The Big Reveal

[Quil]

The staff here at the Dragovale Community are suggesting an exciting change to one of our internal policies!

There is a continually updating file that contains game rules, upcoming features, dragons, and events. This file is downloaded to your device every time it changes, and some of our team here have the ability to read this file. Historically the game was much simpler, the only important information was spoiler info and it was decided not to release that news.

We did not want to cause confusion for new players, reveal spoilers for those who don't want the info, and we certainly don't want people asking Backflip about features or changes that were never meant to be seen by us. And we don't want to spoil the game for Backflip and take away from the excitement of their announcements. And not everything in the file always gets released, or it may be at a much later time. For example, the Gaia leveling was seen more than two months ago, but none of us knew if or when it would be available.

However the game has become increasingly complex and there is information available that may help players get the most from their gameplay, such as with breeding rules and odds, mystery eggs, and event chests. So, we are considering sharing and discussing some of the data from the file, in this quarantined thread, to keep all spoiler info in one place only.

The file is massive in size and difficult to decipher, as there are nearly half a million variables found in the data. It contains the cloning and breeding odds for many dragons, but does not answer all the questions about breeding, and creates a whole lot of new ones.
The thread I started on testing dragon breeds is now starting to help answer many of those questions. We believe this is information that everyone should have if they want it, as it can change how you play the game, and shows how what you breed impacts the odds for specific dragons. It will also give you a better feel of whether its worth spending gems (or cash) based on the odds.

Another example of important information in the file is which dragons are on the exclusion list for mystery eggs. For example, the new Rigel dragon is still on the list. Knowing what's on the list can help players determine if they want to spend resources on mystery eggs, or hold off spending. All of this is information that Backflip should be providing to their players, for example loot box odds being displayed in game are a requirement at least by Apple.

Not everything is in the file. These are rules, but the application code itself determines how to use the rules and likely has many other rules stored internally. We don't see that piece of the puzzle. There are also variables that are passed to the game at login, which we don't see. For example, rules are set in the file to remove visions for players on specific devices and payment tiers, but we can't tell you what device or payment tier you're assigned to in game.

The plan is to begin with a series of posts here that will discuss the actual breeding odds and the mechanics behind the scenes in game. A lot of my findings are from a mix of our own testing with what we're seeing in the files. We'll also be happy to discuss other existing game mechanic questions here, though again, we don't have all the answers.

Please feel free to reach out to us with any questions or concerns, we certainly don't want to alienate anyone from the community that doesn't want to know this information. We will do our absolute best to keep things separate, and we still do not plan to spoil upcoming changes to the game. We would like to invite you to give your views on this planned new direction. If there is a clear majority in favor of seeing this information then we'll begin the process. If there are concerns, lets discuss them.

Thank you!

Edit: Here is a spreadsheet containing the base breeding and cloning odds for each dragon.

[nachokitty]

Love it. I always want more information.

[greggerman]

I think this is a great idea and would be a wonderful forum for people who want to know more about how the game works. I, for one, have dedicated untold hours to this game and am always looking for ways to maximize the utility of that time. Please do proceed!

And anyone who is not interested doesn’t have to read the thread.

[olbiddy]

I think this is a great idea.  The game has become more confusing, Rigel being 1 example, as rules seem to have changed a lot.
I also think this would ease a lot of frustration players are feeling

[cecilia]

How wonderful....

I feel as if I've been admitted into the secret dungeon of MI7 or the equivalent.

Thanks and congratulations to all the sleuths working on behalf of us lesser mortals.

Cecilia at the Duckpond

[Quil]

We've had a lot of views and a few posts, and the comments so far appear very positive. There's so much information to share, I could write page upon page about it and still not hit all the topics. I'm going to try to break things down into smaller posts, but I tend to ramble on with details so they may not stay very small.

The best place to start is likely a discussion of what kinds of things are in this file, and how are we going to use that data. As mentioned earlier, this is a huge file with roughly half a million values in it. You can try to look at it manually, but it's a lot to take in and not formatted.

I've done computer software development and database work for decades, and I jumped at the chance to help parse this thing. Other mods here have certainly read through the file and have a good understanding of it, but I wanted to do the next step. I'm currently running a Linux server where I parse the file and load the data I'm interested in into a SQL database. From there I can run queries to look at breeding percentages and other information on all of the dragons. I'm currently working on building a breeding simulator, similar to the sandbox, but without the really nice front end. I'm just not a web guy. But I'm trying to get the results to bring back accurate percentages, which is my first end goal.

So what does the file have on breeding information? It has required combinations, broken down into elements, dragons, rift requirements, and other things. And it has a breeding chance percentage. For most things, but not everything. And to make it more confusing, some dragons have more than one. Others have additional rules, such as having a lower chance during an event and a higher one later. Dragons you're allowed to breed with are listed. And elements provided to a breeding combo are also there. Cloning chances, which differ based on using one or two of the same parent, and whether its in the normal, social or rift cave is in the file. For some dragons. Some have only a few listed, others have them all.

The rules on how these percentages work exactly are not found here. Bonus calculations for having higher level dragons, epic caves, etc are not in the file that I've found. What happens with the cloning chance if the value isn't there? If there are three breeding chance numbers, which one is used? Most normal hybrid dragons don't have breeding chances, so how does that work? And what happens when you breed so many elements together that the result set adds up to well beyond 100 percent in chance? These are all questions we've attempted to answer with our breeding tests, and I believe I have answers to most of these except the bonuses.

It's an interesting study, and there are times when cloning a dragon in the social cave is better than a normal cave, and vice versa. Or when just breeding for a dragon is better than trying to clone it with one parent. 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. This is exactly why we've decided to share this information, the game has gotten crazy complex and this can help players make better decisions.

[braff20]

Super excited about this. Can’t wait to see the great information you will share.

[Quil]

Lets take a look at some of the numbers behind the scenes, and how they play out in the game. This is just to give you an idea of what we're looking at here. I'll start with a newer golden hybrid, the Fever dragon. This was a dragon that came out during the Color Guard event. The file has two breeding hints for it, elements cold and fire. In the same area, it has a breeding chance of 0.035. That's 3.5 percent. What's extra intersting about this dragon is that there's an event driven variable too, with the breeding chance of 1 percent during Valentines Day 2019 (Color Guard). So yes, we can confirm that the Fever dragon was harder to get when it was part of it's own event. It then became easier after the event ended. 3.5 times easier.

In another area, we can see the cloning chances with all six values available. Normal cave and rift cave, both parents as Fever, is 30%. Normal and rift cave with one Fever dragon is 1%. You can see the significant drop there, and why for some dragons cloning a single parent isn't as good as actually breeding it again. But with Fever, you can do both - Fever + Fire would give you 3.5 for breeding and 1 for cloning, for a total of 4.5%.

The social cave has different values. You can socially clone Fever with two parents at 15%, and 3.5% extra for having the right combo to just breed it. You're twice as likely to get the dragon if you clone it in your own cave instead of the social one! Now, if someone else has a Fever and you do not, you can try to clone it in social with one parent at 4% for the social clone and the 3.5% for the correct breeding odds at a total of 7.5%.

Our testing numbers with Mountain + Poison helped reinforce the 3.5% chance. I need to input the last couple days into the data from our testing thread, but Fever is noticably lower than the other golden hybrid results. Those hybrids are all defined at either 4 or 5% and Fever is the lowest, and it shows in our tests. I love it when testing helps confirm our numbers!

Now Fever is at 3.5% with just cold and fire elements combined, but as long as you don't add too many more results to the final set then you can try for more dragons too. Lets say you need Fever and the newest Xanthium dragon, a very low 2.5% dragon with earth and metal. If you breed all 4 elements together then you'll still have the 3.5% chance for Fever but also a 2.5% for Xanthium. Your odds do not lower, but you now have a 6.0% chance of getting one or the other new dragons. You're better off trying for two at a time if you need several dragons, than trying for just one.

That's not the case when adding too many elements and results (you're safe up to at least 20 usually), and I'll get into that in another posting.

There are still things we don't know about breeding a Fever. I can't tell you what happens when I use a level 21 dragon vs as level 10, does that raise the 3.5% odds? That's something we'll need to check in our testing thread. I'll need to put together a list of questions we have, and work out how we'll get each of these tested over time. My guess, from my own observations, is that the bonus is very miniscule.

[adlerist]

Jun 4, 2019 16:55:50 GMT -6 Quil said:

However the game has become increasingly complex and there is information available that may help players get the most from their gameplay, such as with breeding rules and odds, mystery eggs, and event chests. So, we are considering sharing and discussing some of the data from the file, in this quarantined thread, to keep all spoiler info in one place only.

When it comes to event treasure chest odds, breeding odds, mystery egg contents, no Tolzar after purchases, etc, players don’t appear to consider this information spoilers (I’ve never seen anyone complain about or label these discussions as spoilers). The info isn’t new — if anything it’s confirming players’ long-term understanding of how DV functions — while spoilers, by definition, refers to new content, e.g. unreleased dragons and features. So I don’t think you have to be too concerned about spoiling, and I suspect spoiler-free players will appreciate reading about important and consequential information regarding the game.

If there are posts about unreleased content and/or features, it’d probably be best to keep them separate in a clearly titled 'SPOILERS!' thread so nobody accidentally clicks on them, especially since there’s no option for a spoiler cover. But it doesn’t appear there’s any intention to create these spoilt posts.

Since most mobile games seem to have at least a few players who have access to the game files, I’ve wondered if DV had any. There’s an occasional poster on Reddit, and there’s probably more on FB, but since I don’t do FB, that hasn’t been very helpful. It’s nice to know DV file readers have quietly been here, and it’s always positive to know we’re getting along just fine without FB.

[jadeitedraggyv]

I’m at a stage of experience and gameplay that would benefit greatly from information such as this. My park is by no means optimised, nor do I want it to be, though I’d prefer to know about certain mechanics so that I can prepare myself.
There aren’t too many “new additions” that have really excited me of late, and I’d much prefer to know things like breeding odds so that I can make a determination of breeding vs event currency or gems.
I appreciate your time with this Quil. Thank you!

[crumble608]

great thread, please go on
and thankyou, Quil 😘

[Quil]

Jun 6, 2019 3:09:54 GMT -6 adlerist said:

If there are posts about unreleased content and/or features, it’d probably be best to keep them separate in a clearly titled 'SPOILERS!' thread so nobody accidentally clicks on them, especially since there’s no option for a spoiler cover. But it doesn’t appear there’s any intention to create these spoilt posts.

SPOILER: Click to show

Thank you Arkynos for showing me how to make spoilers!

[Quil]

Next up, lets look at the Rainbow dragon. This is where things can get really crazy with the large number of possible results.

The file definitions for this guy:
Hint: min number of elements is 4
Breeding Chance: 1%
Cloning Normal x2 Rainbow: 25%
Cloning Rift x2 Rainbow: 10%
Cloning Social x2 Rainbow: 5%
Cloning Normal or Rift with 1 Rainbow: 0.5%
Cloning Social with 1 Rainbow: 1%

Keep in mind when you're trying to clone a Rainbow, you're also introducing 4+ elements which means you can also breed it without cloning, adding 1 percent to the total.

Rainbow adds so many dragons to the result set (as many Epic dragons do), can we truely trust the numbers in the file? Think about it, there are roughly 60 golden hybrids available in the result set, 10 opposites, and 25 epics, not to mention the 15 or so rift only dragons when bred in the rift. That's roughly 100 rare dragons, each with their breeding percent chance set in the data. If the goldens average about 4-5 percent chance each, then you're looking at 250+ percent chance at them alone, which makes no sense! You can't have 60 dragons each at 5 percent, something has to give.

This is where our Rainbow + Double Rainbow testing has come into play. I wanted to see what happens to these percentages when the sums of all the dragons add up to a large number (440+ percent). To fit a range from 0 to 430 into a 0-100 percent range, you'd need to multiple it by 100/440, or .227. So that a dragon that was at 5 percent ends up at around 1.1 percent. Now the new scale works, but rare dragons are significantly rarer than their number in the file would imply.

And our tests match this result. You may be asking how I arrived at the 1.1 percent chance without having massive numbers of results. We have less then 1000 results, and we might have only 5 to 10 of each specific dragon. There's really not enough data when looking at one dragon at a time, some at 5 percent have 10 results while others have 4. This is where knowing which dragons are supposed to be 5 percent works in our favor. I can group all the 5 percent dragons together in one result set, and do the same for the 4, 3, 2, 2.5, etc. That gives me much larger result sets, and I can see as a grouping that the 5 percent dragons dropped to 1.1 and the math adds up perfectly with that scaling.

Now what about cloning and where does that fit in with the numbers? Do cloning numbers also get scaled back when you have more results? These are questions I also tried to answer with some testing. Using two rift rainbows, the above says we should see Rainbow 10 percent of the time from cloning, and 1 percent for normal breeding. But the 1 percent is actually lowered to 0.22 percent with the scaling. If the 10 percent is scaled back too, then we'd expect to get rainbows about 2.5 percent of the time. If it's not, then we'd get them a little over 10 percent of the time. The tests are currently showing about 11 percent, with a small sampling of 100 or so. That's enough to deduce that cloning is not impacted by the number of results and the cloning chance is static. It likely does a clone check first, and if it fails, runs against the full list of the odds (at the 22 percent reduction).

So if you're trying for that Fever dragon we talked about earlier, and you use two rainbows, then your odds of getting one is only 0.8 percent instead of 3.5. Make sure you stick to a result set where the sum of the numbers is under 100, and you'll keep your 3.5% odds.

[Arkynos]

Jun 5, 2019 19:10:10 GMT -6 Quil said:

<snip>

The best place to start is likely a discussion of what kinds of things are in this file, and how are we going to use that data. As mentioned earlier, this is a huge file with roughly half a million values in it. You can try to look at it manually, but it's a lot to take in and not formatted.

I've done computer software development and database work for decades, and I jumped at the chance to help parse this thing. Other mods here have certainly read through the file and have a good understanding of it, but I wanted to do the next step. <snip>

hi yes this sounds like me

Before Quil was modded, I had found myself in charge of file duty almost by default due to inactivity of any of the previous mods who had the technical expertise to read it. Now, I am by no means an expert myself, but a passing knowledge of html (from being part of the Neopets generation ) on top of a term of Matlab at university let me get enough of the gist of what was going on to be moderately useful for things like event thread preparation. So here's a couple of my thoughts:

- I have "read" the whole thing. Firefox has a pretty-print option that makes it more readable: when you figure out that a chunk of code relates to, say, creating the menu that lets you select a questing dragon, I decided you're probably not missing too much by skipping to the next block. It still took hours.

- For a fair while I looked for changes by CTRL-F-ing certain terms in each new build of the game and checking for differences in the number of found results. Searching through 500 odd dragon names to find the one new one is not the most fun I've ever had, but I was happy to make myself useful by doing so.

- There's a lot in there that is just unnecessary to know. Even if you just narrow it down to dragon data alone, there's things such as animation timings, model scaling, and orb positioning that while vital to making the dragon work as intended I doubt anyone is interested in the numbers for.

- As Quil mentioned, we're still not going to talk about in-game things that haven't been released. The recent example of the Easter Island upgrade that appeared in the wish menu briefly is an example of why. People reacted to that slip up, then saved event currency for something that never appeared in the live game, and we don't want to cause other things like that to happen.

- Another more fun example: a dragon name appeared in the file in September that I was presuming would be a release in one of the autumn events. It never progressed further than just a name, and said name was removed sometime in the last few weeks. I always held out hope for it to show up later, but alas.The name is in the underlines back there. RIP.

- It will presumably be Quil who takes the lead in revealing things here (as the raw power of the SCIENCE he is doing with this is well beyond my abilities) but I might drop in for an occasional clarification or random bit of trivia.

[crumble608]

quote from Quil

Next up, lets look at the Rainbow dragon. This is where things can get really crazy with the large number of possible results.

The file definitions for this guy:
Hint: min number of elements is 4
Breeding Chance: 1%
Cloning Normal x2 Rainbow: 25%
Cloning Rift x2 Rainbow: 10%
Cloning Social x2 Rainbow: 5%
Cloning Normal or Rift with 1 Rainbow: 0.5%
Cloning Social with 1 Rainbow: 1%

Wow, figures like these are really helpful for deciding where to breed for which dragon.
but I have a question:
back in time when the epic breeding island was introduced, BFS told us it would give a better chance for rare dragons.
So, shouldn't there be different percentages between normal cave and breeding island ?
Or do these better odds only stick to breeding odds and not to cloning ?

[nagini]

This is very interesting, but will take a lot of time to read carefully and absorb. But I am so glad you are making it available, thank you! I am especially curious about how some dragons have different odds definitions for different situations and others don't. I wonder if it can be correlated to when they were introduced, for example the earlier dragons having different odds in the original and second breeding caves, but no difference between the caves for the newer dragons. (This would suggest that a later team of game developers was not following through on the ideas originally implemented by a previous team, but they were just left in the older dragons' code; this would be good to know, helpful to understand what gameplay "rules" became outmoded at what point).

[Quil]

Jun 7, 2019 2:44:28 GMT -6 crumble608 said:

Wow, figures like these are really helpful for deciding where to breed for which dragon.
but I have a question:
back in time when the epic breeding island was introduced, BFS told us it would give a better chance for rare dragons.
So, shouldn't there be different percentages between normal cave and breeding island ?
Or do these better odds only stick to breeding odds and not to cloning ?

The numbers there are for cloning chances. It's very likely that there is the same cloning chance in the Epic and Normal caves and they share the percentages. What the Epic island does is in increase the breeding chance of rare and epic dragons. How much it increases them is not in the file, at least not something I've found yet, and it's likely built into the application code itself and not the data file. The breeding chance on rainbow is 1 percent, and it does not define a cave.

It's also likely the epic island adds a small number, like 1 percent more, to each rare result. Or it could be 5, or 1/2 or even just 1/4 a percent. We'd need to test it to be sure, and testing outside the rift is difficult. I only have half a dozen people helping with numbers, and you can easily pull 4 to 10 breeds a day in the rift. But it's difficult to do more than 1 breed a day in the epic island on secondary accounts, so unless we get a huge army of testers then we may not ever know.

Also think of it in another way. Rainbow + Rainbow produces rares 98% of the time. If all the results are given a bonus, then they'd all have to be scaled proportionately and the numbers would remain the same. We'd probably need to test it with a cold+lightning, which has a small result set, and compare Moon results to the expected results. If it's only giving it a 1 percent or less bonus, we'd need 1000+ results to really start capturing the amounts. If it's bigger, like 5 or 10 (which I highly doubt), it would show up more quickly.

[Quil]

Jun 7, 2019 9:23:09 GMT -6 nagini said:

This is very interesting, but will take a lot of time to read carefully and absorb. But I am so glad you are making it available, thank you! I am especially curious about how some dragons have different odds definitions for different situations and others don't. I wonder if it can be correlated to when they were introduced, for example the earlier dragons having different odds in the original and second breeding caves, but no difference between the caves for the newer dragons. (This would suggest that a later team of game developers was not following through on the ideas originally implemented by a previous team, but they were just left in the older dragons' code; this would be good to know, helpful to understand what gameplay "rules" became outmoded at what point).

They've gone back and updated the odds through the years. In fact in the earliest files there were no odds on anything. The file from release time looks very different, back when rainbow was the only rare, but odds were not added until much, much later. Even when gemstones were first introduced, odds didn't exist in the file. My guess is that back then the odds were set based on the rarity of the dragon.

When odds were first introduced for rainbow, it was after the social cave was released and match the numbers from today (1% and 5%). They only added them for the social cave, because if you recall back then two dragons of the same type always produced that dragon in your own caves, and there was no way to clone from a single parent.

Despite it feeling like dragons are rarer now, the numbers for breeding haven't really changed in a very long time for existing dragons. There are just a lot more results, and as we've seen above if the results add up to over 100 percent than it scales down. So dragons with huge result sets can actually be rarer than they were originally. But gemstone dragons? While I can't see how common they were in the really old days, they've still sat at 3 percent for many years with result sets small enough to not impact the number.

[Quil]

I mentioned before that some dragons have multiple breeding chances in the file. I would appreciate feedback from everyone if they have any ideas or experiences interpreting what these different numbers may mean.

First is the Air dragon, fire + water. It has two breeding numbers, where chance one is 100 percent. Chance two is 33 percent. I tried water and fire in one of my caves, and I did not get an air so did not land in chance one. 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? Or maybe it used to be 100 percent, now it's 33, and the latest number in the file wins.

One dragon has 3 numbers, it's the only one with 3. Motley has 4 percent, 1 percent, and 0.5 percent listed in that order. It feels like Motley is so common these days during BeB, but then again it can happen with any breeding combo so maybe it's still 0.5 percent and just feels worse.

Most of the other dragons with 2 numbers are more difficult to test, as they are rare and the numbers are close. Blue Moon: 4 and 3. Celtic 3 and 2. Chrysalis 6 and 4. And maybe a dozen others. Holly is at 12 and 3, the next biggest range.

If we can figure out air, maybe the rest will make sense.

[adlerist]

Jun 6, 2019 13:12:04 GMT -6 Quil said:

Jun 6, 2019 3:09:54 GMT -6 adlerist said:

If there are posts about unreleased content and/or features, it’d probably be best to keep them separate in a clearly titled 'SPOILERS!' thread so nobody accidentally clicks on them, especially since there’s no option for a spoiler cover. But it doesn’t appear there’s any intention to create these spoilt posts.

SPOILER: Click to show

Thank you Arkynos for showing me how to make spoilers!

So there is intent to post spoilers after all?!


For anyone who wants to know how to create a spoiler cover on your post:



Now go forth and spoil, spoil like the wind.

[adlerist]

Jun 7, 2019 10:50:39 GMT -6 Quil said:

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.

[adlerist]

Jun 7, 2019 2:44:28 GMT -6 crumble608 said:

back in time when the epic breeding island was introduced, BFS told us it would give a better chance for rare dragons.
So, shouldn't there be different percentages between normal cave and breeding island ?

Jun 7, 2019 10:12:04 GMT -6 Quil said:

What the Epic island does is in increase the breeding chance of rare and epic dragons. How much it increases them is not in the file, at least not something I've found yet, and it's likely built into the application code itself and not the data file.

According to the Epic Breeding Island’s in-game description, 'rarer hybrids might come about more frequently' on the island than the breeding cave. I tracked results to test the Epic versus Enchanted locations while breeding 82 Bogberrys for the Gargantuan Island. Basically, the odds are ~5% higher to breed rare hybrids on the Epic Breeding Island for Bogberry + Bogberry. However, this doesn’t necessarily apply to epics. There was a ~2% advantage of breeding an epic on the island with the Bogberry combo.

It’s interesting to note that the parent dragons affect the Epic Island’s breeding bonus. I’ve bred 89 (and counting) Portent dragons with Portent + Portent on the island and in the cave. Since Portent holds all the elements, there’s no clear advantage using the island as every offspring is a rare hybrid in both breeding locations per the Rainbow + Double Rainbow discovery. As far as breeding epics (all epics, not just Portent), there has been no noticeable benefits (in fact there’s a .25% advantage using the cave, but this will fluctuate with more breeding attempts). A lot of players are under the assumption that there’s a higher chance of breeding epics on the island, probably because it’s called the Epic Breeding Island. But it seems fair to say the breeding odds advantage on the island benefits rare hybrids, and maybe will help your epic chances.

I took the Bogberry opportunity to test dragon levels as well to determine any breeding benefits at levels 11, 16, 18, and 20. But I’ll save that for another post, lest this one becomes too long (or even longer).