimuli related to CA 19, but without knowing that CA 19 means 'animal', a vital link in the chain is missing. The Norn's brain doesn't associate that CA with anything in particular, so the instincts and stimuli won't work right. I'm not sure exactly what would happen, since I've never put my Norns in that situation, granted - but I'm 99% sure it would result in either a lot of Norn confusion or else there would be no visible effect at all from adding the EMITs and the genes.

ENUM with the agent's classifier, KILL TARG, and NEXT...

Now, the agent itself! NEW:SIMP to create a new simple object, then a classifier in the 1 1 class and that blank sprite. Since this agent's only job is to make bugs and critters smelly, it doesn't need to look like anything or be perceptible to Norns.

Then... well it's supposed to say TICK 2500, to make it fire about every two and a half minutes or so. But I typo'd and wrote TICK 500. So right now it's firing about every 20 seconds. Whoops. Proofread your script! XD Someone else had to point that out to me. I'd totally missed it. It's probably contributing to some lag at the moment, albeit not so much that I notice compared to the lag produced by the number of Norns I have active right now. Still, I intend to fix that to its intended value.

Then comes MVSF. That's where the engine should put the new agent. I copied this from another agent - I think the New Toy Smell agent actually. If I read it right it places the new agent wherever the Hand happens to be at the moment of injection, but I might be wrong. It doesn't entirely matter where this particular agent physically goes, anyway, since its function will work no matter where it is on the map.

SLOW comes next, and that's another bit of the copied code so I forget what that does. I'd look it up but it's 4:30 in the morning as I type this and I'm feeling distinctly lazy. Sorry! :p I'll edit this later to fix it if I don't forget.

And that's the installation script. Pretty easy, right? You have no idea how proud I am of myself for figuring all this stuff out. For the longest time, CAOS would just dissolve into a big, incomprehensible mess when I looked at it, like numbers tend to. But I'm trying to learn to think of them as words instead of number-like things, and that's been helping.

Anyway, so far all we've done is make the agent. We haven't told it to do anything yet. So here's the timer script.

scrp 1 1 36100 9
enum 2 13 0
emit 19 0.9
next
enum 2 15 0
emit 19 0.9
next
endm

SCRP tells the engine to associate the code to follow with the agent we just created, using its classifier (1 1 36100 in this case.)

The ENUMs are telling the engine which agents are to be affected by the script. In this case, it's critters and bugs.

EMIT is the core of the agent. This is the bit that tells the engine to make the specified agents produce a CA. In this case, I chose to have CA 19 emitted at a rate of .9. This is very high, but I did that on purpose to make up for the fact that CFF/TWBs have a harder time focusing than normal. (I realized later that because bugs and critters move around, navigating to them is harder to begin with, another reason to make them smellier so even Norns that are pretty far away from them can still get moving in the right general direction.)

NEXT of course moves us to the next operation.

ENDM closes this part of the script.

I'll probably want to remove it at some point, if only to replace it with an updated version of the script. And what if I release this to other people? They may want to swap it out for the New Toy Smell or something else at some point. Or they just might want to get rid of it. So it needs a remove script:

rscr
enum 1 1 36100
kill targ
next
scrx 1 1 36100 9
enum 2 13 0
emit 19 0
next
enum 2 15 0
emit 19 0
next

RSCR tells the engine this is a remove script, of course.

ENUM, as noted before, specifies which script is to be affected.

KILL TARG and SCRX destroy the agent. To be thorough, we also specifically tell the engine to make bugs and critters emit CA 19 at a rate of 0 - in short, to stop emitting it.

And that's it! It's a very simple thing, but I'm crazy proud of myself for putting it together. I've tried a few CAOS projects before, but I'd always end up getting lost in stuff I couldn't make sense of. This is the first time I've been able to read it all and keep reading it and have it not fall apart. Take that, dyscalculia! >D

And how does it work? Like I mentioned, since bugs and critters move around, they're harder to use as a navigational tool than food, fruit, or seeds. So they're only somewhat better at that IQ test. But they no longer do that confused 'pacing' thing - they rise up or drop down confidently to snatch previously out of reach birds and fish. And the flying ones are making much more use of their ability. Though since installing C12DS they still spend a lot of time on the ground, more individuals are spending more time in the air, and flying to greater heights more often. This is often driven by prey-seeking behavior. Even when not hungry they like to follow flying birds or fish and seize them. This is less evident in the ones with the old hunting instincts silenced, but they still exhibit a distinct tendency to do that, which I interpret as them following the scent. Their genes still need some tweaking to balance Hunger Overwhelmsion and suchlike, but the agent seems to be working more or less the way I hoped it would.

So I'm pretty happy. And so are they! :)

Poor Lectern Celerity. It's a tossup whether he'll starve to death before his organs fail. But either way he'll probably hit 14 hours, which is about the most a Dragon Norn can ever hope for, so he was a very successful Norn all things considered! I just wish he would leave those yellow birds alone and eat a Balloon Pigeon.

EDIT: After finishing this post, I went to look at my game and found a Dragon - an old male, over 13 hours old - chasing the yellow birds in the sky above the garden. He'll die pretty soon. His brain is at like 20-some percent. So he's not getting the hint and moving on like the younger ones usually do. I felt bad for him and tried to redirect his attention by moving the yellow birds out of his eyesight and putting a Balloon Pigeon close to him. Before, that always used to work. But when I moved that bird he was on it like a shot! He followed it like he had an invisible leash attached to him. I was really surprised, and experimented. Sure enough, unless I moved so fast it whipped out of sight before he could decide to follow it (and thus got distracted by a closer bird before he -could- decide to follow), he would follow the bird he'd originally focused on no matter where I moved it. Fascinating! The agent is definitely working. It also suggests a new application I hadn't thought of - I could make a hungry Dragon Norn go anywhere by leading it carrot-on-a-stick style with a critter or bug in the hand!


Monday, July 11, 2016

A Most Forlorn Norn

Sooo I spent today working on my dragon Norns for the most part, tinkering with little aspects of their genome. I also made my very first working agent today! I'm finally beginning to understand CAOS! I'll make a proper post about that later, because I'm pretty proud of it even if it's a very simple thing. For now I've got something else to cover.

That poor C2 Norn from the post before last has been analyzed. A couple of people were kind enough to point me toward Slink's D-DNA Analyzer, which helped greatly in deciphering the mystery of this Norn's problems. I was going to do this tomorrow, after sleeping... but heck with it! Let's dive in. File 1 is the mutant Gen2 Norn, and File 2 is what I used to create her Gen1 parents.

Firstly, as I suspected, she is suffering from a brain mutation:

621 Different in File 1   6   0 Emb   B MutDup      128  Lobe #= 7 State Rule: state TRUE type0 PLUS type1 <end> <end> <end> <end> <end> <end> <end>, WTA= 1 Relax State Rule:  <end> <end> <end> <end> 0 <end> <end> <end> <end> <end> <end> <end>
621 Different in File 2   6   0 Emb   B MutDup      128  Lobe #= 7 State Rule: state PLUS type0 PLUS type1 <end> <end> <end> <end> <end> <end> <end>, WTA= 1 Relax State Rule:  <end> <end> <end> <end> <end> <end> <end> <end> <end> <end> <end> <end>

I am just beginning to learn how the Creatures brains work, so I can't parse out the exact meaning of this change. It may not have any real meaning beyond reducing a once-functional gene into muddled gibberish. The important point is that the seventh lobe of her brain simply does not work. Recall that I mentioned that she never tries to move or do anything other than speak and never looks away from herself? This right here is the culprit. The seventh lobe in C2 is the Attention lobe. With a damaged or broken Attention lobe, she is unable to focus on anything and so by default looks only at herself. Either that or the mutation actually forces her to fixate on herself - unsure whether it's specified or a default reaction. Regardless, this is serious brain damage. I will probably never re-import her because of this alone. She just has zero chance whatsoever of having anything resembling an enjoyable Nornish life.

In fact this gene appears to have two mutations. The second one might have been caused by the first though - don't know enough about the brain to be sure. The State Rule has changed, and so has the Relax State Rule. Not good stuff.

But that wasn't her only problem. The shivering turns out to be due to a couple of simple point mutations. Firstly:

 12 Different in File 1   5   0 Emb   B MutDupCut   128  Organ#=1 Creature, Drive Levels Hotness, chem=Coldness, thresh=0, nom=0, gain=122, features=Analogue
 12 Different in File 2   5   0 Emb   B MutDupCut   128  Organ#=1 Creature, Drive Levels Hotness, chem=Hotness, thresh=0, nom=0, gain=122, features=Analogue

This one borked her ability to process temperature chemicals. Instead of the Hotness drive being controlled by the Hotness chemical, as in File 2, it was controlled by Coldness. Essentially this made her unable to perceive warmth. If a place had a lot of Coldness, she'd feel warm, but otherwise she feels very cold because she cannot detect Hotness chemical. This would be a useful adaptation for a Norn living in a cold place, but for her it was a liability. She was shivering because the Norn Hatchery area is warm - unable to sense Hotness chemical and without any Coldness chemical to paradoxically warm her, she was left feeling constantly cold.

That's not her only temperature related mutation:

304 Different in File 1  32   0 Emb   B MutDupCut   128  Organ#=19 Creature, Sensorimotor, Invol 4='shiver', chem=Coldness, thresh=141, nom=1, gain=255, features=Analogue
304 Different in File 2  32   0 Emb   B MutDupCut   128  Organ#=19 Creature, Sensorimotor, Invol 4='shiver', chem=Coldness, thresh=141, nom=0, gain=255, features=Analogue

This is a very small mutation, of course, but it's interesting that she'd have multiple areas of the genome concerning the same function (temperature detection/reactions) mutate all at once in the same individual. (Both parents were genetically identical Gen1 Norns so none of these mutations were inherited.) Essentially, this gene states that when the Coldness chemical reaches a quantity of 141 or higher, the Norn will begin to shiver.

Now here's an interesting change.

158 Different in File 1 112   0 Emb   B MutDupCut   128  Organ#=15 Organ, Injury chem=Antigen4, thresh=16, nom=0, gain=50, features=Inverted Digital
158 Different in File 2 112   0 Emb   B MutDupCut   128  Organ#=15 Organ, Injury chem=Antigen4, thresh=16, nom=0, gain=50, features=Digital

This gene states that Organ 15 is injured by the presence of Antigen4 if its quantity is 16 or higher. Normally, this receptor is Digital rather than Analog, meaning that when the signal strength of the chemical the receptor is sensitive to is greater than the given threshold number, the receptor will fire, and its firing strength will be equal to the nominal plus or minus gain. In this case, in the normal/healthy genome, that means that when the chemical Antigen4 is present in too great a quantity, the receptor fires at a steady rate and intensity; the result is organ damage, as the receptor belongs to the organ's Injury locus.

However, this girl has changed to Inverted Digital. An Inverted analog receptor fires more weakly the stronger the signal gets, rather than firing more strongly the stronger it gets as with a normal analog receptor. Now, Digital receptors don't vary in their strength like that - they're either firing or not firing. So an Inverted Digital receptor will fire when the chemical is BELOW the threshold instead of firing when it's ABOVE the threshold. This poor Norn is similar to a C3DS Toxic Norn in that she actually requires Antigen4 in order to prevent her Organ 15 from dying!

As if all of that wasn't enough...

591 Different in File 1 137   0 Emb   B MutDupCut   128  Organ#=14 Organ, Injury chem=Antigen1, thresh=16, nom=0, gain=50, features=Digital
591 Different in File 2 137   0 Emb   B MutDupCut   128  Organ#=14 Organ, Injury chem=Antigen0, thresh=16, nom=0, gain=50, features=Digital

This gene is sort of a mixed bag. It causes her Organ 14 to be vulnerable to Antigen1 when it wasn't before, but in tradeoff, the organ loses its vulnerability to Antigen0. Kind of a wash, ultimately.

And finally, that's it. All other changes were in pigment and pigment bleed genes. So there you have it - I was right that her main problem was brain damage due to a catastrophic mutation, though I would not have guessed that an entire lobe had gone down in flames. I also had no idea she was slowly dying of lack of Antigen4, too. All in all this is a highly nonviable Norn, and I think I'd be merciful not to ever import her again. It's sad, but it's for the best.

Sunday, July 10, 2016

Rambling About My Dragon Norns

Possible notes toward the readme.txt for the next release of my dragon Norns...



First and foremost, the most important thing that people who are used to keeping more 'normal' strains of Norn need to know is this. These Norns are predators.

This means several things. Firstly, although they are excellent at caring for themselves in general, they have some special needs and problems that normal Norns lack. Specifically, they are very poor at navigating, much moreso than normal Norns, despite their lack of obsession with buttons and lifts. This is because most Norns rely a lot on food smells to navigate, but my various dragon Norns are carnivores, and unfortunately, critters and bugs do not emit a scent of their own. They don't even seem to emit the protein scent given off by fruit.

The result is that these Norns will often appear to be much stupider than they really are if you try to put them in situations that require them to be good at navigating by scent. For example, they will fail the conventional IQ test every single time - and not merely because they will die a brief four minutes after hatching if not fed. They can navigate just fine when it comes to tracking down fellow Norns, but if they manage to get themselves too far out of sight of food, they may become lost and starve to death because they cannot smell their way back to it, unlike other Norns.

Additionally, although they are able to derive some small amount of nourishment from the usual 'food', 'fruit', and 'seed' classes, they are designed to be primarily carnivorous. Critters, bugs, and even those pests and beasts which are edible will provide much more nourishment and will be sought out by most dragon Norns exclusively. 'Food' class items are an important secondary category and will be commonly browsed on, but fruit and seeds will typically be ignored most of the time by most individuals. (With that said, I have noted several dragon Norn testers that, over the course of their lives, developed a fondness for fruit or seeds, some of them even liking one of those things in preference to meat despite getting minimal nourishment from them - they'd just eat a whole bunch of seeds or fruits to make up for it. The long lifespan of a dragon Norn allows for more personality development than might otherwise be true in a species with such strong instincts.)

In short, if you put one of these Norns in a barren metaroom and expect it to find its way to another, better-stocked metaroom, it will die, through no fault of its own. It'd be like taking an ordinary Norn and turning off all the genes that allow it to smell things, then putting it through the IQ test. It's not merely a guaranteed failure, it's unfair because the parameters of the test require the Norn to be able to do something based on information it just doesn't and can't have. Also, you can pretty much never keep these guys in an unmodified world. You will need to add at least a few additional critter and bug agents in order to keep your dragon Norns properly fed, especially if you're going to keep more than a few of them.

However! With all of that said, you will find dragon Norns to be very hardy and tough if you provide them with an adequate environment. The Norn Terrarium comes close, as it has a number of critters normally, though the Flooded Norn Terrarium is better, with its fish and crabs and swimming pig things. All you need to do is add Shrimptons so they'll have a bug and the Hydro Norn Terrarium is a perfect habitat.

Although they are attracted to pests as a source of food, and are not particularly harmed by geddonnase, stingers are not an ideal meal. Using Stinger Honey is a good way to make stinger hives a more viable food source to balance out the pain they cause.

Dragon Norns are aggressive beasts. They are not actively murderous, usually, but they are heavy-handed with each other; just the evening of my typing this document I witnessed a newly hatched baby advance on her father and proceed to beat him soundly about the head and shoulders! (Literally! She swam up in the water to his head to attack him rather than hitting at his lower body.) They do have an instinct that inhibits them from hitting each other, but it doesn't work well, to be honest - they quickly learn that hitting other Norns is just as fun as hitting anything else, in terms of how the actual hitting feels. However - aside from personal feuds that develop due to social disputes and disagreements - Dragon Norns are not usually Norn aggressive to a dangerous degree. If they are struck themselves, their bodies produce Grendel Nitrate, which makes it easier for them to strike out in return even if they like the Norn that hit them.

I did this so they'd be more able to defend themselves against Grendels who seemed friendly at first only to lash out later, and Grendels encountered by Norns who were in a good mood at the time (since a Norn in a good mood is apt to view any freshly met creatures as new friends until forced to think otherwise - by which point it may be too late when dealing with the more vicious and powerful grendel breeds). But it's interacted with other Dragon traits to produce a peculiar side effect. In brief, Dragon Norns take being hit by other Norns in stride, and they also take it even less seriously the older they get. Babies, children, and adolescents are more emotional and may engage in dramatic battles and arguments, but once they hit Youth and especially the long, long Adulthood, the more fond they tend to become of the other Norns in their social group in spite of any moderate hitting that may go on. Norns who dislike each other tend to do so on the basis of other disagreements or extremely intense combat; a few slaps here and there isn't even enough to register on their scale of like/dislike. It's almost as if they learn to regard hitting as just another way to interact with each other - not always a fun way, but not necessarily a bad way, either.

However, when it comes to Grendels and Ettins, Dragon Norns are not to be considered safe. The tougher Grendel breeds are OK to keep with these Norns - they may fight each other, and might even become very vicious, but I have found that, one on one, they are usually not strong enough to kill each other. (Do not allow groups to gang up on individuals, though, or you'll end up with a dead Norn or Grendel.) These Norns have neuroemitters as well as stimuli and instincts which prime them to feel a host of negative emotions in the presence of both other species. Ettins make them feel angry, bored, and crowded, whereas Grendels make them feel angry and crowded but not bored. They usually do not act like Hardman Norns, though, because they lack the artificial and constant flood of Anger that Hardmans suffer from. They will usually not go out of their way to track down Ettins and Grendels, and can even coexist with them peacefully if there is only one non-Norn in the group, plus the Ettin/Grendel behaves itself and gives the Norns their space. How individual Dragon Norns react to an Ettin or Grendel depends on the Norn's personality, but they have the capacity to learn to be truly savage killers of non-Norns.

Beyond that, Dragon Norns are great at taking care of themselves. If you provide them with a sufficiently food-rich environment, they will feed themselves without needing to be told to, and will usually eat only what they need without gorging themselves. They will usually rest when they need to and sleep when they need to without being prompted. Even as totally unattended wolflings (I count them as wolflings rather than ferals because they require a slightly modified environment to even have a chance to survive due to their carnivory) they are capable of surviving up to 12 hours or even more - the oldest known Dragon Norn tester was over 14 hours old at time of death. In short, all you need to do is set up an appropriate environment with the correct temperature ranges and a sustainable population of edible critters and bugs. Then put the Gen1 eggs in there and leave them alone. They will hatch and see to themselves without needing your help.

Which is not to say that they aren't any fun to nurture. Their very long lifespan, the vast majority of which is spent in the Adult life stage, allows them to learn much more than a typical four-hour Norn ever can. By the time a typical Norn is starting to develop noticeable learned personality traits, it is very old and about to die, whereas a dragon at five or six hours of age still has at least another five or six hours to look forward to. This makes hand-raising and light-feral runs of these Norns very rewarding indeed. It's great fun to begin with a group of four or six genetically identical Gen1 Dragons, hatch them, and observe how similar they are as babies - then watch their individuality slowly develop over the next few hours. By the time they are six hours old, most Dragons have a noticeable and distinct set of learned personality or character traits. For example, some learn to be highly social, remaining close to their family, while others learn to enjoy wandering. Some spend most of their time on the ground while others are constantly flying. Some develop a great fondness for eating bugs or pests exclusively, while others learn to like and seek out fruits or seeds - sometimes even in preference to  meat, despite receiving very little objective nutrition or drive satisfaction from eating such things. It's a textbook example of how "nature" (genes) and "nurture" (life experiences/learning) come together to create the total personality.

Saturday, July 9, 2016

C2 Mystery Norn

How is it possible I haven't posted since April?? Where did the time go? Yikes!

Well, today I'm posting because I'm pretty mystified. I've been playing with C2 today since it's the version of the game I've spent the least time playing as a rule. I edited the basic 'norn.gen' file a bit and made some Gen1s. One of the Gen2s from that group is really odd and since GeneCompare only works for C3DS Norns I haven't yet been able to figure out what's wrong with this Norn.

Basically, she hatched lying down and never moved. She never looks away from herself. Other Norns have come up and interacted with her both positively (tickles, talking) and negatively (hitting) with no effect in terms of getting her to move. She also began shivering shortly after birth for no apparent reason. She makes no attempt to take any actions except to speak, and once after I tried injecting her with a metabolism transplant she sat up for a few seconds, then fell asleep for a while.

I'd hoped that was a good sign, but alas, she woke up in the lying down position and never moved again until I gave up and exported her out of mercy. I took a quick look through her genes but I haven't spotted anything yet since I'm having to work manually. Gah, wish I had a C2 version of GeneCompare!

It seems like she probably has some kind of debilitating brain mutation, but I can't be sure. If anyone reads this later and has any ideas I'd love to hear them!