Bogberry offers 4 elements: fire, plant, water, and light. That keeps the result set low, for a total of 35.5 percent when adding the rares and epics today. And 4 lower if you did this prior to Smoulderbrush's release. The rest of the goldens are at 5 percent, with rainbow at 1 and double rainbow at 0.5. Are you saying that in the Epic cave, the six golden hybrids were 10 percent each, and rainbow was around 3 percent? 82 isn't really enough attempts to nail down a change with the rainbow exactly, but it is with the goldens when taken as a group.
What you're saying leads me to believe that the Epic cave doubles the odds. 5 percent becomes 10, 1 percent becomes 2. That's a simpler equation than adding different numbers to each dragon (especially when the numbers aren't in the data file), and I believe simpler is more likely when looking at coding things. We could certainly use more testing here, but that's a great start.
And you're right, Portent x2 isn't going to impact the odds at all. The result set is huge, like rainbows, and the odds already add up over 100. They'd scale back down to exactly how they would be without doubling them.
I'd be very interested in hearing how the levels impact the odds too.
This is a great idea. Anything that helps breeding that elusive dragon is helpful.
Thank you Quil for taking the time to crunch the numbers and share the results with all of us.
I'm happy that you are not planning on posting spoilers about upcoming events and new dragons, especially if you have access to that information many weeks in advance of it actually happening. It is nice to still have a few unknowns.
Another interesting set of dragons to look at are the normal hybrid dragons such as Poison. We solved most of how the normal hybrids work in the testing thread, but seeing what's in the file reinforces our findings. Some of that information we found is going to be repeated here, just to make it all clear.
While the breeding hint is plant + fire, it has no defined breeding chance in the files. Nothing at all set for any normal hybrid. Cloning also looks different:
In normal and rift caves: 1 poison parent is 1% In the social cave: 1 poison parent is 10% Social with 2 poisons: 20%
And that's it, nothing is defined about cloning two poisons together in normal or rift caves. There are many dragons of all types missing various cloning numbers. That led me to believe there might be a default. And what are the breeding odds for a dragon that doesn't have odds listed? That's why we test!
Our Poison + Poison test revealed the following results. 61 percent Poison, 31 percent Flower (the other normal hybrid), 4.7 percent Pepper and 3.1 percent Smoulderbrush. Now the file shows Pepper at 5 and Smoulder at 4 and our numbers are pretty close given the small testing size. It appears there is a default 30 percent cloning chance for Poison x2. Then the percent chances for Pepper and Smoulder take up 5 and 4 percent and any left over is split between the two normal Poison and Flower dragons (30.5 percent).
So normal hybrids take over the left over chance after checking for cloning and for the rares with defined numbers. When we tested Poison + Mountain, I would expect the normal hybrids results to be at 48 percent and we found them to show up 51 percent of the time. Why the difference? It could be population size being too small to be accurate (under 500), it could be some of the results were done in the rift and included rift dragons depending on the rift alignments, or it could be off due to my next finding.
So what happens to normal hybrids when you breed two rainbows, such as our Rainbow + DR test? They take the leftover percentages, and two rainbows will fill up and go over the total 100. That's where we have to scale them back by 22 percent, as done in the previous posting. I would expect normal hybrids to just completely fall off the result list, but they didn't. After a large number of tests, a normal hybrid showed up roughly 2 percent of the time. Given there are 80 normal hybrids, your odds on getting any specific normal hybrid with rainbow + rainbow is lower than the rarest of rare dragons.
I'm guessing there is always a 1-3 percent chance (looks like 2) to get a normal hybrid, and if that check is added to all breeding pairs, that could explain our 3 percent difference in mountain + poison on normal hybrid results.
Another interesting note, there are 7 golden hybrid dragons that's don't have breeding odds. Coldforge, Conifer, Frostflower, Permafrost, Dandelion, Cyclone, and Subterranean. They all act like normal hybrids, none showed up as results with two rainbows, and I ran a cold+metal test and found Bonescale was low at 4 percent (5 in the file) and Coldforge, Steel, and Mine all split for 30 percent each. I'm guessing this is a bug.
Any ideas on where fire + water is 100 percent of the time an air dragon? Maybe the first time you breed one? Maybe in one of the other caves?
I just bred Fire + Water in my new level 32 park on the Epic Breeding Island and got Steam. I had not bred an Air dragon in that park, only purchased. So new accounts that haven’t bred Air doesn’t give the 100% guarantee.
It was always my understanding that the Air dragon couldn’t be bred with Fire + Water, but it could be bred with Coral or Salamander in the combo, unlike the other element dragons.
So I took one of my small parks and ran some tests too. I owned an air dragon, so sold it and tried fire + water in the normal cave as I don't own the epic breeding cave there. Got an air dragon. While it was in the nursery, I tried again and got something else. I sped up and sold the egg from the nursery and tried again. I ran over 10 tests in a row, as long as I don't own an air dragon or have one in the nursery then I always got an air. Once I owned one, it would be something else.
Could you check in your normal cave too when you get a chance, selling any existing air dragons first? Perhaps the multiple numbers work only in the normal cave and are based on how many you own.
I'm still not very clear on exactly how cloning works in the file, and it's one of the biggest gaps I have in determining the actual odds. I'm hoping others have some ideas based on experience.
We talked about the Poison dragon, and that it's missing some of the cloning numbers. It has both social (1 and 2 parent at 10 and 20 percent), and single parent normal and rift odds at 1 percent each. Our testing of roughly 500 breeds in the rift show that the cloning chance is 30 percent with two poisons.
There are a lot of other dragons that don't have cloning numbers in various places. For example, Leap Year and Celtic dragons have only social numbers - 5 percent for 2 parents, 1 percent for 1. I know they can be cloned in the normal breeding and rift caves, but the question is what is the percent? Do these rarer dragons also default to 30 percent when cloning in a normal cave? And what's the default number when cloning with just 1 parent and it's not defined? Most dragons that have values fall in either 1 or 5 percent, so I'm guessing it'll be one of those two.
Here's the counts of how many dragons have each clone type defined. Ones with 'parent' in the name are single parent cloning: CLONE_TYPE count(*) ----------------- -------- noramlclone 1 normalclone 234 normalparentclone 447 riftclone 229 riftparentclone 429 socialclone 511 socialepicclone 3 socialparentclone 517
The Auroch dragon is bugged and has the misspelled normal name. Morpho, Serket, and Trilobite all have the socialepicclone flag. This was an older flag removed from dragons a while back, and I'm not sure how or if it works today. It refers back to the days when you needed an epic dragon to clone against another epic.
I just bred Fire + Water in my new level 32 park on the Epic Breeding Island and got Steam. I had not bred an Air dragon in that park, only purchased. So new accounts that haven’t bred Air doesn’t give the 100% guarantee.
It was always my understanding that the Air dragon couldn’t be bred with Fire + Water, but it could be bred with Coral or Salamander in the combo, unlike the other element dragons.
So I took one of my small parks and ran some tests too. I owned an air dragon, so sold it and tried fire + water in the normal cave as I don't own the epic breeding cave there. Got an air dragon. While it was in the nursery, I tried again and got something else. I sped up and sold the egg from the nursery and tried again. I ran over 10 tests in a row, as long as I don't own an air dragon or have one in the nursery then I always got an air. Once I owned one, it would be something else.
Could you check in your normal cave too when you get a chance, selling any existing air dragons first? Perhaps the multiple numbers work only in the normal cave and are based on how many you own.
I have a small Level 21 park. I made it 'accidentally' while signing into Google Play to see if I could play my IOS games on Android.
After reading this post about Fire & Water, I bred my Level 11 Fire & Level 11 Water in the Epic Breeding Island and first breed I got Air. I already have a purchased Air dragon in the park. Second breed - while Air was in the nursery gave me a Coral. Third breed - after selling the newly incubated Air dragon - gave me another Coral. Happy to do a few more breeds if you would like the data.
So I took one of my small parks and ran some tests too. I owned an air dragon, so sold it and tried fire + water in the normal cave as I don't own the epic breeding cave there. Got an air dragon. While it was in the nursery, I tried again and got something else. I sped up and sold the egg from the nursery and tried again. I ran over 10 tests in a row, as long as I don't own an air dragon or have one in the nursery then I always got an air. Once I owned one, it would be something else.
Could you check in your normal cave too when you get a chance, selling any existing air dragons first? Perhaps the multiple numbers work only in the normal cave and are based on how many you own.
A breath of fresh Air! Sold all my Air dragons in the level 32 park, bred Fire + Water in the Enchanted Breeding Cave... and got good ol’ refreshing Air!
To further the experiment, I bred another Fire + Water on the Epic Breeding Island while the first Fire + Water was still breeding the Air dragon in the cave... and got Steam, again.
You cracked the code! If there’s an Air dragon in the park, or if there’s one incubating, or if there’s one being bred, players will not receive the 100% Air guarantee from Fire + Water. There cannot be any hint of Air anywhere.
Never would’ve thought Air has even more breeding quirks. Let me know if you need any more breeding tests in the Vale. I’ll try to check the forum more frequently so it’s not such a long wait till I post results. Very interesting, indeed!
Friend ID: adlerist#5274
Due to a potential limit on the number of friends a player can have before issues arise in the game, I will only be able to add Forum members. Thank you, for your understanding, and please consider joining us!
Oh sorry, MinyaZ! I didn’t see your post, as I was posting at the same time. Ah well!
Friend ID: adlerist#5274
Due to a potential limit on the number of friends a player can have before issues arise in the game, I will only be able to add Forum members. Thank you, for your understanding, and please consider joining us!
After reading this post about Fire & Water, I bred my Level 11 Fire & Level 11 Water in the Epic Breeding Island and first breed I got Air. I already have a purchased Air dragon in the park. Second breed - while Air was in the nursery gave me a Coral. Third breed - after selling the newly incubated Air dragon - gave me another Coral. Happy to do a few more breeds if you would like the data.
Okay, so this changes everything. No Air in the park isn’t the requirement to breed Air with Fire + Water. Dang, it really seemed like that was the catalyst. So what’s so different with MinyaZ’s park? MinyaZ, did you sell the Air dragon prior to breeding Fire + Water?
Last Edit: Jun 8, 2019 19:23:42 GMT -6 by adlerist
Friend ID: adlerist#5274
Due to a potential limit on the number of friends a player can have before issues arise in the game, I will only be able to add Forum members. Thank you, for your understanding, and please consider joining us!
After reading this post about Fire & Water, I bred my Level 11 Fire & Level 11 Water in the Epic Breeding Island and first breed I got Air. I already have a purchased Air dragon in the park. Second breed - while Air was in the nursery gave me a Coral. Third breed - after selling the newly incubated Air dragon - gave me another Coral. Happy to do a few more breeds if you would like the data.
Okay, so this changes everything. No Air in the park isn’t the requirement to breed Air with Fire + Water. Dang, it really seemed like that was the catalyst. So what’s so different with MinyaZ’s park? MinyaZ, did you sell the Air dragon prior to breeding Fire + Water?
I'm not sure I'm really clear on the steps taken here. The first breed was an air, while an air was already somewhere in the park. That's fine, as there is a 30 percent chance of air when you own one. The air in the nursery and getting Coral makes sense too. Then on the third breed you sold the newly incubated air and got a Coral. Except I don't really follow if you sold your original air too, or if that was still around. If there was another air anywhere in the park, menus, etc then getting the Coral still makes sense.
Okay, so this changes everything. No Air in the park isn’t the requirement to breed Air with Fire + Water. Dang, it really seemed like that was the catalyst. So what’s so different with MinyaZ ’s park? MinyaZ , did you sell the Air dragon prior to breeding Fire + Water?
I'm not sure I'm really clear on the steps taken here. The first breed was an air, while an air was already somewhere in the park. That's fine, as there is a 30 percent chance of air when you own one. The air in the nursery and getting Coral makes sense too. Then on the third breed you sold the newly incubated air and got a Coral. Except I don't really follow if you sold your original air too, or if that was still around. If there was another air anywhere in the park, menus, etc then getting the Coral still makes sense.
Thanks!
Sorry, for the confusion. Tried hard not to cause any....
At all times I had an Air in the park, so Quil yes, it makes sense that I got a Coral for my second and third breed. It even fits into the 33% expected outcome.
With the results that you and adlerist are getting breeding Fire & Water with no Airs in the park, it would seem that both the breeding numbers you found in the file are relevant. Chance 1 = 100% if you don't have any Air dragons and Chance 2 = 33% if you do own one.
So the few dragons that have the two numbers likely just mean that it's a bit easier to get them the first time, and harder with subsequent attempts. I wonder what it means with Motley and 3 numbers, I'm going to guess it's if you own 0, 1 or 2. Or it could be 1 and an egg. Given how small and close the last two numbers are, I don't think we'll ever have a correct answer.
The more I work on this 'exact' breeding number simulator, the more I realize just how complex it can be. I was already taking into account which cave you use and which rift elements are available in the rift cave, but now I'll have different results based on what dragons you may or may not already own. Crazy!
There's another data element in the file related to breeding odds, but I'm not really sure yet of it's meaning. Most dragons have a value for 'maxthreshold'. This is a number between 0 and 255, and almost always changes based on the breeding chance for the dragon.
Dragons with a 0.5% percent breeding odd have a value of 221. Dragons with a 5% chance have 22. 1% is 110, 20% is 5, and 3% is 36, etc. But it's not always the case, and there are several where the odds and threshold don't match up. I thought for a while that the threshold would be used when the numbers go over 100 percent, to help scale it down, but I can't make any of the math work right with what we've seen in tests. I know it's not used straight up for breeding odds as a simpler way of using numbers instead of percentages, as Coldforge is bugged and is missing its percent, but its threshold is 22 (5 percent) and it acts as though it doesn't have any breeding percent in our tests.
Anyone familiar with what max thresholds are when dealing with odds? I just don't want to miss any clues from the file.
So the few dragons that have the two numbers likely just mean that it's a bit easier to get them the first time, and harder with subsequent attempts.
I have definitely noticed that to be the case with some dragons. Seems like when I get a dragon in the first few breed attempts it is hard to get a second. Conversely, when it takes 100+ attempts, once I get that dragon I will get multiples in a row right after.
If you want to see a practical way to use this data (and I'll post the numbers once I get everything formatted in a nice way), here's a real example of something I did today:
I need the new re-released kite dragon in the rift, never got one there. I also still need a xanthium egg in the rift. Now I can breed the kite elements (metal, air, lightning) and try for kite at 4 percent until I get one. Or I could try for Xanthium with metal and earth until I get the egg at 2.5 percent. But if I breed metal, air, lightning and earth all together right now, the sum of the rares is 47.5 percent. That means that using all 4 elements keeps Xanthium at 2.5 percent and Kite at 4 percent and they don't scale down.
The key: It doesn't hurt your odds on getting either one of these dragons by using all 4 elements and trying for both at once!
I now have a 4+2.5 = 6.5 percent chance of getting a dragon or egg that I need, instead of just 4 or 2.5.
Now if you're trying for them outside the rift, you'll want to consider the lengths of the fails. Adding all 4 elements adds rainbow and DR to the mix and time is important when not using rift caves.
Let talk exclusion lists. There are multiple exclusion lists in the file, and multiple ways dragons are added/excluded from the various rewards.
Some of the values are just not clear on exactly how they get used in the game. Others have more complex definitions. If you look for the mystery eggs, there are various headings such as "magicegg_rift_rare" with all of the various level costs in the file, maxing out at 22,500 Etherium for this rare one. It has a loot table id of "loottable_magicegg_rift_rare", and when you go look that value up you'll find it has loot with a value of "magiceggreward_rift_rare" = 1.
The magiceggreward_rift_rare is defined with values of "include_owned": true, "isegg": true, "rarity": "rare", and a "not" list of the basic 10 element dragons. As far as I can tell, these eggs can contain all rare dragons.
Now if you look at the reward egg for Galaxy dragons, it's similar but it has an additional value of "include_ids", listing out 18 different galaxy dragons. Rigel is NOT in the list. You can't get it from a Galaxy egg. Epic and commons do not have this include list. Anyone know if you can get Galaxy eggs from normal mystery eggs in the rift? I've not bought enough regular eggs to find out. If you can, then Rigel may be available that way.
Chrysalis mystery eggs also have an inclusion list with only the original dragons, none from the recent event.
Now the daily login bonuses has it's own rules for every single day set. On day 2, the Galaxy egg day, it's flagged as a milestone, galaxydragon, and it has a rewards defined. The rewards says fallback: polariandragon and "not": solarflaredragon / rigeldragon. So Rigel is not available in the daily login rewards. The other daily eggs do not have 'not' lists set. I've not seen anything in the files that state these need to be new dragons, but we know they are and I've probably just missed it and it's hiding somewhere else. It's a big file. Or it could be "include_owned" defaults to false, and they just don't list it. That's more likely. This is why Pearlescent would show up for some people on login, it's not excluded from the epic list.
Also, under the heading 'gamestentbuilding_0', there are two exclusion lists that are being kept up to date. There's an exclusions list and an exclusions_egg list. It appears to be related to the winter 2017 event originally, but they keep updating the dragons in the list. For example, Rigel and Pearlescent are in the egg exclusion list, but not the dragon one. This building value seems to get called during all sorts of events, so it may be reused code and lists of dragons available during most event rewards. I can't tell you yet if these lists are tied to the bought mystery eggs or not, adding additional exclusions, or if it's separate. It seems separate.
As a side note, I find it crazy that they've got information for events going back years and years in the file. It's so bloated in size in part due to the fact that it has tons of stuff from events going back to 2015. I thought it might be because they want to reuse things from year to year, but they always get new entries and clean out just part of prior events. The file is at least twice the size as it needs to be, if not 3x or more. Or even a lot more. I'm sure the game would speed up with less data to load, be prone to less errors, download faster, and use less memory (and help with crashing) if this file were cleaned up.
As a side note, I find it crazy that they've got information for events going back years and years in the file. It's so bloated in size in part due to the fact that it has tons of stuff from events going back to 2015. I thought it might be because they want to reuse things from year to year, but they always get new entries and clean out just part of prior events. The file is at least twice the size as it needs to be, if not 3x or more. Or even a lot more. I'm sure the game would speed up with less data to load, be prone to less errors, download faster, and use less memory (and help with crashing) if this file were cleaned up.
^^THIS!^^ Dear BFS: The Administration here at the DragonVale Community would be willing to rent out Quil for a nominal fee to make this happen.
There is a 'luck' value on dragons. Most / all of the gem dragons all have a value 'eligibleforluck'. Most of the galaxy dragons also have it, as do a small sampling of other dragons such as Ouroboros, Bronze Oly, Eve, and Forest.
Different rarities have a luck_threshold. 38 for epic, 23 for gemstone, -1 for hybrid, and 14 for rare.
There's a luck definition that has 'bonus_breed_chance' = 10, with 'num' = 6.
It sounds like it should give these dragons a 10 percent extra chance after 6 attempts, but how it really works is unknown as we do not see the code. I have a feeling that this isn't working this way, as Rigel is eligible for luck and has a 0.5 percent breeding chance (which will be lower with the large result set). If it was 10 percent, we'd all have one. It's not clear to me exactly if you when the luck kicks in. And does the 6 mean attempts? Does breeding in another cave for something else reset it? All valid questions.
Or maybe it does work, and since there is no luck threshold for 'Galaxy' it doesn't kick in for Rigel. And if the thresholds are used instead of the 10 percent base chance, then it's a different bonus for each type. For example, gemstones at 22 threshold is just slightly over 5 percent. So maybe it adds 5 percent instead of 10. So many questions.
For example, someone was asking about trying to clone their Pearlescent the other day and how much trouble they were having, when it's in fact better to try to just breed a second than clone one.
I would never have guessed that this was the case. I assume you are saying this because you’ve actually seen the numbers in the file. Care to share the Pearlescent numbers with us? I, for one, am curious. 🧐
Last Edit: Jun 9, 2019 21:52:55 GMT -6 by greggerman
Dragon Warden at WestWyrm Isles Dragon Sanctuary, est. Sep 2011 Gladly accepting friend requests for Coop Breeding. My daily gems and treats are spoken for by my long-time Game Center friends. GCID: da real dub g DV Friend ID: Greggerman#1656 DV Friend Link: dragongame-prod.appspot.com/friendid?friendid=MuBigOo5OjUBoH0iMHF4 House of the Mountain Sun ⛰☀️
Pearlescent breeding chance: 0.5 percent with starwing / prism and the right alignments. Elements provided: fire, plant, earth cold
Normal Clone with 1 or 2 Pearlescent parents: 0 Social Clone with 1 or 2 Pearlescent parents: 0
Rift clone with 2 Pearlescent parents: 10% Rift clone with 1 Pearlescent parent: 0.25 percent.
So as you can see, breeding a Pearlescent in the rift with a plant gives you a quarter of a percent chance of getting one. Breeding one with a starwing and prism is double the chance, at half a percent, but a large result set can scale that number down.
Prism offers water and light to the result combo. Starwing offers earth, cold, fire and plant for a total of 6 combined elements. You'd think this might put you over 100 and start scaling back the results, and it does, but only to 113.5 percent. Add in the 1 percent chance to clone starwing, the 2 percent chance of getting a normal hybrid, and whatever chance cloning prism has (it's not defined, we need to test this but I'm guessing 1 percent) and you're at 117.5%. That means you need to scale the odds back by 100/117.5 = .851. Even if cloning prism is 10 percent or more, you're still well below the 200 percent that would even the odds.
Breeding a Pearlescent = 0.5 * .851 = 0.4255 percent. Cloning it with 1 parent = 0.25 percent.
Breeding is better! Add in BeB and the number jump significantly to 311.5%, dropping your chances down to 0.16 percent. So during BeB, cloning is better.
This is exactly why I wanted to do this project and work out all the details. There is no way a random player could possibly guess the above information without the data and the community like ours to help interpret how it all works.
Quil The games tent exclusions are for Name That Egg/Name That Dragon! Which explains a lot of why they keep getting updated for events and such.
I also think the Prism chance could be higher than 1%. I'm reminded of my experiences attempting to breed Rift Misfortune, that Apocalypse has an undefined Rift Parent Clone, and that I was getting at least one a day. If I were to guess, I'd place the fallback percentage at either 5 or 10%.
This may just be confirmation bias of course! Proper testing would be a good idea.
Post by greggerman on Jun 10, 2019 18:15:35 GMT -6
Quil First of all, let me reiterate how much I appreciate the work that you're doing by digging through the file and doing your best to make sense of it! 🌟
Have you looked at the numbers for Rigel? It's clearly not available at the moment, but assuming that there is a BEB coming up, it hopefully will be. I was looking at the Sandbox and it looks like including Rigel as one of the parents drastically increases the number of fails and might outweigh any benefit gained by cloning. And that's assuming there is a possibility of cloning with Galaxy dragons in the Rift. I definitely did not receive any parent clones during my quest to breed my Rigel, and I don't recall hearing about anyone else who did either.
Dragon Warden at WestWyrm Isles Dragon Sanctuary, est. Sep 2011 Gladly accepting friend requests for Coop Breeding. My daily gems and treats are spoken for by my long-time Game Center friends. GCID: da real dub g DV Friend ID: Greggerman#1656 DV Friend Link: dragongame-prod.appspot.com/friendid?friendid=MuBigOo5OjUBoH0iMHF4 House of the Mountain Sun ⛰☀️
'... overwhelming goodwill, and witty humour, profound grace and compassion blended with the spice of relentless taunts.'
Note: This is an independent forum and is not affiliated with Deca Games, the developer of DragonVale.
It was created by and is managed by DragonVale fanatics.
All who observe the following brief guidelines are welcome to this great community: no personal attacks & no spam.
