Token Previewing

Vladdy said:

I doubt that it would be extremely "expensive" to deal with the previewing issue.

Let's assume the goal is not to completely fix it but to improve the situation and making it more fair. As @fmftint said the pull sequence in terms of 4* or 5* pulls won't change per store. But different stores have different sequences that provide the opportunity to optimise your overall 5* pull rate by managing you resource expenditure (if you do preview).

Let's think about that they would use one sequence for the classic legends store (and any possible future store with a 20 CP per pull cost) and another sequence for latest legends and any other store with a 25 CP per pull cost.

That way you should not be able to improve your 5* pull rate by previewing special stores (just the pulled 5* characters are different but not the amount).

Maybe I am missing something but that could be an "easy enough" improvement I think.

Well....let's consider this for a minute.

Currently each store as it enters the game generates some string of results for each user at random.  It's locked in as soon as the store is - I assume - in the code which is downloaded to your device and you next open the app (probably even before it appears on your screen, since that happens instantly).  Which seems like a fairly straightforward thing to code for.  It's done, it works, they've got it down to the point where it is probably extremely routine.

What you are describing is to have a new store somehow interact with the existing locked in Latest pull string - something that the game does not have anywhere else between stores.  And to make the results vary (in terms of pull results) depending on which storefront you clicked on.  So the same string in Latest might give you a BRB next while pulling in the Sinister CP store gives you a Sinister, let's say.  Oh, and to make sure that the new store accurately advanced along the results chain and interacted with pulls in Latests and any other CP store (s) (have they ever had 2 special CP stores open at once?  I think they did once, anyway).  Done wrong and you could see how new exploits/bugs could arise (chain doesn't advance, you get 2 5s instead of 1, etc).  

This seems like a system that would require a lot more coding (as each new store would have to be set up correctly) and QA (to make sure that the new store did indeed give you a Sinister instead of a BRB because oops!). It might not even be possible to build a store that gave you different results from that chain, just because you made the pull on a different screen with the existing game code and engine.  

Hence, expense is added to every single new CP store.  (Not to mention that to verify the store was accurately pulling from the string, you would need to preview the pair of stores, lulz.)

To me, it  seems it is not a low-cost endeavor to do this and might be far more expensive (in terms of ongoing time/money spent to ensure proper operation) than the current workflow they have around building new CP stores.

Vladdy said:

bluewolf said:

Well....let's consider this for a minute.

Thank you for taking the time and share your thoughts about my comment.

When 5* characters leave/enter latest legends the string remains the same. So I guess that they don't lock in the specific pulled character but more likely the index and the store knows what character currently occupies that index.

Anyway I am still not convinced that it is that bad as you seem to think but maybe I am too optimistic.

Yes, there are definitely problems that they would have to deal with if they want to change something about previewing. But I highly doubt that they do want. So I guess it will just stay with us.

All very true.  Previewing has been around and known for years.  If you think about it, they resolved the re-roll exploit with the locked-in string vs stopping players from previewing at all.   

This leads me to believe they consider it to be closer to an annoyance vs a drain on potential revenue.  That, or the resources/changes required to eliminate previewing was too big of a hurdle to take on (if even possible with the game as it was built - the engine is new but the base code is still 6 years old).

The fact is that the team is small (about 21 members according to the Credits) and QA is already a major hurdle just from a "test this new character and its interactions with 200 others" point of view.  It is clear that they are limited in their ability to take on side projects which is, from my perspective, where you would have to put the goal of "stop previewing".

I could be wrong about costs and such, not knowing their internal processes, but I think that if a fix introduces a new ongoing complexity and cost to their processes, it would not be easy for them to adopt.