Animal Husbandry Feedback
Before I say anything else, I want to say I love animal husbandry very much as it is. It is such an interesting and complex system that has kept me engaged for 18 months so far. All feedback is written from a place of appreciation and respect and the target audience for this is primarily Srand, but I hope there is some useful information for others.
To date, I have hatched 1,280 bees/wasps, primarily Freeze Wasps. Doing some back of the envelope math, I expect that I have just about a year left in the progression to get to best in show attributes. My best wasp has the following attributes: {'Fe': 87, 'T': 85, 'Ru': 83, 'En': 74, 'Fr': 93, 'I': 78, 'Fe': 77}. Throughout these 18 months, I have been diligent in my hatching (outside of hatches between 3am and 9am) and while I could be further along in theory, it would come at the cost of sleep for a marginal gain.
I am writing this feedback with the knowledge that bee genetics are currently slightly broken and will undergo a redistribution of points and possibly what genes affect stats. By broken I mean at least one and likely 2 attributes cannot currently reach the 100 point max (ignoring the possibility of mutation). I hope that this redistribution will minimally affect the progress of bee breeders currently in flight b/c it has been a long road with much more ahead.
Also, I would like to shout out to Azizah - her and I have been discussion what we know about bee breeding since the start and she has done a lot of work in understanding the current genome and its limitations!
Lastly, before diving into the specifics of my feedback, I would like to acknowledge that these are end game items for an end game that doesn't exist yet - so I know it puts them in a weird spot of cost and progression as future content will change the value of a council and slowing us down is probably in your interest at least partially.
Feedback:
Overall I really like animal husbandry. I love that there are gene interactions such that a gene that took away an antennae didn't take it away when another gene that affected this became recessive. The system is both simple yet deeply complex, requires dedication and planning, and has a true progression. Mutation as a possibility interests me though if its happened I haven't been tracking it. The control over both phenotype and attributes is very cool.
My more constructive feedback can be broken down into the following:
- length of time required
- cost of progression
- availability of stable slots
- Lack of UI support/requirement of external tooling
I will go into each of these in detail and then provide some things I would and wouldn't like to see added as someone who is excited about animal husbandry. I will conclude with some information about how I process and hatch bees with hopes that it will be helpful when considering UI elements.
Length of Time Required.
If I could do it all over again, I think I could shave some time off the process, but not by a significant amount. In its current state this process seems to last between 2-2.5 years. That is just to finish the attributes of your bees, it does not account for any cosmetic choices that you might want. And that's 2-2.5 years of logging in every 2-3 days to tend to the brood. This is a huge ask of players and since there is a deep cosmetic component to this a single player can expect, in its current state, to spend at least 3 years finishing their bees/wasps in both attribute and aesthetic. Moreover, for players that enter the game late, I assume it would be difficult to find motivation to go through this process knowing you had 2 years before your bees were truly marketable. They might be able to sell in process bees at a steep discount the way actively breeding players are today, but for many I think a 2-3 year commitment will be a hard sell.
An additional time sink is going out and finding the genes not already in your gene pool. Analyze genes has a decently long CD and (unless you're a fairy) requires you to have mez on your bar in addition to animal handling (which notable doesn't have a mez) or risking your pet killing the bee/wasp before you can analyze it.
The typical fat requirement also adds time here - I can get 10 typical fat in around an hour of farming so that's an hour of farming every 2 days. Not terrible, but in the end typical fat alone will account for ~450 hours of farming and typical fat is both rare and expensive on the player markets.
Cost of Progression
Back of the envelope math I hope to finish shortly after my 2,250th hatch for attributes. I have 20 stable slots (have purchased the 1 mil slot not the 5 mil) and use typical fat for all breeding attempts that are not the full moon. Historically I have seen typical fat range from 5-10k each in the player markets. I will use 5k for simplicity. Each breeding attempt has cost between 500 and 2k councils not counting the registration fees. For simplicity I will use 750 councils as my average cost of mating attempts, but this is more of a guess than a true number.
If I'm correct that near my 2,250th hatch will have best possible attributes (which is also a guess) the total cost is as follows:
Stable slots: ~1,550,000 councils
Typical Fat: 2,000 hatches (assuming typical fat not needed for ~10%) at 2* 5,000 councils each = 20,000,000
Breeding costs: 2,250 * 750 breeding and registration cost = ~1,650,000
Total cost not counting egg incubator mats and xogrite costs = ~23 million councils
Now the benefit of this cost is, once best in show bees are obtained they can be produced for a few hundred councils. However, these are only useful to players that main AH so we're stuck with a near infinite supply (up to 5 every 2 days), a large sunk cost, and a medium-low demand. AH mains might buy multiple species of bee/wasp, but in order to break even a player would have to make 23 million in sales. For horses, I am not terribly concerned about someone making back this money since horses are universally desirable, but for AH pets the time and council costs present a pretty large barrier to entry.
Again, this is an end game item for an end game that doesn't exist so its possible in the future this 23 million will feel like nothing, but currently it a significant amount.
Availability of Stable Slots
Currently we are locked in at 5 simultaneous breeding attempts (per account). That requires 5 males, 5 females, and 5 slots dedicated to the offspring. In theory you should have 5 good or best bees (breeding bees) mating with 5 bees that have 1 or more gene you're trying to fold into your pool (folding bees). If all of your breeding bees were male and all your folding bees were female the odds of hatching males is higher, so your next hatch you could see 5 folding males, 5 breeding males and 5 now useless folding females. To this end you'd want an additional set of breeding bees to offset the sex ratio of your actively breeding bees. That is, you'd want a back up of up to 5 breeding females in this example. There go 20 stable slots easily and with max husbandry my 21st stable slot would cost me 5 million councils.
But those 20 bees aren't enough, what happens when you "finish" folding a gene in? You need to immediately start the next gene in line - so ideally you also have 1-2 wild caught bees on the bench ready to be folded in.
And currently your mount takes up a slot and if you actually use animal handling each pet you want to keep takes up a slot. This is another place where end game councils would make having 22+ stable slots more realistic, but if animal husbandry 81-125 comes with more simultaneous breeding attempts you could easily eat up any excess stable slots (at 3-4 slots per attempt). I have given up my mount and all AH pets for this progression.
Lack of UI elements/requirement of 3rd party tooling
This one is pretty self explanatory, but its not easy the way its currently set up to process bees. You can save the gene to file if you have VIP which helps, but still you need to load them into some other tooling to process. Not having VIP means translating by hand which is error prone and puts a soft VIP requirement behind husbandry for its save to file functionality.
Things I'd like to see added
- Reduce required number of genes to perfect
Increasing the average effect of each recessive gene could go a long way in shortening the progression and would make each gene more meaningful. Currently, it looks like a "finished" bee will require ~120 recessive attribute affecting genes (assuming Int and Fert will need some extra from what's available today) - lowering this to ~100-105 could go a long way. This would also increase the variability we'd see in bees. Assuming Toughness is the least impacted attribute with 0 always dominant genes I will use that as a baseline. Currently there are 206 possible combinations of Toughness that meet or exceed 100, but of those only 98 for the minimum required number of genes. This seems like a lot, but for how complex and varied gene interactions are, I would be interested in greater variability. Though adding more options at the same amount so you need a smaller % of genes would also increase variability. I do not pretend to have a full understanding of the impact on aesthetic of the genes so perhaps 98 is more than enough, but reducing the standard deviation of how much each gene impacts an attribute and increasing the average would intuitively increase the variability and reduce the total genes needed. At the very least, I hope that bees are the most complex genome (possibly tied with horses) such that later species have a naturally easier progression.
- Control over outcome sex
This is the one I want the most. This solves a lot of issues including number of needed stable slots, complexity of breeding, and length of progression. It removes one 50% coin flip from the equation which is in itself huge in the progression - I doubt it would reduce the time spent by 50% but I could see 20-33% reduction. It also removes the need for fall back breeding bees, so instead of an ideal of 21-22 stable slots for breeding that would be reduced to 16-17. If this is implemented I could see it being a craftable "pheromone" given to females before breeding. If this precludes the ability to use super egg incubators, it would not shorten the progression so ideally this would be used before mating and super egg incubators would still be usable after successful mating attempts.
- More Stable Slots/Breeding attempts
Somewhat self explanatory as it naturally speeds up progression. However, this increases the rate at which you use fat which is expensive and increases the complexity/time spend processing and re-breeding bees. This is a somewhat obvious option, but not the most favorable to me.
- Ability to sell wild caught bees for breeding
This would reduce the time spent catching wild caught bees, but it would also give players not directly involved in breeding bees a use for arthropod genetics while creating a market for specific rare genes. This suggestion is specific to wild caught bees, not hatched.
- Recipe for lumpy fat wax
This would cut costs by up to 20million for the breeding process.
- Analyze Genes have a shorter CD and impervious to interruption
Aside from having arthropod genetics lvl 100 (14 hours and about 1k xogrite) and animal handling 70, you also need to have a mez on your bar that can, at the very least, handle freeze wasps which are surrounded by elites if not being able to handle FR. Simply making it so this spell can't be interrupted would allow players to use whatever combat skill they prefer to analyze.
Honorable Mentions
- An item that has a % chance to increase outcome attributes while not controlling genes - this could let players in progress hatch better bees than they have, they then have to decide if they sell these bees or if they keep them for breeding. Fun idea, doesn't change much.
- Control over outcome species - similar to outcome sex but less important
- Item that encourages mutation across all genes - no specific gene control, but just a fun little randomization and see what happens.
- Increase success chance per point of fertility - I would prefer a lumpy wax recipe, but this would also be welcome.
- Completely separate breeding slots from pet slots - not sure about this one, but it could allow a rebalance of stable slot pricing.
- Add sex and stats to genome save file for easier parsing
- Have save to file for genes not be VIP exclusive
Things I would NOT like to see added
- Gene specific control via items
- Increase chance of rolling recessives via items or otherwise
- Ability to sell hatched bees for future breeding
How I process my bees
When hatching a new bee I save the gene to file and then load it into a notebook using parsing methods I wrote in python. I refer to genes in the following way: [Chromosome (1-10)][Grouping (A-J)][Index (1-4)] so the very first gene seen in the genome would be 1A1 and the very last would be 10D2. I filter down to only genes that affect an attribute and calculate the following numbers for each bee: how many attribute affecting genes have at least 1 recessive allele (empty circle or partially filled circle) and how many attribute affecting recessive alleles are there in total (sum mixed genes + sum recessive genes * 2). The total allele count can never be more than 2x the total gene count and a bee with 2x alleles than genes is considered "pure" as it cannot improve without introducing new genes. I then print out a sorted list of all my bees split by sex so I can easily rank my bees from closest to perfect to farthest (sorted first by gene count and then by allele count descending).
However, often times my "folding" bees hatch a bee that is better than the folding bee, but doesn't have the gene they're folding. To that end, I flag any bee who is the sole carrier of recessive alleles for an attribute effecting gene as "unflushable". In the case that only 2 bees have alleles for a gene I care about there is a use case for flagging them both, but most times I am actively comparing them as one is parent and one is child so I skip that.
When I am considering breeding 2 bees together I compare:
- genes that one has and the other doesn't
- genes they both have where one has more recessive alleles
- genes where they are both mixed
as follows:
Under each section I list the gene, number of recessive alleles on the given bee/wasp, and the count of recessives fo this allele across my brood. In this case, m1273 is a folding bee with gene 9B2 (count across brood = 1) and is un-flushable. f1266 has one more gene represented than m1273, but otherwise 9B2 is mostly folded. Any cases where both bees are recessive on the gene are excluded.
I can then run compare_2 between m1273 and the hatched child to compare them or look to see if 9B2 was passed along and then compare their count genes and count alleles to decide who to keep and who to flush. When they are sufficiently folded, I will mate the folding parent with the offspring for a 25% chance to move 9B2 to recessive or simply hatch enough near pure bees with 9B2 as mixed and start treating those as breeding bees.
I think the UI could use this type of comparison and summary statistics, though I by no means think my solution is the best implementation. Additionally the ability to highlight or filter to attribute affecting genes (possibly of a specific attribute) would help visually bypass the natural noise the genome presents. Presenting some odds ratios would also be helpful I think.
To anyone still reading, thank you and apologies for the novel. To Srand, I hope at least some of this perspective was useful and thank you again for such an interesting and exciting system, I've never seen anything like this. Happy to answer any questions!
