ACNH Guide On Hybrids

@Sloom

They should not have a flower of the same species next to them to be cloned. What I did was I paired the hybrids all together but making sure no same species of flowers are touching each other and water them. They could could clone themselves without a partner flower.
 
Can the mods make this a sticky thread? It's one of the most useful threads on this forum, and most people wouldn't even know it's here if it's not being constantly bumped. It also helps for lurkers to find it, since I believe most guests only look at the first page.

Appreciate it, but I don't particularly think this thread is sticky worthy. Bookmarking it is probably the best someone can do, and I'm trying to bump it at least once or twice a day so people come across it. It's also in my signature for people who notice. XD
 
  • Like
Reactions: Raz
Appreciate it, but I don't particularly think this thread is sticky worthy. Bookmarking it is probably the best someone can do, and I'm trying to bump it at least once or twice a day so people come across it. It's also in my signature for people who notice. XD
That's fair. I don't see signatures because I use TBT on mobile like 99% of the time.
 
I disagree with OP about this thread not being sticky worthy. there is literally no better explanation i’ve been able to find.

+1 for sticky, and thank you very much OP. super well written and easy to understand. just good stuff.
 
Evening bump!

Anyways, to those who are asking for this thread to be stickied... I do appriciate it, but there's not much I can really do. Personally don't think they would anyways, and I can't really ask them since they'll likely deny the request.
 
I don't think this thread is important enough to be stickied. A lot of people want their threads stickied, and stickied threads are only useful if there are very few of them. However, the staff may be open to including a link to this thread in the stickied FAQ thread if this guide is considered reliable and well-written.
 
I don't think this thread is important enough to be stickied. A lot of people want their threads stickied, and stickied threads are only useful if there are very few of them. However, the staff may be open to including a link to this thread in the stickied FAQ thread if this guide is considered reliable and well-written.

Basically this.

Also, afternoon bump 'cause I slept in (again).
 
Thank you for this thread!! I have a REAL basic question- if you plant 2 hybrids of the same color together (like black rose & black rose), will they always be able to clone themselves? Is that the best way to increase your hybrids once you get at least 2 of each?

Is that why I see people buying 2 of each hybrid color?

Like, that seems logical, but then again I just had to study game flower genetics so I just thought I would double-check!
 
Thank you for this thread!! I have a REAL basic question- if you plant 2 hybrids of the same color together (like black rose & black rose), will they always be able to clone themselves? Is that the best way to increase your hybrids once you get at least 2 of each?

Is that why I see people buying 2 of each hybrid color?

Like, that seems logical, but then again I just had to study game flower genetics so I just thought I would double-check!

Depends on genetics, but if the pair contain the same genetics they'll always produce the same result. To clone, you only need a single flower and water it and it will have a chance to clone itself with the same genetics, but a pair will normally produce more frequently.

To get more complicated, though:

So let's say in your case you bought two black roses that were spawned from two red roses that were from seeds. The black roses will have an R 'gene' of 2. When you write this with the rest of the genes, it ends up being 2-0-0-0 (R, Y, W, S) But on the other hand, you bought two black roses and one has the genes 2-0-0-0 and the second one has genes of 2-0-1-0, the child will always be 2-0-1-0.

Basically, it depends on the genes. But generally, hybrids will tend to produce the same colour if paired with the same colour in most cases.
 
UPDATE!

Due to a friend of mine off this site who enjoys messing around with the genetics results of ACNH's flowers, they have found a new method of obtaining Blue Roses!

This third method is about the same difficulty as the second method and is a bit shorter as well, but it requires a bit of multitasking and keeping track of a few things. You'll find it below the Method 2 spoiler!
 
UPDATE!

Due to a friend of mine off this site who enjoys messing around with the genetics results of ACNH's flowers, they have found a new method of obtaining Blue Roses!

This third method is about the same difficulty as the second method and is a bit shorter as well, but it requires a bit of multitasking and keeping track of a few things. You'll find it below the Method 2 spoiler!

Morning bump! : )
 
There’s also another way to get blue roses from a user on reddit which they’ve confirmed works

White and white for purple

yellow and white for white*

White* and purple for purple* or purple dud that useless

Red and yellow for orange

purple* and orange for orange*

orange* and orange* for blue or red**

red** and red** for blue
Guess you’d call this the medium route?

You can test the purple by breeding them with seed yellow, and the purple* will produce yellow half of the time.

I am using this method currently. Have only been testing purple two days (1 then 3 of them) so I have not confirmed any yet.

BLUEROSEGUIDE.png


Edit: this is not my guide. Let me know if it is not okay to post it here and I will remove it.
Edit2: I thought it was probably okay to post since the image credits its creators.
 
Last edited:
You can test the purple by breeding them with seed yellow, and the purple* will produce yellow half of the time.

I am using this method currently. Have only been testing purple two days (1 then 3 of them) so I have not confirmed any yet.

BLUEROSEGUIDE.png


Edit: this is not my guide. Let me know if it is not okay to post it here and I will remove it.
Edit2: I thought it was probably okay to post since the image credits its creators.

Yup, and the third method I recently added that was created by a friend is basically a reduced version of this, and happens to be the "BackwardsN Method." Though, the one I have adds an extra step to reduce RNG.

Friend created a program that tested theoretical best methods, and we got the BackwardsN Method.
 
Yup, and the third method I recently added that was created by a friend is basically a reduced version of this, and happens to be the "BackwardsN Method."

Friend created a program that tested theoretical best methods, and we got the BackwardsN Method.
That's really cool. I was writing one in C# after I read a guide on the genetics, but figured people were already doing this and felt redundant. I don't have practice flattening recursive algorithms, so it was a bit of a headache for me, but it was fun to work the problem for a bit. Really cool that your friend completed a program to do this. Wish I was using the new method to breed my roses, but too far in now.
 
That's really cool. I was writing one in C# after I read a guide on the genetics, but figured people were already doing this and felt redundant. I don't have practice flattening recursive algorithms, so it was a bit of a headache for me, but it was fun to work the problem for a bit. Really cool that your friend completed a program to do this. Wish I was using the new method to breed my roses, but too far in now.

The BackwardsN method is theoritically better as it provides more chances at Blue Roses, but the method I have adds in a few steps to reduce RNG, since the BackwardsN method has a few 12.5% chances, compared to the one I have that adds the extra step or two to increase odds to 25% from 12.5%, but has less Blue Rose chances.
 
The BackwardsN method is theoritically better as it provides more chances at Blue Roses, but the method I have adds in a few steps to reduce RNG, since the BackwardsN method has a few 12.5% chances, compared to the one I have that adds the extra step or two to increase odds to 25% from 12.5%, but has less Blue Rose chances.
Yeah, I prefer your method. I don't mind the extra steps if it removes some of the bottlenecks.
 
Yeah, I prefer your method. I don't mind the extra steps if it removes some of the bottlenecks.

Yup, for visual this is a comparison:

1588457849326.png


It essentially removes one RNG bottleneck in favour for better odds of the correct hybrid red rose.

Additionally, BackwardsN's method can have a slight modification done to it to increase odds:

Once you get a 1220 Red from the two 1210 Orange, you can replace one of the 1210 Orange with a 1220 Red for increased 1220 Red odds (25%), and increases Blue Rose to 12.5%
 
Last edited:
Back
Top