Lights And Logic

Will someone please explain the logic behind the following statements:

1)You must cap azure and pearl to get time extensions.
2)Occasionally getting X lights reduces chance to get Y chests, so don't build X light(while constantly building Y light).

For 1:
You need to get to maximum azure potency for time extensions. Azure light affects potency and drop rate(chance if mob drops chest, it will be blue). Pearl light affects overall drop rate (chance mob will drop any chest).
If you are capped on azure (and no pearl), there is nothing preventing you from getting time extensions, except luck and kill speed.
If you are capped on pearl (and no azure), then you won't get time extensions until you bring azure up to the point where extensions are an option.
If you are capped on pearl and blue, then the only thing between you and time extensions is luck and kill speed.

For 2:
Azure, ruby, amber lights affect potency and drop rate. Pearl affects drop rate. From what we have been told through developer interviews:
Pearl light determines the chest drop rate of a kill. Color lights affect the % that a chest, if it drops, is a particular color. The base values for each % is a non-zero value (this was not part of interviews, but can be seen due to getting occasional colors that are not built). Therefore, getting color lights increase the weight (%) that a particular color chest will drop, and decrease the chance that other colors will drop (down to a minimum, non-zero value).
So if you are constantly building (for example) azure and pearl (which constantly slants the % drop towards azure), what is the logic behind not building any ruby/amber light whatsoever? At the very least, you are working towards getting ebon lights, and improving the quality of your gold chests.

did... wait... what?

You asked the question and answered it at the same time? lol.

Well, for number 1:
You need to get to maximum azure potency for time extensions. Azure light affects potency and drop rate(chance if mob drops chest, it will be blue). Pearl light affects overall drop rate (chance mob will drop any chest).
If you are capped on azure (and no pearl), there is nothing preventing you from getting time extensions, except luck and kill speed.
If you are capped on pearl (and no azure), then you won't get time extensions until you bring azure up to the point where extensions are an option.
If you are capped on pearl and blue, then the only thing between you and time extensions is luck and kill speed.

And for number 2:
Azure, ruby, amber lights affect potency and drop rate. Pearl affects drop rate. From what we have been told through developer interviews:
Pearl light determines the chest drop rate of a kill. Color lights affect the % that a chest, if it drops, is a particular color. The base values for each % is a non-zero value (this was not part of interviews, but can be seen due to getting occasional colors that are not built). Therefore, getting color lights increase the weight (%) that a particular color chest will drop, and decrease the chance that other colors will drop (down to a minimum, non-zero value).
So if you are constantly building (for example) azure and pearl (which constantly slants the % drop towards azure), what is the logic behind not building any ruby/amber light whatsoever? At the very least, you are working towards getting ebon lights, and improving the quality of your gold chests.


Hope this helps you figure out them trixy lights!

I'm trying to figure out if my logic is skewed, or the people who I frequently end up with in Abyssea doing the "REDZOR R BDZ" "NEED BLUNPRL 4 TE" or letting red chests with amber, red, azure (and other colors) depop are idiots.

I would prefer to give them the benefit of the doubt, and hope that my understanding of lights is lacking, than make a post about how many people make me want to /facepalm.

I see what you did there...

So you were trying to avoid being an *** about it, but ended up being an *** about it anyway? Makes sense.

If being open to discussion while providing the reasoning behind the stance is being an ***, then yes, yes I am. However, our definitions of "being an ***" seem to differ. Should we start a thread? :D

Well I think a lot of people believe that getting Amber light (more Gold chest) or Ruby light (more Red chests) negatively affects your chance to get blue chests. I tend to think they same thing really. If you build Amber or Ruby then when a chest drops instead of it being blue there's a higher chance of it being something else.

I think of it like this...lets say you can only have 10 Fruits and you have 10 Oranges. In order to get some Apples in there, you have to remove Oranges. (lol silly analogy, but makes sense to me.)

It was found in the DATS that there is a "recording system" for lights. As you correctly described, there are dual functions at work when applying the effect of lights on chest. First is the magnitude of the chest. The second is the ratio of the chest. The "recording system" showed that the game stores up to 1000 lights. The significance of this is that it is assumed once you reach 1000 lights, older lights will start to drop off. This is significant because it means that building lights may not only mean you begin to equalize the proportion of lights (For example, if you want TE ONLY and you start building gold lights, the chances of getting a blue box might change B-R-G from 900-50-50 to 500-250-250), but you also actively erase previous lights.

This potential danger often emerges in AF3 PTs when the PT, having sufficiently built their time, begins to solely build ruby and amber lights. This is a potentially dangerous situation because you risk dropping the portion of blue lights to less than would be expected if it recorded an unlimited number of lights. For instance, if it were unlimited, we would expect that having 998 blue lights (Not 1000 since we expect that red and gold have at least 1 unit in the ratio since they can drop even if you don't build any of those lights) and building red would mean you have (998/1001) chance of getting a blue box. However, the addition of the 1000 limit and the assumed "drop-off" of lights means that you're blue box percent is actually (997/1000). This is initially insignificant, but it becomes increasingly important as the lights continue to build. Take, for example, a scenario where you gain 997 blue lights. If you follow this with 499 red lights, the ratio isn't 998-500-1, it's 499-500-1. Consequently, the further you build lights, the more rapidly you diminish the ratio.

So to summarize, up to 1000 total lights (997 kills, if 1kill = 1 light, which is probably not quite the case), the affect of building lights is cumulative; after that amount, the effect of building lights is sacrificial(adding one light removes another) in nature?

*edit* Which becomes even more fun, if its not a replacement effect, but as stated above, oldest lights getting replaced. If you did 333 blue 333 red 333 gold (in that order), you would end up with 332-334-334 (blue red gold)?
Also, is pearl light calculated in the 1000?

In short, that is the current theory. Support for this, as mentioned, is the statement from SE and the 1000 lights register. However, there are some potential assumptions we're making in adopting this theory. We assume that the older lights drop off once over 1000 lights have accumulated. It could be that there is a threshold that lights never fall beyond. For example, if we have 997-2-1 and the "oldest" light is red (2), we expect that gaining an azure light will bring the ratio to 998-1-1. However, it could be the case that lights do not "drop-off" until they exceed a certain threshold. Under this assumption (Different from our earlier assumption), we expect the ratio to remain stagnant. Contrarily, if we had 333-333-334, the threshold were 200, and the "oldest" light was an azure light, these two assumption would both agree that adding a ruby light would make the ratio 332-334-334. A modified version of the theory that would oppose the previously mentioned versions could be that the highest light drops off. If this were the case, then we would expect amber light to reduce; thus, leaving us with 333-334-333. We don't have enough information to assert that one of these assumptions is better than the other.

Edit: First, I would like to clarify that in my previous posts, I equated gold light to amber light. This was with respect to the color of the box and not the XP light. It's assumed that pearl, like gold (XP) and silver (Cruor), are not susceptible to the 1000 light limit. The reason for this is that they work on magnitude whereas the 1000 light limit works on ratio. I previously explained this, so just look at the first part of my post. If the magnitude of pearl is maxed, it will stay maxed (Just like other lights). When a box drops, there is no additional calculation for pearl light (No pearl chests). Similarly, chest drops do not need to go through a second based gold (XP) light. However, your question is still relevant since if it did include other lights, we could argue that getting nothing but 997 pearls in a row will reduce the ratio to 1-1-1. Experience tells us otherwise, so it is assumed that Non-Azure/Amber/Ruby lights do not affect the 1000 lights ratio.