Mutations stopped on my Rexes

[EgorAkaErgo]

Hello,

I am breeding my own line of rexes. Just for fun. I raised stats to 55HP and 56DM. My breeding fleet has 200 lo level zero mutation females. My current male has 204 mutations.

I noticed, that I stop getting any further mutations.

I usually have some mutation every breeding. Now for at least 5 last breedings (200 females every time) I have no any mutation.

 

I dont understand why? 

[GrumpyBear]

Because, you have half the chance of getting a mutation back when your count was under 20 on the male.   Now that the male is over 20, he has no chance of supplying a mutation.  You have to get it from the females.

 

If you want to have a consistent chance of mutations, you need to increase the harem size.  or.... you break the mutation counter by increasing it to a value that is beyond the capacity of the variable the game has setup for the counter.  THis takes about 42 generations of incest between brother and sister rexes from a count of 1, 1+1=2...2+2=4......n+n ~1.1billion ~ -1billion , counter broken. 

 -1b < 20 hence , mutations begin again.

[EgorAkaErgo]

I have 163 females. All 0 mutations.

Rex has 2 important stats: HP and MD. What is probability to get 1 mutation on either HP or MD? 

I did 6 rounds of breeding with zero mutation on HP and MD, since this EVO event.

I am afraid that something is broken in the code with this color event patch

[EgorAkaErgo]

As far as I know,

1 if one parent have more then 20 mutations, and another parent has less then 20 mutations - further mutation is possible.

 

2 if both parents have less then 20 mutations, then probability of mutation is 5%.

 

3 if one parent has less then 20 and another has more then 20, then probability of mutation is 2.5%

 

4 one of 5 stats can get mutated: HP, STA, FOOD, OXI, WEIGHT, DMG

it is also possible that color can get mutated.

 

QUESTION 1: if color shares probability with stats mutation?

 

5 even if color shares probability with stats, and I am looking for mutation on either HP or MD, then probability of useful mutation is 2.5% * 2/6 = 0.77%

with my breeding fleet headcount 163 females, probability of useful mutation is 125%

 

considered that I had 6 unsuccessful rounds of breeding, probability to have useful mutation is 753%

 

which means that I had to have 7 or 8 mutations already. Which is perfectly matches what I had before current EVO event starts.

please advise me, where I a wrong?

[GrumpyBear]

22 hours ago, EgorAkaErgo said:

As far as I know,

1 if one parent have more then 20 mutations, and another parent has less then 20 mutations - further mutation is possible.

 

2 if both parents have less then 20 mutations, then probability of mutation is 5%.

 

3 if one parent has less then 20 and another has more then 20, then probability of mutation is 2.5%

 

4 one of 5 stats can get mutated: HP, STA, FOOD, OXI, WEIGHT, DMG

it is also possible that color can get mutated.

 

QUESTION 1: if color shares probability with stats mutation?

 

5 even if color shares probability with stats, and I am looking for mutation on either HP or MD, then probability of useful mutation is 2.5% * 2/6 = 0.77%

with my breeding fleet headcount 163 females, probability of useful mutation is 125%

 

considered that I had 6 unsuccessful rounds of breeding, probability to have useful mutation is 753%

 

which means that I had to have 7 or 8 mutations already. Which is perfectly matches what I had before current EVO event starts.

please advise me, where I a wrong?

* the math below has been fixed *

It might help to re-read https://ark.wiki.gg/wiki/Mutations

Lets just look at 1 stat.  Melee.  This is 1/7 of all stats on a rex.  The stat you want has a 55% chance of being inherited. The parent that can supply a mutation is the female, not the male.   So now, you have a 50% chance of the roll working from the prefered parent.  This parent has a 2.5% chance x3.    This 7.5% is broken up into 3 parts, "The probability for at least one mutation in an offspring whose parents both have less than 20 matrilineal and patrilineal mutations is 7.31%, for at least two mutations it is 0.184%, and for three mutations the probability is 0.00156%" [from ark.wiki.gg]

 

Its 2am for me here, hopefully someone can correct whatever I miss here.  But the way this works to me is (1/7) x (55%) x (50%) x (7.31%) = probability of success. I'm getting a grand total of 0.2872% chance of success on each egg.   You would need well over 300 eggs per breeding to even approach a probability of success.  If you want to talk guaranteed success, you probably need to double or triple that.

* that 50% could also be incorrect.  Part of me wants to think that should be 45% .  Since you are way past 20 mutations, double and triple mutations could be included.  So i'd probably bump the 7.31% back to 7.5%, keep the 50% because I"m probably wrong on the 45%.

That's still only 0.295% chance of success per egg.  Each egg's roll is independent , therefore you don't get to just take 100 / 0.00295 and calculate the # of eggs you need.  You would need to increase your sample size to start to improve your odds at being successful.  This means going way beyond just 300 eggs. 

[Pipinghot]

14 hours ago, EgorAkaErgo said:

5 even if color shares probability with stats, and I am looking for mutation on either HP or MD, then probability of useful mutation is 2.5% * 2/6 = 0.77%

with my breeding fleet headcount 163 females, probability of useful mutation is 125%

 

considered that I had 6 unsuccessful rounds of breeding, probability to have useful mutation is 753%

 

which means that I had to have 7 or 8 mutations already. Which is perfectly matches what I had before current EVO event starts.

please advise me, where I a wrong?

You're thinking about probability the wrong way. No matter how much it's possible to do something, that does not guarantee it will happen the number of times you think it should.

Let's look at an easier example. If you roll a 6-sided die the chances of rolling a 1 are 1/6 (16.66%), but if you roll it six times that does not mean that your chances of rolling a 1 are   6 * 1/6 = 1  . Instead the chances of rolling a one are    1 - (5/6 * 5/6 * 5/6 * 5/6 * 5/6 * 5/6)  =  1 - 0.33  =  0.67  . If you roll the die 6 times there is a 33% chance that you won't roll any 1's. Heck even if you roll the die 12 times there's still an 11% chance you won't roll any 1's, or if you roll the die 18 times there's a 3.8% chance you won't get any 1's.

This is why Grumpy is showing you that you have a smaller chance of success than you think you should. You can't simply multiply the % chance by the number of attempts, probability doesn't work that way.

I suggest to you that if you usually get a mutation from each breeding then you've been pretty lucky so far, all that's happening recently is that your luck is balancing out. Mind you, it's always possible that something has gone wrong but if that had happened I would expect to see a bunch of posts all complaining about this same thing. Since we haven't seen a bunch of posts this pretty much means that mutations are working just like they normally do and your luck is just balancing out.

 

[EgorAkaErgo]

9 hours ago, GrumpyBear said:

"The probability for at least one mutation in an offspring whose parents both have less than 20 matrilineal and patrilineal mutations is 7.31%, for at least two mutations it is 0.184%, and for three mutations the probability is 0.00156%" [from ark.wiki.gg]

this is probability that two mutation will happen SIMULTANOUSLY. Not "at least one of two" which I was talking about. 

 

57 minutes ago, Pipinghot said:

If you roll a 6-sided die the chances of rolling a 1 are 1/6 (16.66%), but if you roll it six times that does not mean that your chances of rolling a 1 are   6 * 1/6 = 1  . Instead the chances of rolling a one are    1 - (5/6 * 5/6 * 5/6 * 5/6 * 5/6 * 5/6)  =  1 - 0.33  =  0.67 

you are messing up things. What is probability? Probability is the chance that something happen. Probability of 1/6 means that in average it happen once out of each 6 trys. This is by definition of probability.

However, you are calculating different thing: you are calculating the situation, when you already draw 5 times unsuccessful and now going to draw one more time, and you are calculating probability of success in this time. This is different.

[Joebl0w13]

9 minutes ago, EgorAkaErgo said:

Probability of 1/6 means that in average it happen once out of each 6 trys.

This is absolutely not true. The chance of rolling a 1 on a 6 sided die is 16.66%. No matter how many times you roll it. It's still just 16%. You might roll that die 100 times and never once land on a 1 (although the chances of that happening are quite low).

The basic formula for probability is:

P(A) = (# of ways A can happen) / (Total number of outcomes)

I suggest you read this because you don't understand how probability works.

https://www.cuemath.com/data/probability/

[GrumpyBear]

6 hours ago, EgorAkaErgo said:

this is probability that two mutation will happen SIMULTANOUSLY. Not "at least one of two" which I was talking about. 

 

you are messing up things. What is probability? Probability is the chance that something happen. Probability of 1/6 means that in average it happen once out of each 6 trys. This is by definition of probability.

However, you are calculating different thing: you are calculating the situation, when you already draw 5 times unsuccessful and now going to draw one more time, and you are calculating probability of success in this time. This is different.

Each roll is independent. Every roll would be ~16.66% , no matter how many times you roll it.

*I was wrong on the final %, I was shifting the decimal 2 places too far.  My answer looked rediculous, that's my mistake.

I think my logic on the breakdown was pretty solid.  I wasn't robust on the including 2 mutation w/ a desired outcome of 1 melee mutation, and the 3mutation scenarios, but I think it should have been clear that doing so would only clarify the difference of 7.31% and 7.5% - not a huge improvement.  We could just assume this 7.5% is the actual, since its a bigger # than our actual #.   

 

From there its pretty simple  (1 out of 7 stats is the one I want) x (55% chance that stat came from the correct parent) x (50% chance the parent who is successful can actually supply a mutation) (7.5% chance of a mutation as defined by WC *ark.wiki.gg , and our assumed upper limit).

so that comes out to (1/7) x (0.55) x (0.5) x (0.075) = all the gas you have at making "the ONE desired" mutation happen.   Each egg would have only this chance.  Stacking 330 of them would not give you 100% chance of the mutation you want happening.   It would improve the chances, but not guarantee anything.  

 

Speaking as someone who has hatched over 1000 eggs for that 21st mutation on blood wyvern melee,  I have some idea of the efforts needed to get the results.  But my math here isn't meant to pass a math class, its just to get a rough idea of what we are working with.

[EgorAkaErgo]

I can confirm broken mutation: just did 217 carchas breeding - zero mutation.

Map = Lost Island. Definitely broken mutation since this EVO event.

Curious, if someone still getting mutations other then color and what map you are on if you still do?

[GrumpyBear]

1 hour ago, EgorAkaErgo said:

I can confirm broken mutation: just did 217 carchas breeding - zero mutation.

Map = Lost Island. Definitely broken mutation since this EVO event.

Curious, if someone still getting mutations other then color and what map you are on if you still do?

How are you spotting these mutations?  Breeding mechanics shouldn't be effected by any map they are on.  I've never heard of that happening in 5 years.   Most the problems we see, recur in some fashion or another.  I've never heard of them breaking mutation mechanics* , * - I can't say breeding because there are some issues with some animals sometimes (ex. aberrant dimorphodons lay regular fertilized dimorphodon eggs still to this day).

I can’t say what you are inferring from your experience is impossible ,  but after 5 years here,  a post like this usually resolves once the OP learns more or notices what they were missing, be it a mod or replacing bad advice with more solid info, or realizing the wrong male was in the mix mucking things up...

 

Let's focus on your process, I don’t want to assume ,  because I am not here to dismiss your experience.  It might help to know how you are spotting mutations;

• Are all breeders matching?  Are they all level 1 breeders or just random ones left over from breeding to mix stats?  
• how do you spot the mutations,  hatch and look for color? Incubator?  
• some Pictures of a sample ancestry , stat/inventory could possibly help spot something missed.

 

Extra maybe relevant detail - the new animal had issues with eggs recently that has already been addressed by a patch.  Could be there’s still something broken with them…Also, I bred my maewings a day before the thanksgiving event began and got a mutation.  So I did get one less than a week ago , when I last bred something. I also bred ferox last week and caught an unwanted mutation when I was trying to identify stat values.

 

***Just some extra established facts on mutations that might help give context;   

• Every mutation comes with a color change. 
• Not every animal has all 6 color zones, sometimes that color change goes to a zone that doesn't exist. 
• Some color changes go from dog brown, to ugly brown - not every color change is noticeable even if it happens in a zone you can see.   
• Run speed can be mutated, you will see absolutely no change to the animals inventory with the listed stats and values.   
• the incubator can lie.    I have had eggs hatch with mutations that the incubator did not register.   
• IF one animal has 1710 stamina and the other has 1785, the baby could inherit the 1710, mutate it to 1785 - and you would have no idea what stat mutated unless you had a strict control over the stats going into a mating.

 

[I've cleaned up this post because my brain kept editing it in my head and it was too messy, I hope its coherent now]


The 3 biggest tools I had for breeding was

• strict control over stats on breeding   I liked zero points in all stats except food on breeders.  And my stud would be matching with the exception of the target stat.  
• I would use about 250 breeders until 20 mutations. 
• To go past that, rolling the counter above 1 billion was the only way to keep the breeding setup the same. This process has to be started while making the stat mix on the breeders to be ready to take over after the 20 mutations limit is hit.  This timeframe for all of this depends on the growth rate of the target animal, and your level of commitment.

 

[Pipinghot]

4 hours ago, EgorAkaErgo said:

you are messing up things.

No, I'm explaining them. If you didn't understand that explanation feel free to go ask your math teacher. I'm sure that someone who teaches probability will have better ways of explaining this if you don't like my explanation.

4 hours ago, EgorAkaErgo said:

What is probability? Probability is the chance that something happen. Probability of 1/6 means that in average it happen once out of each 6 trys.

That's right, on average, but knowing the average never guarantees that it will happen when you think it should, nor does it even guarantee that it will happen as often as you think it should. Averages also include unusually long stretches of good and bad luck, in addition to what feels like normal luck. No matter what the average is there will still be good and bad luck, and no amount of arguing will change that fact.

4 hours ago, EgorAkaErgo said:

However, you are calculating different thing: you are calculating the situation, when you already draw 5 times unsuccessful and now going to draw one more time, and you are calculating probability of success in this time. This is different.

No, I'm not "calculating the situation" because that's not a thing,  I'm showing the calculation for the cumulative probability over a series of rolls. This is the same method of calculation that should be used for the cumulative chance of having a mutation when you hatch a bunch of eggs.

When you're calculating cumulative chances you don't use addition, you don't add 1/6+1/6+1/6+1/6+1/6+1/6, instead you multiply 5/6*5/6*5/6*5/6*5/6*5/6 and subtract from 1. If you don't understand the concept of cumulative probability that's ok, most people don't, I'm not trying to insult you, I'm trying to help you understand why the model you have in your head is incorrect. Earlier you said, "considered that I had 6 unsuccessful rounds of breeding, probability to have useful mutation is 753%" and that is super wrong, it means that you don't know to to do probability calculations. Again, I'm not trying to insult you, I'm trying to help you understand that you're thinking about this in the wrong way.

4 hours ago, EgorAkaErgo said:

when you already draw 5 times unsuccessful and now going to draw one more time, and you are calculating probability of success in this time. This is different.

When you calculate each individual roll, one at a time, it's always a 1/6 (16.66% probability). However, when you want to calculate the cumulative chance of success or failure from a large number of attempts grouped together then you have to use equations designed to calculate cumulative odds. Individual odds and cumulative odds require different logic, you're not understanding the logic of cumulative probability.

[GrumpyBear]

Its been a little while since I did the math here, but I think its something like 1 - (1 - P)^n  for the probability that it will happen after n number of independent events.

So w/ some ball park #s @ 1egg , 1 - (0.997)^1 = 0.3% chance.  This falls in line with my initial est. , going further; @100 eggs , 1 - (0.997)^100 ≈ 1 - 0.740 or roughly a 26% chance of the desired outcome occuring.   

 

I think this formula works.  It won't ever end up with an answer over 1, so it seems to fit the situation.  I also had to correct the final % i came up w/ was wrong , i moved the decimal 2 places when I shouldn't have.  The # I should have come up with was 0.295%, not 0.00295%.

[EgorAkaErgo]

I guess I figured what is happening: developers decided to increase probability of color mutation... and this lowered probability of all other mutations. Stupid, IMHO

P.S. Other users on Reddit confirmed that they also experience similar issue - mutations stopped or become very rare since this EVO event.

[GrumpyBear]

25 minutes ago, EgorAkaErgo said:

I guess I figured what is happening: developers decided to increase probability of color mutation... and this lowered probability of all other mutations. Stupid, IMHO

P.S. Other users on Reddit confirmed that they also experience similar issue - mutations stopped or become very rare since this EVO event.

Links speak more than your 8 post count.  I did a search and can't find a thing posted within the last week on the topic.

Every mutation has a color change, every color change has a mutation.  THey are linked together like a pair of pants, its one thing.   

 

I think I can solidly state, If you post a picture of the ancestry and stats of the 2 parents that created a baby with a color change and no mutations.  Include the same pictures of the resulting baby, I can show you what mutated with extreme confidence.  Anyone here on the forums would be able to see what you are seeing and point to what you aren't seeing. 

Its a lot of information for anyone, it helps to have a BS in math w/ most games - but this game takes the cake.

 

I single handedly burned over 2000 dino id #s a week on official for 2 years solid all by myself.  I know a thing or two about figuring out whats what on breeding.

 

 

[Pipinghot]

3 hours ago, EgorAkaErgo said:

I guess I figured what is happening: developers decided to increase probability of color mutation... and this lowered probability of all other mutations. Stupid, IMHO

No, that absolutely did not happen. No matter what anyone says anywhere, including reddit, it didn't happen.

Every mutation includes both a stat change and a color change, that is fundamental to the core code of ARK and has been since the very beginning. There is no such thing as separate probabilities for a color mutation vs. a stat mutation, every mutation changes both a color and a stat. First the game calculates whether a mutation happens and then it calculates which color and which stat get mutated.

3 hours ago, EgorAkaErgo said:

P.S. Other users on Reddit confirmed that they also experience similar issue - mutations stopped or become very rare since this EVO event.

I just took a look at reddit and I don't see any such confirmation, if you've found anything worth sharing you need to post the links.

What I do see are old posts from previous patches where people have said the same things you've said... until it was proven to them that they were wrong, but nothing from this patch other than your post. Again, if this was happening the way you think it is, there would be multiple threads on the forums, there are hundreds of people breeding in ARK every single day, people who hatch dozens or even hundreds of eggs per week, if this was a problem that was caused by the patch there would be at least half-a-dozen threads on these forums.

We're trying to help you here, but you're working really hard on refusing to be helped.

[Griffin998]

Like everyone's said, color and stat mutations are linked. And I'm not even going to touch the probability issues. If you don't know this...