Well the reason I don't like the "egg group" in Pokemon is because each pokemon is a different species and species don't mate with other species in most cases. Some do, but they almost always produce infertile offspring.

Basically the way my monsters are broken (with real world creatures) down are...

Kingdom (Animal) -> Family (Feline) -> Species (Lion)

Kingdom (Animal) -> Family (Feline) -> Species (Tiger)

Kingdom (Animal) -> Family (Canine) -> Species (Wolf)

Kingdom (Animal) -> Family (Canine) -> Species (Fox)

A Lion and Tiger can Mate, but a Wolf and Fox cannot.

Tigers can mate with other tigers, Lions can mate with other Lions, and Wolves can mate with other Wolves, Foxes can mate with other foxes.

And Lions nor Tigers can mate with foxes or wolves.

Pokemon are just crazy even though they do have the above type of categorization to some degree. The egg groups are just all over the place where some species cannot mate with itself. Some Species can mate with other kingdom. etc...

The 1 Kingdom that can mate between all or most of the families/species is Humanoid and that is pretty much following mythology rather than genetics... for fun ^.^

As far as fun goes, I think both have equal merits as to how fun they are. There is definitely something to be said to about merging 2 species together and getting some weird new species, but there is also fun to be had in having constraints placed on what you have to work with to get what you want too.

Considered the you fighting with your pets thing... for me it's a no-go for right now anyways mostly due to I couldn't do it if I wanted to. I've considered the whole multi-pet thing too. The problem there is AI and CPU... if the problem is only AI then I'll do it if I can... if it's CPU it's a no go... another part of the plan is to make this for tablets as well as PCs ^.^ so I want pretty much the minimum happening on the screen as possible. Currently my thinking is Pokemon style of grass = trigger into random battle, but if the CPU load isn't bad I see no reason not to have a set of pets fight freely around you with AI controlling their actions while you issue commands... But for now...small steps ^.^

---------- EDIT -------------

My first attempt at creating an stat gene algorithm thing didn't quite work...

I pretty much made it so that I could set the number of genes on each genome and then randomly generate which attribute and what value...with the value ranging from -3 to 3... and then compare the 2 chromosomes... which resulted in mostly negative stats... So yeah that didn't work

I tried to limit the negative stat genes so that only 1 in 3 times a negative number comes up does it get through... better, but not by much.

Then I removed the negative values so i could only get 0-3... and the results are meh due to some stats not getting any... which sucks and far too uneven.

So the next thing I'm going to do is make it so that each area is handled by 1 chromosomal pair... That should make it so i can get a more controlled distribution of points in the stats in a given area as well as in what area the stats go in. I think I'm also going to limit the number of negative points to maybe no more than 1/3 of the overall genes in a given area, maybe up to half. Maybe I'll take them out as they cause problems all together, but i dunno, depends on if after I finish this it doesn't look right still.

Well the reason I don't like the "egg group" in Pokemon is because each pokemon is a different species and species don't mate with other species in most cases. Some do, but they almost always produce infertile offspring.

Given that the whole concept of battling pet monsters is unrealistic, not to mention "elemental" animals which are pure fantasy, I don't really see why you'd be concerned about realism of interbreeding. If your world is a magical world, there's no reason that world wouldn't have fertile interbreedings between dogs and cats, or whatever pairing.

Well the reason I don't like the "egg group" in Pokemon is because each pokemon is a different species and species don't mate with other species in most cases. Some do, but they almost always produce infertile offspring.

Given that the whole concept of battling pet monsters is unrealistic, not to mention "elemental" animals which are pure fantasy, I don't really see why you'd be concerned about realism of interbreeding. If your world is a magical world, there's no reason that world wouldn't have fertile interbreedings between dogs and cats, or whatever pairing.

That's as true as "It's fiction, you can write whatever you want" in as much as it is true, a good story generally doesn't just do whatever it wants,

But I'm not saying that it won't for all times be prohibited, just not through breeding. If I ever wanted to add it the magical qualities of the world would allow it through a magical rare ability so that it is possible but not through normal means... Also the interesting part of this is that a play could then create a new monster species or have a species that is a 1 of a kind thing that can't be duplicated which makes it that much more special and valuable.

Ok, I am going to throw a random idea here.

What if we make the inheritable(s) combines to form other inheritables.

For example, if both parent have a BaseStats + 10 gene, there is a chance that the child will inherit a baseStats + 15 gene.

Similarly, say one parent have fire breadth ability and the other have a rock throw ability, perhaps the child might gain a fireball ability.

This forces the player not to just breed things that they can see, but also try new combination in hope to get new stuffs. =D

There's a serious balance problem: bias towards getting stronger monster than their parents causes grinding, bias towards weaker monsters makes breeding useless.

I would separate power level (starting low and fixed because it's a junior monster, and increased by training, special treatments etc.) and statistics/features (scaling up as monster power increases).
The "genes" could contain weights for stats, proportional to "baseline" stat values (excluding the increases that contribute to a trained monster's actual values); inheritance could give the offspring the average between the weights of each parent, plus some random noise, with initial stat values obtained by normalizing weights and multiplying for some constant. This way one could breed, for example, two monsters with high strength to get offspring with high strength, possibly even higher than either parent, but at the expense of other stats, making all offspring equally good (and as good as the parents).
Players would breed (and train appropriately) monsters with particularly unusual stats for their race in order to adapt to specific opponents or to support specific strategies (e.g. enough stamina to do a certain powerful attack one more time per fight), not to get more powerful monsters (which should involve training, experience and new monster acquisition).

Ok, I am going to throw a random idea here.

What if we make the inheritable(s) combines to form other inheritables.

For example, if both parent have a BaseStats + 10 gene, there is a chance that the child will inherit a baseStats + 15 gene.

Similarly, say one parent have fire breadth ability and the other have a rock throw ability, perhaps the child might gain a fireball ability.

This forces the player not to just breed things that they can see, but also try new combination in hope to get new stuffs. =D

As to the first suggestion... How I have it...

gene = type - value - dominance

Compare chromosome1 to chromosome2.

if the gene from chromosome1 has greater dominance than the chromosome2 gene, the chrom1 gen is expressed. .
If the opposite is true, then chrom2 is expressed.

But if both are equal both are expressed.

When breeding you have a 50/50 chance of getting either gene

All the monsters of the same species have the same genes when caught from the wild save for 10 which are randomly selected and modified.

The same is true when breeding... 10 genes are modified.

Currently I am liking the results of monsters stats with just +1 as base stats so every gene = +1 in a wild monster.

When breeding this will change as a value can be added to or taken away by 1 point.

This makes it interesting as breeding can cause stats that are not dominant to be expressed or for their value to go negative or for them to randomly switch what stat it is going towards ^.^ So even a monster at base may have no special attack or defense, if bred right they could in theory get some.

I like the idea in theory for combining moves, but I'm separating moves from breeding. The monster will get moves that one would assume an animal would get from being in the wild. The trainer can then teach them any move that they are able to learn... the moves then require training to raise power / lower stamina cost. So hypothetically instead of 2 moves coming together to form 1...the monster might have the natural ability to do the initial 2 moves, but then trainer would have to teach the monster how to do the combined one.

There's a serious balance problem: bias towards getting stronger monster than their parents causes grinding, bias towards weaker monsters makes breeding useless.

I would separate power level (starting low and fixed because it's a junior monster, and increased by training, special treatments etc.) and statistics/features (scaling up as monster power increases).
The "genes" could contain weights for stats, proportional to "baseline" stat values (excluding the increases that contribute to a trained monster's actual values); inheritance could give the offspring the average between the weights of each parent, plus some random noise, with initial stat values obtained by normalizing weights and multiplying for some constant. This way one could breed, for example, two monsters with high strength to get offspring with high strength, possibly even higher than either parent, but at the expense of other stats, making all offspring equally good (and as good as the parents).
Players would breed (and train appropriately) monsters with particularly unusual stats for their race in order to adapt to specific opponents or to support specific strategies (e.g. enough stamina to do a certain powerful attack one more time per fight), not to get more powerful monsters (which should involve training, experience and new monster acquisition).

You mean...

p1 : HP * w

p2: HP * w

c1: HP * (p1.w + p2.w / 2)

Considered that... I don't like it as much as I like the full gene thing. The results of what you are suggesting keeps all the monsters roughly the same which, if you're concerned with balancing issues. I would say overly concerned, would help keep that in check, but I like the idea that you could in theory take a monster that you like and through breeding make it something crazy powerful in the way you like.

You also said that it keeps all the monsters equal, regardless of whether they're bred or not. I don't like that. A monster that is bred should be stronger than it's previous generations, but also carry with it the risk of being weaker. Those who get into it and really try will be rewarded for it. Those who don't, won't. I don't see a problem. The breeders will have an edge. And the trainers will have an edge...and a trainer who is a breeder will have an even bigger one. The more time and effort you put in, the bigger the edge you will have. I see that as a good thing. Not a bad thing. I also like the idea that because different people like different things it could result in an originally weak monster unexpectedly being uber strong and further allowing players to make that decision to go against the game's defaults and make a weak monster strong and strong monster weak.

There are only 2 restrictions in the system I have now... You can't change Element (which restricts moves) or Kingdom/Family/Species. 1 of which if I can figure out a satisfactory way to handle I'll change, and the other I've pointing out I could make it so it isn't so much a restriction as it is a rarity, but I haven't decided that yet.

Here is the stat generator for the Kingdom of "animal"... this seems to create decent stats.

http://durakkenstudios.com/Navitae/GenerateStats.html

Every stat point is generated by...

Gene from Chromosome 1a's dominance vs Gene from Chromosome 1b's dominance

if the former is higher its gene expressed.

If the latter is higher its gene is expressed.

If they happen to be equal both genes are expressed.

All genes can have a value of 0 or 1, save for the basic stats which are always 1 for this.

Each group is a different chromosome. Different kingdoms have different numbers of genes in each chromosome and their distribution.

If I knew how to change model details more fully I'd consider the "skins" and "size" to be separate and the attributes like size would need its own chromosome which is why it's separated

The results of the stat distribution...is that about half the number of genes are 1 and half are 0. And the dual expression comes up maybe 1 in 10 to 1 in 20 reloads.

Now I need to come up with the distribution for the other kingdoms which are...

Spirit

Insect

Dragon

Plant

Humanoid

And then I need to generate the xml data for the chromosomes/genes for the base monsters to use for random spawns.

You can do a lot with chromosomes, but the question is how it is communicated to the player in an understandable way,

especially if a player can only see the stats of a monster and not it's individual chromosomes.

I'd keep it simple and add one (small) negative stat and one (bigger) positive stat to each chromosome.

Small side-effect will be that incest will give a chance of getting the same chromosome double, and giving a monster double the penalty,

which can be quite the disadvantage and actually is why incest is bad irl.

(it can also be advantageous as the positive stat also goes double, so you get specialized monsters)

You can do a lot with chromosomes, but the question is how it is communicated to the player in an understandable way,

especially if a player can only see the stats of a monster and not it's individual chromosomes.

I'd keep it simple and add one (small) negative stat and one (bigger) positive stat to each chromosome.

Small side-effect will be that incest will give a chance of getting the same chromosome double, and giving a monster double the penalty,

which can be quite the disadvantage and actually is why incest is bad irl.

(it can also be advantageous as the positive stat also goes double, so you get specialized monsters)

Considering I just explained in briefly and it is very understandable and covers just about everything I don't think their's a problem of communicating nor of understanding considering how simple it all really is.

I don't want players to see the genes. That would destroy quite a bit of the adventure.

The reason incest is bad is because...

1) Lower fertility rate of the first gen

2) Higher chance of Expression of bad recessive genes

3) Genetic homogeneity is means that if something figures a way to kill you, it effects everyone.

1) Will be in the game

3) Is a non-issue for a game

2) The way this works is that you have something like this...

Parent1 = T|t && Parent2 = T|T

Child = TT || TT || tT || tT

The expressed will be T, but let's say the child inherits the t from Parent1 so it is tT

Parent1 = T|t && Child = t|T

IChild = Tt || TT || tt || tT

So when the parent and child mate the chance of t being expressed is 25% where as before it was 0%. More over the chance of dual expressing goes from 50% to 25%.

Supposing that this t is actually a good thing then it is a good thing, but if it is bad then it is a bad thing...

Assuming it's a bad thing you'd stop breeding there, but if it is a good thing what you'd do is attempt to create a breed with tt is 100%... in other words the family line would be like this

Parent1 = T|t && Child = t|T

IChild1 = Tt || TT || tt || tT

IChild2 = Tt || TT || tt || tT

And assuming both get tt then you'd want

IChild1 = t|t && IChild2 = t|t

2ndgenIChild = t|t

Of course a breeder who is just looking for the t|t and doesn't care about the rest would try to get t|t from 2 genetic lines and then bring the lines together so you only get the thing you want expressed to be expressed.

There is a possible problem with what I've got which is that in the 2nd or 3rd generation of pets will have a great number of double expressed genes due to only 20 or 30 mutations from the base of that monster will have occurred... and there are 1000 gene pairs and 6,000 possible mutation points, so the likelihood that a dominance gene is effected is 1 in 500.

So what will happen will be something like this

Caught monster 1 = HP:1:500 | STM:1:632

Caught monster 2 = HP:1:500 | STM:1:632

Bred monster will be one of the following =

HP:1:500 | HP:1:500

HP:1:500 | STM:1:632

STM:1:632 | STM:1:632

STM:1:632 | HP:1:500

Statistically that is

CM1 = HP 0/STM 1

CM2 = HP 0/STM 1

BM = HP 2/STM 0 || HP 0/STM 1 || HP 0/STM 2 || HP 0/STM 1

1/2 the bred monster genes will double expressed resulting in a 50% increase in stats 50% of Gen 1 bred monsters.

That's an interesting result...

Anyways... i'm trying to figure out how to auto generate the genes while still having a control of the stats. I suppose I make it so that I can set a stat and then have it randomly assigned to slots in the chromosome. but that wouldn't work with the comparative method. Oh well I'll figure something out.

Hrmmm seems I've run into the problem too many columns for spreadsheet software ^.^

i guess I'll just have to figure a new thing out.