Very low supply of HMP, Riftcinder and Riftdross

[Demonjustin]

although ffxidb drop information is not correct since everyone in the party is not posting results to the collective.

I'm curious, why do you think this leads to inaccuracies in FFXIDB's drop information?

[bungiefanNA]

I'm curious, why do you think this leads to inaccuracies in FFXIDB's drop information?

For one, everyone must be running Windower or Ashita with a plugin loaded that monitors the drops and submits them to the database. That means console users are automatically excluded from the statistics.

[detlef]

For one, everyone must be running Windower or Ashita with a plugin loaded that monitors the drops and submits them to the database. That means console users are automatically excluded from the statistics.

Are you suggesting that the game has different drop rates depending on your system? Or are you saying that console and pc users cap lights/use cells in a different way?

[Shirai]

I'm not sure about the rest of you, but I'd still take the word of an automated system collecting data from thousands of different instances over that of a mental list from an individual who'se statistics are based on them playing a few characters.

Unfortunately, the latter seems to have the biggest group of followers here...

[Byrth]

FFXIDB is inaccurate in two cases that I know of:
1) In a "layer zone" aka instance like Salvage or Delve. -- Packets are sent by the server multiple times because SE's server infrastructure is bottlenecked and they make it playable by being incredibly redundant in cases where they think the packet might have been dropped. This means that item packets can be sent twice, which results in higher than realistic drop rates. For instance, if you look up Khrysokhimaira Elder you will notice it has a 125% drop rate on Linen Coin Pouches. That's because the drop rate in that zone is overestimated by about 25% due to high traffic/lag. Orthrus Seether, on the other hand, is only listed as having a 105% drop rate because Bhaflau is less laggy/popular.

2) When determining the precise level of TH. -- FFXIDB fails to notice non-traditional forms of TH, like Bounty Shot, TH Trait on BLU, pets with TH, and Fellows with TH. This results in higher than realistic TH0 drop rates, but has no affect on higher tiers of TH. Also, it assumes that you automatically have the highest level TH in the party regardless whether they acted on the mob, so in that sense it biases drop rates downwards. Overall, FFXIDB underestimates the effect of TH.


As far as I know, Voidwatch drop rates should be near 100% accurate. XIDB has also collected data on alignment, but no one has bothered to parse it.

[bungiefanNA]

Are you suggesting that the game has different drop rates depending on your system? Or are you saying that console and pc users cap lights/use cells in a different way?

I meant that events with individual loot pools, with console users in the party, won't have their items reported. So the stats on a full party getting drops may be incorrect. Drop rates are calculated on the server, not your system, so of course consoles won't have different drop rates. However, FFXIDB doesn't talk directly to the server, it relies on us having a program to record the activity from the server, which can't be easily done on a console.

[Kensagaku]

The thing is that drop rates are not calculated by party; they're calculated by number of drops over X runs. Whether it's 18 people over ten runs or ten people over 18 runs, you've still got 180 different data points. It may take more runs if not everyone is calculating, but overall a trend will become visible. It doesn't matter if not everyone is recorded; it will not change the overall trend.

Example, say you have all 18 people in an alliance parsing over 18 fights. That's 324 points of data. Now let's say you have ten people doing it instead of 18; it'd take 32-33 fights to get the same amount of data, but you're still getting the same general trend of data over time. The more fights you do, and the more data you get, the more the pattern and exact drop rate becomes obvious. It doesn't matter how many people there are doing it, except to determine how long it takes to get enough of a sample size to prove a relative rate. And with how long VW has been out, likely the rate is pretty accurate by now.

So yeah, it doesn't matter if there are console users or users that aren't giving item drop information; the rate will eventually be determined simply due to having a large enough sample size, since droprates are the same across all players.

[Demonjustin]

Well it seems I'm late to the party. Regardless, I'll give my own explanation.

Roll a die 1,000,000 times and record the results of every roll. You should see each side come up roughly 16~17% of the time.
Roll a die 1,000,000 more times, this time record only the results of every third roll. You should still see each side come up around 16~17% of the time.

The reason for this is that probability isn't being affected by your lack of gathering results from those other two rolls. You're ignoring two thirds of the rolls but you're not excluding specific numbers or anything, they're entirely random, it's as though these rolls never took place. If you excluded specific numbers you'd be tampering with the data, but since it's entirely random data you're missing and each roll had an equal chance to produce the results you're going to get an accurate number all the same.

FFXIDB is missing information from some players, yes, but this doesn't prevent it from getting the overall picture. The only way it would is if the chance for the dice to land on a specific number was different for each character/console or if it specifically ignored certain information and thus, tampered with it's own data.

[Olor]

So this weekend I see a shout for morta (I think, anyway it was a cinder dropping NM), I join. It's me and the shouter. After about 15 mins of NOT ANOTHER SINGLE PERSON EXPRESSING INTEREST the shouter decided to change it to akvan. It filled up okay after that... but yeah cannot express enough how uninterested most people are in doing this content. This was on a Saturday afternoon (Pacific time) so should be a great time to scare up people for a shout.

People don't feel like gambling their time when cinder rates are so low and dross is worth so little. I don't blame them. I just joined cause I like VW.

[InpendingDeath]

Heavy Metal is trivial now that it requires T3 mats for reforged armor, only thing it takes is a couple weeks granted you have the gil to buy from people. Maybe requires you to make your own VW shouts when there are no shouts going on which everyone is capable of doing with little to no common sense.

Riftcinders/Dross are the real problem. I'd rather do 1500 plates twice than plates+cinders because no one sees value in doing these anymore.

SIMPLE FIX! Make Empy Armor reforge require Sealord Skin (Bismarck drop) or Belladona Sap (Morta drop).