With the new cave checkboxes, I can see the odds are getting much more accurate. It looks like you're now including cloning odds in each of the caves for the results, which is another important reason to have those selections and it's great to see them added.
I noticed one issue that is likely causing some percentage discrepancies - When a dragon can be cloned from a combination, you're returning the cloning odds on those two clones and calculating the percentages on all of the other results. It's missing the percentages added to the cloning chance for those being cloned.
this whole system is pretty mental and you are a genius for having deduced your way through it. i've been following some of your notes from years ago.
Example: Abyss + Trench in the Upgraded Cave. It's showing 1% for Abyss (the cloning chance). 1% for Trench (cloning chance). 3% for Dapplethorn (correct). And 95% for Ooze. But Trench should be 1% for cloning + 5% for breeding chances = 6% total. And then Abyss is missing its breeding chance, which is split evenly between Ooze and Abyss as the two possible Hybrids. They would split the 90% remaining at 45 each, giving Ooze 45% and Abyss 45+1 = 46%.
you are correct, its currently missing this component. that's a whole 'nother refactoring.
which i'll have to work in to rewriting everything from scratch. these recent changes have pushed the whole system above some threshold of (probably system memory) which is causing the time-to-update to absolutely explode. like, from dozens of minutes (grudgingly acceptable) to a few hours. i'm going to have to figure out a completely new approach. have some ideas.
but this is the sort of thing which killed robotron's version years ago.
I also have one other suggestion. When you have some dragons with really low odds, like with Rainbow + Rainbow, rather than show 0.0% as a result perhaps have it say 'Extremely low' (or something similar). It seems there are a lot of 0 chances, which may confuse users, when in reality it's closer to 0.01% and off the scale, rounding to 0. There are so many at the 0 range that you'll get a lot of them, just not the one you may be specifically targeting.
that's a really good idea. maybe i'll floor the value at 0.1% or something, to keep everything consistently percentage.
Oooh! A new Social Breeding checkbox to resolve the Galaxy cloning issue in the Rift and Vale. Absolutely brilliant, drow !
When I suggested changes to the Parent Finder results for the galaxy dragons in the Rift, I didn’t mean to invoke so much extra work on you. But the implementation is terrific! I love how, in the Rift breeding results, if the Social Breeding box is checked, it reverts to the Vale breeding results — chef’s kiss! This works perfectly for the galaxy dragons.
The only thing I noticed were the galaxy exceptions. Lunis and Solis are not showing in their possible breeding combos.
And if Lunis is forced, it shows it cannot be bred.
'Lunis + any dragon' can result in Lunis since it fulfils its own breeding combo. And 'Lunis + Lunis' can also result in Lunis.
However, the breeding/cloning of Lunis and Solis has been complicated since they cannot be cloned in the Rift, only bred. What if, Lunis (and Solis) show in their Rift breeding combos when 'if I don’t have one to breed with' is unchecked. And if the box is checked, Lunis (Solis) doesn’t show. This would be kind of similar to the combo results for other dragons but have a different meaning with the galaxy co-op cloning rule. I know this is bending the way the sandbox works since a checked 'if I don’t have one to breed with' refers to breeding, and unchecked refers to cloning. I’m suggesting the reverse for Lunis and Solis, but I think it makes sense. Oddly enough.
The other exception, Rigel, is showing that it’s not breedable.
But those three dragons are the only difficult ones. The other 800+ work terrifically with the new Social Breeding option.
Super superb work, Drow! You know how much I love checkboxes. This just made my year. 💕
All hail, Drow the Genius!
Is there any chance this could be implement? The ingenious solution Adlerist mentioned sounds like it’d work. It will be confusing for players if Lunis and Solis are not in their breeding combos. We’ll start seeing complaints on Reddit and Discord about players getting Lunis and Solis, and that the sandbox is wrong. They will definitely ask if/why Rigel isn’t breedable.
Thank you for working on this, Drow! I know a lot of players depend on the sandbox.
Is there any chance this could be implement? The ingenious solution Adlerist mentioned sounds like it’d work. It will be confusing for players if Lunis and Solis are not in their breeding combos. We’ll start seeing complaints on Reddit and Discord about players getting Lunis and Solis, and that the sandbox is wrong. They will definitely ask if/why Rigel isn’t breedable.
if the breeding engine is properly implemented, then all these odd little edge cases should resolve themselves based solely on the breeding data.
i've just spent several hours rewriting the sandbox calculator into new hotness. it seems to handle lunis and solis correctly now, and odds should be closer to correct. but i'd sure appreciate other eyes on it.
again, this is just the sandbox, <https://dvbox.bin.sh/> the parent finder is still old and busted.
if the sandbox looks good, then i can translate the new typescript into perl, and start rewriting the backend engine which handles the parent finder.
Is there any chance this could be implement? The ingenious solution Adlerist mentioned sounds like it’d work. It will be confusing for players if Lunis and Solis are not in their breeding combos. We’ll start seeing complaints on Reddit and Discord about players getting Lunis and Solis, and that the sandbox is wrong. They will definitely ask if/why Rigel isn’t breedable.
if the breeding engine is properly implemented, then all these odd little edge cases should resolve themselves based solely on the breeding data.
i've just spent several hours rewriting the sandbox calculator into new hotness. it seems to handle lunis and solis correctly now, and odds should be closer to correct. but i'd sure appreciate other eyes on it.
again, this is just the sandbox, <https://dvbox.bin.sh/> the parent finder is still old and busted.
if the sandbox looks good, then i can translate the new typescript into perl, and start rewriting the backend engine which handles the parent finder.
The same issue from Parent Finder is arising in the sandbox. When the Rift Cave is selected, Lunis does not appear in the results for 'Lunis + Lunis' and 'Lunis + galaxy element'. Same for Solis. Rigel is not showing in the results for 'any galaxy dragon + any galaxy dragon'; it appears to be unbreedable.
The omission of Lunis and Solis may be due to the galaxy co-op cloning rule. But they can be bred in the Rift, so they should be in the results. I’m not sure why Rigel is unbreedable. Maybe because it requires owning all the galaxy dragons, and that requirement hasn’t been added.
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!
i guess i don't see how 'Lunis + Lunis' or 'Lunis + galaxy element' matches its breeding hint, 'Moon + any galaxy except Lunis or Solis'? unless this is a known exception handled in the DV code?
Rigel must be handled special in the DV code, since its all zeroes in the data, i'll have to code in an exception.
I don't know if the code in-game suggests otherwise, but the hint shown in-game definitely doesn't have the "except Solis and Lunis" exception. Any galaxy at all can fulfil that part.
if you click on the Info button in the hint, its there.
I don't know if the code in-game suggests otherwise, but the hint shown in-game definitely doesn't have the "except Solis and Lunis" exception. Any galaxy at all can fulfil that part.
if you click on the Info button in the hint, its there.
Ack! I apologise. I must have been reading quickly and missed that line! I'll delete my comment 😅
It's looking much closer, really great to see that. You've made quite a lot of progress!
I've been rewriting the Compendium simulator in scripts over the past few days so it can be used in a parent finder and called from outside the sheet. I've been just looking at all the weird cases and comparing stuff between the new and old. Checking some of these against the new Sandbox I found one oddity, and the rest seem very close:
Fire Rift + Dargon in Rift, or Fire + Dargon outside of Rift: This is a case where you have a single element in a breeding combination. This works very much like opposite dragons, where there are no normal hybrids in the result. Opposites have only golden hybrids with fixed chances each. Opposites solve the issue by adding the two primary dragon elements and they'll have the dynamic odds just like hybrids. Single elements work the same way, with just the single element dragon being added and taking all of the dynamic result odds. Fire Rift + Dargon in the rift will mostly produce "Fire" dragons. Fire + Dargon in the normal cave will mostly produce Fire too, and Dargon will be at its cloning odds instead of adding any additional breeding chances. No dragons with fixed odds will ever take additional 'leftovers' from a combination. They'll either be at those odds or lowered if too many results occur.
i guess i don't see how 'Lunis + Lunis' or 'Lunis + galaxy element' matches its breeding hint, 'Moon + any galaxy except Lunis or Solis'? unless this is a known exception handled in the DV code?
Rigel must be handled special in the DV code, since its all zeroes in the data, i'll have to code in an exception.
Apologies, since Lunis was not showing in any breeding combos in the Parent Finder, I got it in my head that it required the moon element, not dragon (slowly losing my mind). 'Lunis + Lunis' and 'Lunis + galaxy element' will not result in Lunis.
However, 'Lunis + Moon' will result in Lunis. Because Lunis has the galaxy element, it fulfils its breeding requirement. In the breeding hint description, it mentions 'except Lunis and Solis' because it’s referring to cloning. Yeah, I know it doesn’t read that way. When the Lunis and Solis debacle occurred in 2021 with the onset of the RBC bug it caused, the reps went round and round with Deca trying to keep Lunis and Solis breedable in the Rift at the wish of many players and trying to get the galaxy cloning rules squared away. At the time, we were told that the DV Team came to a solution that allowed Lunis and Solis to be breedable everywhere (except the RBCs) but only clonable at the co-op. This led to the 'except Lunis and Solis'.
I wrote the in-game Lunis, Solis, and Solarflare breeding/cloning descriptions/hints while we were in discussions. I asked Deca to remove the 'except Lunis and Solis' if the dragon could be in the outcomes as a breeding result, not cloning. It was never removed.
Lunis (and Solis) are correct in the sandbox with the results for 'Lunis + Lunis' and 'Lunis + galaxy element'. It’s only 'Lunis + Moon' that should have Lunis in the outcomes for all breeding locations, including the Rift Cave. And the same for Solis + Sun.
Based on what players have reported on this thread, Rigel and Tranquil are showing as unavailable in the sandbox. Since it won’t be possible to check every dragon, there may need to be a comb-through (not to be confused with a comb-over) to see which dragons are appearing as unbreedable. But as I said yesterday, it looks chef’s kiss.
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!
Is there any chance this could be implement? The ingenious solution Adlerist mentioned sounds like it’d work. It will be confusing for players if Lunis and Solis are not in their breeding combos. We’ll start seeing complaints on Reddit and Discord about players getting Lunis and Solis, and that the sandbox is wrong. They will definitely ask if/why Rigel isn’t breedable.
if the breeding engine is properly implemented, then all these odd little edge cases should resolve themselves based solely on the breeding data.
i've just spent several hours rewriting the sandbox calculator into new hotness. it seems to handle lunis and solis correctly now, and odds should be closer to correct. but i'd sure appreciate other eyes on it.
again, this is just the sandbox, <https://dvbox.bin.sh/> the parent finder is still old and busted.
if the sandbox looks good, then i can translate the new typescript into perl, and start rewriting the backend engine which handles the parent finder.
Is the sandbox going to have accurate percentages? Or fairly accurate percentages? Is that the goal? I know players have wanted this, but you always seemed so skeptical about it happening.
Fire Rift + Dargon in Rift, or Fire + Dargon outside of Rift: ... Fire Rift + Dargon in the rift will mostly produce "Fire" dragons. Fire + Dargon in the normal cave will mostly produce Fire too, and Dargon will be at its cloning odds instead of adding any additional breeding chances.
oooh, thanks for this. resolved.
after a week of some very late nights, everything has been re-engineered/rewritten/rebuilt and i think i'm done breaking things.
breeding odds should be much closer to reality. the social cave checkbox does things. updates to the parent finder data repo should happen reasonably quickly (its under an hour!) and no longer demands destroying the old repo while its running. result sorting works properly (floating point comparison stinks, kids).
if you notice anything else awry, please assume i don't know about it yet. thanks!
and some trivia! the dragonvale sandbox has been online for a bit over 7 years now (yikes). in that time, the game has gone from 360 dragons to 835, and the number of breeding pairs from 54,941 to 349,030. the sandbox code has gone from 2759 lines to 11,538 lines. the backend repository from 440 MB to 52,133 MB.
Fire Rift + Dargon in Rift, or Fire + Dargon outside of Rift: ... Fire Rift + Dargon in the rift will mostly produce "Fire" dragons. Fire + Dargon in the normal cave will mostly produce Fire too, and Dargon will be at its cloning odds instead of adding any additional breeding chances.
oooh, thanks for this. resolved.
after a week of some very late nights, everything has been re-engineered/rewritten/rebuilt and i think i'm done breaking things.
breeding odds should be much closer to reality. the social cave checkbox does things. updates to the parent finder data repo should happen reasonably quickly (its under an hour!) and no longer demands destroying the old repo while its running. result sorting works properly (floating point comparison stinks, kids).
if you notice anything else awry, please assume i don't know about it yet. thanks!
Thank you so much for all your hard work. Already a wonderful information source and your improvements sound fantastic.
Fire Rift + Dargon in Rift, or Fire + Dargon outside of Rift: ... Fire Rift + Dargon in the rift will mostly produce "Fire" dragons. Fire + Dargon in the normal cave will mostly produce Fire too, and Dargon will be at its cloning odds instead of adding any additional breeding chances.
oooh, thanks for this. resolved.
after a week of some very late nights, everything has been re-engineered/rewritten/rebuilt and i think i'm done breaking things.
breeding odds should be much closer to reality. the social cave checkbox does things. updates to the parent finder data repo should happen reasonably quickly (its under an hour!) and no longer demands destroying the old repo while its running. result sorting works properly (floating point comparison stinks, kids).
if you notice anything else awry, please assume i don't know about it yet. thanks!
Thank you greatly for all your dedication, Drow!
You are cherished among DV players everywhere.
Honestly, don’t know what we would do without Drow and Quil. We’d just be breeding dragons aimlessly in the dark. Many cheers!
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!
and some trivia! the dragonvale sandbox has been online for a bit over 7 years now (yikes). in that time, the game has gone from 360 dragons to 835, and the number of breeding pairs from 54,941 to 349,030. the sandbox code has gone from 2759 lines to 11,538 lines. the backend repository from 440 MB to 52,133 MB.
Sounds absolutely delicious!
(We are growing old with this game.)
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!
Fire Rift + Dargon in Rift, or Fire + Dargon outside of Rift: ... Fire Rift + Dargon in the rift will mostly produce "Fire" dragons. Fire + Dargon in the normal cave will mostly produce Fire too, and Dargon will be at its cloning odds instead of adding any additional breeding chances.
oooh, thanks for this. resolved.
after a week of some very late nights, everything has been re-engineered/rewritten/rebuilt and i think i'm done breaking things.
breeding odds should be much closer to reality. the social cave checkbox does things. updates to the parent finder data repo should happen reasonably quickly (its under an hour!) and no longer demands destroying the old repo while its running. result sorting works properly (floating point comparison stinks, kids).
if you notice anything else awry, please assume i don't know about it yet. thanks!
That is AMAZING Drow! I hope you catch up on your sleep. TYSM!!
Fire Rift + Dargon in Rift, or Fire + Dargon outside of Rift: ... Fire Rift + Dargon in the rift will mostly produce "Fire" dragons. Fire + Dargon in the normal cave will mostly produce Fire too, and Dargon will be at its cloning odds instead of adding any additional breeding chances.
oooh, thanks for this. resolved.
after a week of some very late nights, everything has been re-engineered/rewritten/rebuilt and i think i'm done breaking things.
breeding odds should be much closer to reality. the social cave checkbox does things. updates to the parent finder data repo should happen reasonably quickly (its under an hour!) and no longer demands destroying the old repo while its running. result sorting works properly (floating point comparison stinks, kids).
if you notice anything else awry, please assume i don't know about it yet. thanks!
Wow, this is a huge surprise! The percentages are slightly different from the Compendium, but they’re very close. Players should definitely get a good idea of their breeding chances and which combos are best. Thanks for doing this! I used the old sandbox and have been using your version since you started it. It’s been really amazing watching it evolve!
Wow, this is a huge surprise! The percentages are slightly different from the Compendium, but they’re very close. Players should definitely get a good idea of their breeding chances and which combos are best. Thanks for doing this! I used the old sandbox and have been using your version since you started it. It’s been really amazing watching it evolve!
There are always going to be some differences given that we have different criteria and options between the two. For example, the Sandbox doesn't have day or night settings, so Moon and Sun will both show as results for cold and lightning breeds at the same time, and that impacts the other numbers. Same with Weather and such. There are also dragons that have different odds based on how many copies you own, so without linking it to a checklist or settings that let you choose having them or not, the odds will also change in the results. It's the little things with tiny differences. But overall, you can be confident in the numbers given. They've always been rough guesses with the Compendium or the Sandbox. We don't know impacts of dragon levels, extra bonuses in the Epic cave, or other little things that could possibly change things by a little, so we work with what we've got, and they both should now give you a good idea of what your chances are with any given breed.
Somebody was asking for the DC rates today — it’s all in the timing.
I love how the generators are included! Maybe in the future, we can select the number of generators. I have a lot of Snow Generators on one island. And if the element boosts are also added, that would be incredible for DC farming. Oooh!
Thank you for the income addition, and have a fantastic holiday!
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!
'... 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.
and you are a genius for having deduced your way through it. i've been following some of your notes from years ago.
