I recently stumbled upon a mod that reminded me of the way cofl works, but I believe cofl could do a better job. So what i'm proposing is a new command that searches the auction house for a attribute and a level, or 2 attributes and respective levels. And then It would pop up 4 different categories for different pieces so - Helmet, Chesplate, Leggings and boots. Showing the cheapest available match to the command So e.g /attributeprice lifeline 4 manapool 4 and then all the cheapest helmets, chestplates, leggings and boots that match this will pop up in a list. Also a calculation that finds cheapest combinations of attributes to upgrade a piece. So let's say you have a veteran 4 crimson leggings, it will search the ah for the cheapest combination of kuudra leggings to upgrade that piece to x level. So it would find a certain attribute and what item it is that you want to upgrade, then you will tell it the starting level and the ending level and then it will do the calculations.
#Better attribute searching in-game
1 messages · Page 1 of 1 (latest)
i think u should just use the other mod
u can also js use neu search
thats an insane amt of work
I mean why would it be an insane amount of work?
making all that is a huge time sink
for a feature that u can do w a config..
it would take time away from fixes and updates
also discord bots
or neu search
That sounds like half the work of your volatile suggestion.
It's just executing a few searches like the website does or even faster the mod lbin data which is in RAM always
can cofl not access old median data atm?
the mod
does anti manip not store data to detect sudden price changes
There is no median store because there is/was no reason why somebody would be interested in an outdated median
Anti market manipulation keeps a few sale prices around to do it's job but only when calculating the median.
It then does nothing because it would take too long
someone im in vc with did this 5 mins ago
None of those say msnipe
all i know is the median was changed off a few purchases
if the volatility thing is too much work dw abt it
i js figured cofl had some way to store previous days median prices or something
or detect median price jumps
That would be very weird, you mathematically can't do it, especially not alone
3 people
but i guess
if theres no current way for cofl to detect when a median price changes juristically ig it would take a lot more time than i realized
I can store them, I just don't know if that's the best solution, like what timeframe is looked at and how long will it take to load the info back.
From the median calculation step I can calculate if the price recently dropped, but if it did you may just want to buy the item more and it doesn't help you
Which ones?
not sure at the moment
Can you link me to that?
one sec
the bigger problem is median prices skyrocketting in a short amount of time
up or down its still great to be able to see if a median price is moving by a lot in a short amount of time
They basically can't, you need 50% of last two week sales to change it irregularly
ive changed 5+ medians by over 40m in the last few days
its just meta items with low volume'
you can change the median by relisting and having someone purchase after a relist or 2
for some reason
on certain items
it seems like
that might not be the case
Anti market manipulation removes any duplicate buyer or sellers
balloon snake incident
yea
instasell changed like 10 times in an hour cause of volume
What it currently doesn't do is de duplicate item ids
but manipulation aside, there are countless things in the game that skyrocket prices and affect medians
even if they r selling massive amounts
amber polished drill engine
mining v3 was released on alpha and prices spiked like crazy
including median
Except manipulation they should all be valid reasons for the price to increase
That's why median adjusts downwards relatively quickly
but when prices r up median is still up
thats just a very volatile item'
all of that would be extremely helpful to see/know
is this actually a thing or are you just talking about in general btw
That graph is for longer than a day, your volatility suggestion would display low volatility again
that price change was in 1 day
The price change yeah, the median wouldn't mind after just a day tho
idk cause people made soooo much money off those
they remained high for a day or 2 then dropped
its not like they js spiked then crashed
like every other item affected by in game events/metas, price changed drastically on the first day or two, then remained high for another day or 2
then start coming down
what youre saying about median coming back down is completly right, but that is also the dictionary definition of volatile
price shoots up
stays up for a bit
comes back down
or the opposite
median shooting downwards could also cause people to sell items for incorrect prices if they correct themselves a day or 2 later
again if this is all too much of a hassle no worries at all
im just saying there is countless uses for it
not absolutely necessary maybe
but useful
Coming down is for when prices drop from their normal level, median can't shoot up.
Except currently when you manipulate it with 4 or more people, burning 12% of the manipulated amount in ah fees. You need to list an auction for the item to profit, requiring 5% loss on that
im not saying median shoots up when prices drop?
if ur talking about this i mean that when the median plummets due to the player base over reacting/pannicking
thats volatile
the price always corrects after a couple days
Is that an issue on items with more than 20 volume?
yes
this item went from 150m to 260m in a day
it then came down to 215m after 2 days
this happens very often
when ofa was nerfed all items went down like 50% and returned to like 10% lower than what they used to be all within a week
do you happen to have a screenshot of 260m because I can at most find 245m
Anti market manipulation has the last 2 days of sales for that item.
for the oldest 7 the median is 242m how about I add a volatility number based on the difference between the oldest median, long term median (all 80 sales) and the short term (about last 3) to compute the maximum amount of price swings between them.
That would be ~5% right now but higher when it recently swang or has lower volume
How did this get into market manipulation? Since it would be pulling the cheapest on the ah, the only way to "manipulate" it is to buy up all the cheapest shards / armor pieces
about 22% upwards for that spike
not talking abt urs
js not talking in the right channel lolz
mb
that sounds really good
the cheapest would then be another crimson armor pieces probably
the only thing thats needed is just like something in cofl lore or sum that says whether this item is experiencing volatility at the moment
just like a heads up to buyers or sellers
and config users could use it to filter flips that are going thru volatility without having to change config values
the number would be filterable and in /cofl lore, probably color it with green yellow and red at <5% <20% and >20%
If this gets interest I'd be down to spend like 10 cofl coins per search
Or if it's free with premium +
?
Wdym?
sounds like a reasonable thing to put it into prem+
wdym per search
actually lets get back to that
I mean since it would be using cofl servers it's more than fair cofl gets compensation
you want to search for the cheapest crimson armor pieces (all 4) by an attribute?
Do you want to have some kind of level select as well?
or just always the cheapest per level cost, one of lvl 1-5 probably
So there's 2 commands. One to find certain attributes or attribute combinations. And one to find the cheapest upgrade path of a certain attribute
Ig the trickiest part would be the upgrade path. But I think you'll just have to do some assumptions which could make the price a bit higher but probably be the most accurate either way.
the cheapest eg godroll? To my knowledge that would depend on the type of armor as you can only combine attribues but not change them
upgrade path as in a list of active auctions to buy to combine it with?
Yes
akwav idk if it matters but u might be able to make this a lot easier by looking configs that filter attributes used for god rolls and /cofl replay active
thats what i use to craft god rolls
making something similar to that
Like you search for example mana pool mana Regen. And then it goes through all different kuudra armor and shows the cheapest for each piece
yes
just as far as possible or limit it to lvl 7 or so?
no
You'll specify how high you want to go
you can never buy 7-10
taht takes a minute because it reloads all auctions, I can make that faster
they are 3-4x the price it would cost to combine
i see
i just meant it almost already exists js not automated
u could refine it is what i meant yk
look at 6-10
If you want an 8 it'll look up how to make a two 7s the cheapest way
ah, that way around so you only have one result for each type
"how to make"
the process of finding the items would be really easy
youd js have to be careful pricing
anti market manipulation already finds the cheapest attribute cost to cap lvl 7 prices
also a feature like that would REALLY mess with attribute prices
so they would change after this becomes a thing
But cofl would know that
the volatility would go up 🙃
lolz
cofl needs to be able to know if a t6 is more expensive than two t5s
idk if it alr does
it js needs to be able to
and the cost of merging attributes
I mean yeah more people would buy lower level attributes since it makes it easier for them to craft attributes themselves now
it does, as stated above anti market manipulation looks at that
lower level attributes r also not profit 🙃
it caps median at craft cost +10%
But people selling high level attributes would have to adjust
thats what caused the necron boots median being so low
I mean y does it matter?
low attributes arent prof cause price stabalizes at a certain price
i mean crafting wise
only like 3-5 (and sometimes 6) are profitable to craft with
I check every level available
I mean technically not profit. But you don't have to overspend for someone else to craft it for you
i mean profit compared to just buying t4/5
like the only way u save money crafting god rolls is w certain levels
Because in the upgrade path ig you'll have to look for something like (the price) / (2^n-1) where n is the attribute level
And that would figure out the price per attribute 1 shard
only 3-5 work
for saving money
ive crafted like 10 sets of 10 10
6 works if its an attribute that isnt used in multiple godrolls
Why wouldn't it work? Just going to be more expensive, but I'm not worried about that
Since it would be figuring out the cheapest way
Either way
its more expensive than just buying the high level itself.
is what i ean
mean
low levels bottom out at the same price
majority of the time
Is that true though. From when I flipped mp Mr aurora those high level attributes were a rip off
I could do the calculations and we'll see
cause no one wants to sell mana pool 1 for 100k
yk
mana pool 1 is always a good drop
it also heavily depends on time of day
and how many people r crafting
if more than 1 person is crafting 10 10 at a time u cant save money
Still not worried about that
@leaden river make this optional. It would break every filter at the moment
ya this would be cool to see how much it affects the nether armor market
all low level prices would go up and high level gr would come down
good and bad
so we are looking at /cofl cheapattrib mana_pool and /cofl cheapattrib mana_pool mana_regeneration outputting Crimson Terror etc.
How would the attribute path work? just multiple results? wouldn't it be the same?
what are you referring to now?
all attribute prices would change
What do you mean with attribute path?
you said you want to have two commands, the second for combining somehow
this
also
Oh yeah so it would give a list of. So buy these to get it from 1-2 and then buy these to get from 2-3
Etc
this fully depends on prices of god roll high level attribute armor btw
with cost sum next to it?
Yeah that sounds good
@leaden river gr kuudra armor is arguable the most volatile items in the game 😄
so it would have to be dynamic against the price of high level attribute combinations
im assuming thats very easy tho
Then maybe a little warning if the price is above a certain threshold of median
So I will assume you have a lvl 1
Lvl 2 - 400k
-> Auction lvl 1
Lvl 3 - 700k
-> Auction lvl 2
Lvl 4 - 1400k
-> Auction lvl 2
-> auction lvl 2
...
Yes
So I search for example /coflattributeupgrade crimson leggings veteran 1 6
So it would give me a list up to attribute level 6 of kuudra leggings
wouldn't you ommit the crimson since only crimson amror can be combined with each other?
well if thats the cheapest than it will just show that
All kuudra gear can be used to upgrade
which makes it a lot easier
I call all crimson isle armor crimson, so yes
but the attribute prices also vary from armor type
my b
there are also some other pieces which can have attributes
Yes but those aren't combinable
are they of interest?
With crimson, terror etc
which other not but with themselfs
But I mean if the system is in place it could be done with other armors too
But kuudra gear has by war the biggest market
Hence most interesting
and possible magma lord
Maybe rods would be after
and fishing rod attributes
Yeah
akwav the only problem i could see is that if several people start using this it just wont work anymore cause attribute combined prices and gr armor would flatten out at extrmely similar prices
cause atm there is almost 0 people crafting 10 10 armors
consistently
its only profitable right now because no one does it
Hence lock it behind a paywall and watch people buy cofl more
yes but like
you always have the risk that the auction isn't there anymore when you try to purchase it
5 people doing this ruins it
there is no single person consistently crafting 10 10 armors
or even 9-9 etc
But cofl tracks that. And could pop up that it's bought? Or how does it work for flips, does the client send out to other cofl users "hey I bought this auction, mark it as sold"
no it would js get escrowed id assume
Its just that for it to be predictable the order/words need to be consistent so
/cofl attributeupgrade [item type] [attribute] [startLevel] [EndLevel] would have to be the same, and at most one of the arguments can have a space otherwise I can't simply tell which is which
well yes but you chat output won't change after that
just if somebody else executes the command different auctions would be listed
Can't you just call mana pool - manapool
generally yes, currently its mana_pool within the system - thats how hyixel also stored it
and then allow spaces in the item name?
the server receives exactly what you enter
So like a intermediary which changes the command before it's send
Maybe not possible
Or just the servers translate manapool to mana_pool
Yes, rather the other way around, adding aliases, maybe also mp not sure if they overlap
Yeah, I mean since speed isn't a huge factor if it's not the best solution but still a solution it's fine
If you need some help with the math lmk
can volatility go negative - if the price is trending downward or is it an absolute number?
absolute id say
price action downwards or upwards quickly r both volitilty
seeing if an item is volitile upwards or downwards is super useful as well
well yes, but is the volatility metric - as in filter value negative
wdym sorry?
so if the price goes from 100 to 50 it would be -50% and from 100 to 150 its 33%
yes but -50% can js be read as 50%
volitility doesnt depend on direction travelled
js percentage travelled
in that case it travelled 50% downwards
not -50% downwards
So the volatiltiy filter would also trigger for -50 if you set it to >49 ?
volility=standard deviation * √x (where x equals amount of time volitlity tracker watches)
yes
but then you aren't able to filter for a price trend downwards explicitly
how would you select that
I am cooking up a number to be filtered by a filter
okay then
what time range would you expect?
does the amount of time affect speeds or storage
like is there drawbacks to using higher time frames
cause 2-3 days should be enough for 99% of items
possibly even 1 day depending on the item
how many sales
sales dont actually matter in this case
for the most part
considering if an item has super low volume and there is no stable price because of low volume, it is volitile technically
well I am looking at x amount of sales, taking the median of the first and last ~5 currently and use the difference of the two
the short term median, 3 or 5
5 would be perfect for items w higher volume but u could def do 3 for items w lower volume
if an item has 3 sales / day
it could take over a day to update volitility if price spikes
thats how the median works, the threshold to use 5 is 20 or so a day
using the last 3 would be great for low volume items
you asked for a volatility for median
The current median is the last 3 sales
what is the other median
long term median
or the oldest 3 sales
ah
one second
r u not calculating volatility off of the last 3 sales?
oh r u saying if the last 3 sales r different price then another 3 sales?
I am trying to figure out what the correct time frame is to compare the median to to determine the medians volatility
because you said you didn't like the median changing so much, or rather you want to have the volatility metric to see how much it changed
and I am wondering what that time frame is that you think it changed in
The displayed weighted median has 3 compositing numbers to get a stable accurate number
obv ur the expert so i def dont know 100% what im talking abt but could u compare the price of the last 3 sales to the average median before those 3 sales?
or does it not work like that