If I was a betting man, I'd say probably 5 dragons (as it's the most common event dragon number and I don't think we have the weeks left to reach 10) and maybe something non-dragon at the end. Your guess is as good as mine though. 🤷♂️
I wonder if they will release a Jukebox. Shame to let the cool music go away with the event (other tunes seem to be tied to the Twilight Tower setting).
Sorry, been out of town for a week and was extremely busy leading up to it. I'm back now and trying to play catchup.
The testing thread needs a new direction soon. I got behind compiling the numbers over the last few weeks, I'll catch up and try to come up with the next step. The next focus will likely be on determining what impacts levels have on the results.
I am familiar with the Sandbox. The creator of the site is a moderator in these forums, although he has been fairly inactive this year. He has stated that his percent numbers are not accurate to any degree. The sandbox is an amazing tool for several things, for finding required breeding combinations and for determining the possible outcomes of breeding two dragons. You can put dragon A and B together, get a time, and easily see what dragon(s) you have made. It's also great for determining which combos have the least number of outcomes, or the lowest average fail time. Unfortunately it does not tell you the percentages, and it can be misleading with the numbers. For example, there are times when parent cloning is better, and other times when it is not.
We know most of algorithm now. There are a few numbers that I'm off by a percent or so, such as the normal hybrid result chances based on rainbow+rainbow for example, but we know it's around 1 percent. It might be 1.5, 1.25, or possibly even 2 or .5. Or more likely there are forumlas that calculate it that we haven't found. We know what happens when the result set gets too big, and how the rarer dragons scale down with their chances. We also know that despite what most people thought, breeding 4 elements in non BeB gives you the same chances for a specific dragon as breeding 3 because the outcomes for both keep it under the 100 percent cap. I used this to my advantage by breeding for 2 different dragons during some events without lowering my odds for either.
Your Meadow + Luminous test is a another proof that normal hybrids don't have set percentages. Dragons that don't have percentages in the file (normal hybrids and a few bugged goldens) fill in the 'gaps'. If there are 4 results, and the 2 goldens take 4 and 5 percent, then the 91 percent leftover is split between the 2 normal hybrids. We know using the epic cave will double the chances of the rarer dragons. And I believe a high level dragon will double the chances too, but that needs to be absolutely proved still. That's where I hope to go next, comparing dragons at levels such as 20, 16, 10, and 4 and comparing outcomes.
This was found by all of the testing in the earlier pages of this thread. My initial theory revolved around a threshold value in the files, but it quickly proved to be wrong. Weights is one way to describe it, if a dragon has a 4 percent chance in the file but there are so many results that the total adds to close to 200 percent, then the dragon will have a 2 percent chance to be bred.
For Satin and Bogberry, these dragons do not have all of their cloning chances in the file. For example, here is Bogberry: "breedingchances": { "normalparentclone": 0.01, "riftparentclone": 0.01, "socialclone": 0.1, "socialparentclone": 0.02 },
"normalclone" is missing, and so is riftclone. Satin is just missing normalclone, but has a riftclone value of 10 percent. When not having a value, it uses a default. That number is still something I need to work out exactly, but I believe it's close to 1/3 or 30 percent. So cloning this dragon in your normal and epic cave is 3x more likely than cloning it in the social cave, and more common than most dragons who have values set around 10 percent or so.
For Party and Aurora, the breeding chances (not cloning) is 3 and 4 percent. If you're using higher level dragons or the epic cave, these will be doubled or more. With more outcomes, they'll be lower - 754.75 percentages added up total. The numbers currently in BeB would come out to 0.388 and 0.517 percent for level 10 dragons not in an epic cave. I think people notice these more only because they're available in almost all your breeds and when they hit outside the rift, you get upset over the long timer. Testing that I've seen still have them showing very rarely.
Sandstorm and Dodo is a special case, and it's another that the numbers need to be validated. This is a situation where there are no normal hybrids to take up the extra 'room' in the percentages. Dodo, Sandstorm, Air, and Earth are not normal hybrids. That's an easy one to check, maybe I'll run it next on one of my alt parks. If you breed something like flower + poison 100 times, I'd expect to see 4 Smoulderbrush, 5 Pepper, and 45 each of Flower and Poison.
I agree on the cloning missing that needs testing. From what I've seen, those missing values are easier to clone in general and appear around 30 percent. One of our earliest tests is poison + poison, which ended up with 30 - 33 percent more poison dragons. That's where I'm getting my default number so far. Cloning Party dragons should be pretty easy, once you get two you have that 30+ percent chance. Sounds like that's what you were seeing in your park.
Does the 4 percent chance (your example) never increase, even if there are very few results (sum of percentages << 100%)? E.g. Sandstorm+Dodo? (Both chance1=5%, normalparentclone=1%). On top of my head, I wouldn't expect ~90% Air/Earth if I were breeding high level Sandstorm+Dodo?
Only the first breed numbers are in, but it already showed 11 air/earth and 1 hybrid. Maybe the 90 percent earth/air will pan out.
Cloning for all 4 Bright Nights dragons and the 10 Spellform Dragons is all Normal/Rift Clone 30%, Social Clone 10%, Normal/Rift Parent Clone 0.75%, Social Parent Clone 1.5%.
Monte and Sailback are 2% base chance. Toco and Vibe have two base chances each, 1% and 2%. IIRC, it was theorised that you get the higher rate if you already own the dragon but I don't believe that this has been tested.
Darcowl/Outrider/Quantum are 1.75% base, Riptide is higher at 2% base. No, I don't know why. Armament/Dash/Sparkles/Vanguard/Psyche/Bastion have the two base odds, breeding_chance_0 at 0.5% and breeding_chance_1 at 2%.
Fury is 3% base, Normal/Rift Clone 30%, Social Clone 5%, Normal/Rift Parent Clone 5%, Social Parent Clone 1%.
Anniversaries are mostly 3% base, 1% Normal/Rift Parent Clone, 2% Social Parent Clone, Undefined Normal/Rift Clone. Social Clone varies a little: Paper/Leathery are 5%, Cotton/Ironwrought/Wyrmwood are 10%, Garland is 15%.
Embossed isn't like the others. It has 0.5% base, 30% Normal/Rift Parent clone (not a typo), 10% Social Clone, 2% Social Parent Clone.
Last Edit: Sept 11, 2019 15:36:32 GMT -6 by Arkynos
What does it mean when a dragon has one-parent clone odds of 0% and non-zero regular odds (Chance 1), if you try to breed it with one parent? (E.g. Bog + Plant in the Normal/Social cave). Does the 0% override the 4% so that the effective odds are 0%, or is it just 4%+0% = 4% (as long as you don't add too many other elements)?
If it's the latter, I'm not sure I understand the point of defining the one-parent clone odds as 0%... like, why?
If I understand what I've read correctly, it's the sum of the odds. So in your example bog + plant will have a 4% chance of getting bog as an outcome - BUT only when bog is available (green hourglass showing). The rest of the time there will be a 0% chance of breeding bog using that combo. It makes a limited time dragon much harder to breed outside of its availability window.
'... 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.
