Menu
What's New
By:
Filters

Guide to determine the DECAY of any item

MG Mighty said:
Very hard to say since mobs do variable damage so the tests can't be done the way I trully wished.

Anyway look at the following examples:
Knight Impact Protection 12 Damage Decay= 0,776 PEC
Goblin Impact Protection 12 Damage Decay= 0,782 PEC
Pixie Impact+Cut Protection 12 Damage Decay= 0,781 PEC

In them the damage was 12
Several durability, even different types of damage,
Still the variation between them isn't high enough for me to establish any relation besides that total damage might be the most important factor in determining the decay.
Click to expand...
I'm absolutely certain that total damage absorbed is what determines decay.

My question was related to how fast the protection reduces, this can be inferred from the reduction in decay, if we assume decay is a function of damage absorbed.

A reasonable hypothesis would be that armours with high durability would retain their protection longer, and would therefore maintain higher levels of decay longer than lower durability armours.

Please note this wouldn't make higher durability armours less desireable, as I have pointed out in another thread, unless you have an adjusted FAP or better, high protection is cheaper in overall decay (even though it costs more in armour decay), because FAP decay costs more, per point of damage absorbed.
 
Ironheart said:
I'm absolutely certain that total damage absorbed is what determines decay.

My question was related to how fast the protection reduces, this can be inferred from the reduction in decay, if we assume decay is a function of damage absorbed.

A reasonable hypothesis would be that armours with high durability would retain their protection longer, and would therefore maintain higher levels of decay longer than lower durability armours.

Please note this wouldn't make higher durability armours less desireable, as I have pointed out in another thread, unless you have an adjusted FAP or better, high protection is cheaper in overall decay (even though it costs more in armour decay), because FAP decay costs more, per point of damage absorbed.
Click to expand...

Didn't thought of that before like that.
I guess it makes sence.
 
Ironheart said:
I'm no statistician, but from the figures produced so far does there appear to be any link between decay reduction and durability? I would suspect this may be the hidden 'durability factor' that many have been searching for.
Click to expand...

I have a formular that computes the durability from the different damage types and it's value. It's a 6'th order unlinear forumlar. It's been sitting in my harddisk for a couple of days, as I still fail to fully understand it :duh: . It is'nt pretty !. But my formular yeals an error that is to small to be reported in %, but rather ppm (parts per million), so im in no doubt that it's correct :silly2:. Hopefully someone else can understand this nut-formular, and link it to the decay.

I'll post it once I get home, in a new thread

From this formular I can't see anything else that Adjusted Guardian truly is adjusted. I'ts the only one that does'nt follow this formular. If this armor was to follow this 'normal' formular it's durability should have been 947, but it's 2000.

btw. matlab rocks :kos:
 
Nutarix said:
I have a formular that computes the durability from the different damage types and it's value. It's a 6'th order unlinear forumlar. It's been sitting in my harddisk for a couple of days, as I still fail to fully understand it :duh: . It is'nt pretty !. But my formular yeals an error that is to small to be reported in %, but rather ppm (parts per million), so im in no doubt that it's correct :silly2:. Hopefully someone else can understand this nut-formular, and link it to the decay.

I'll post it once I get home, in a new thread

From this formular I can't see anything else that Adjusted Guardian truly is adjusted. I'ts the only one that does'nt follow this formular. If this armor was to follow this 'normal' formular it's durability should have been 947, but it's 2000.

btw. matlab rocks :kos:
Click to expand...
Cool, look forward to reading it. :scratch: :D
 
Nutarix said:
I have a formular that computes the durability from the different damage types and it's value. It's a 6'th order unlinear forumlar. It's been sitting in my harddisk for a couple of days, as I still fail to fully understand it :duh: . It is'nt pretty !. But my formular yeals an error that is to small to be reported in %, but rather ppm (parts per million), so im in no doubt that it's correct :silly2:. Hopefully someone else can understand this nut-formular, and link it to the decay.

I'll post it once I get home, in a new thread

From this formular I can't see anything else that Adjusted Guardian truly is adjusted. I'ts the only one that does'nt follow this formular. If this armor was to follow this 'normal' formular it's durability should have been 947, but it's 2000.

btw. matlab rocks :kos:
Click to expand...

Not to discredit this in advance, but, isnt it ALWAYS possible to create a formula to represent a limited amount of data? Ok, the formula can get very complex, but as it seems so is yours.

But I am also curious ;)
 
Witte said:
Not to discredit this in advance, but, isnt it ALWAYS possible to create a formula to represent a limited amount of data?
Click to expand...

Yes ;)

Let's hope to be surprised :)
 
Witte said:
Not to discredit this in advance, but, isnt it ALWAYS possible to create a formula to represent a limited amount of data? Ok, the formula can get very complex, but as it seems so is yours.

But I am also curious ;)
Click to expand...

You are right, when it comes to fitting data, you can allways find a formular that fits you data, the question however does it fits all you data, how good does it perform ?.

But the thing that made my brain trick me into believing that this formular is correct (except from the guardian adjusted) is the fact that the formular computes the durability to an precision so small, that it's virtual not there. The size of this error looks like it's due to mashine math precision. AFAIK if I thunkated the recreated durability, when the error was zero ;) The formular is not complete tho, as it does not take inaccount of the adjusted gaurdian version. I tryed with alot of different n'th order formulars, but only this one seemed to be correct.

enought with talking ... boss is comming ... I need to work :)
 
Hmmm. Let's say you have 6 data points (x1,y1), (x2,y2) ... (x6, y6) for which you know both x and y. Now suppose you have a seventh data point (x7, y7) for which you know x7 but not y7. But you know there's a relation between x and y.

You can use Matlab to obtain a formula for y given x based on your 6 points (if you are correct in believing there's a relation between x and y). The formula is in general an approximation to y but is exact for the data points. So the fact your formula producing exact results for your data set just means Matlab is working! The fact that it doesn't work for Guardian Adjusted suggests to me that either the approximation is poor and needs more data points or, more likely imo, that y depends on more variables than you've assumed.
 
Nutarix said:
You are right, when it comes to fitting data, you can allways find a formular that fits you data, the question however does it fits all you data, how good does it perform ?.

But the thing that made my brain trick me into believing that this formular is correct (except from the guardian adjusted) is the fact that the formular computes the durability to an precision so small, that it's virtual not there. The size of this error looks like it's due to mashine math precision. AFAIK if I thunkated the recreated durability, when the error was zero ;) The formular is not complete tho, as it does not take inaccount of the adjusted gaurdian version. I tryed with alot of different n'th order formulars, but only this one seemed to be correct.

enought with talking ... boss is comming ... I need to work :)
Click to expand...

The durability of Adjusted Guardian was changed by MA a while ago.
Back then its durability was less than 1000, but the armor wasn't decaying (at least not like it should)
Maybe they didn't change to the correct value of what the general formula would give.

If your formula works for all other armors except that one I say it is good enough for me.
 
Just my two pecs:

- you can use Sweat, fruits or Common Dung to find the exact TT

- you can do it on the field, by using a pvp trade window
 
mrproper said:
Just my two pecs:

- you can use Sweat, fruits or Common Dung to find the exact TT

- you can do it on the field, by using a pvp trade window
Click to expand...

I use Sweat because of the weight it is smaller (and if you test a lot you'll notice that speed is important)

About in the field, I believe that it can be done but you must use the exact stack number or higher in the TT you can add by small ammounts (499+499 Sweat it still appears as 0.00 in pvp trade but it TT 501 appears right as 0,01, just like 499+499, of course to test some kinds of decays doing it in the field is much better but it will require a patient friend to help you)
 
MG Mighty said:
I use Sweat because of the weight it is smaller (and if you test a lot you'll notice that speed is important)

About in the field, I believe that it can be done but you must use the exact stack number or higher in the TT you can add by small ammounts (499+499 Sweat it still appears as 0.00 in pvp trade but it TT 501 appears right as 0,01, just like 499+499, of course to test some kinds of decays doing it in the field is much better but it will require a patient friend to help you)
Click to expand...

Problem is, few really have leftovers of 500 sweat hanging around if they feel like checking a TT value, but we all have shiznit in our inventories and maybe some fruits.

As for the field test, a team of two testers can individually check 2 different items at the same time, without loosing time to get to a TT.

Another issue to look into..

Sweat TT + Nexus TT = ME TT (same for fruit and dung)

does the final product TT includes the 0.00001 TT for the free resources?
 
My god, MG Mighty have you got gold in that head or what ? :) I thought about this thing before but I didn't quite see the approximation of the Trade Terminal that you posted :) guess you solved the problem way before I even figured the start of it :) Gratz. you are a true visionary :)
 
Dragon armor

impact protection: 7

Decay from daikiba: 353 BVS (actually, haimoros ;) )
Decay from cornundo: 353 BVS

7 hp/0.353 pecs = 19.83 damage/pec decay for impact

Pretty near the midway point (20.07) between warrior (impact 5) and pixie (impact 9), it seems.

Is there a place on the wiki for recording these for each armor? I couldn't see it.

Edit: Just for comparison, 1A plated pioneer armor (6 impact total--1 less) decays ~.2465 pecs/hit for an eco of 24.3 damage/pec decay.


Edit:

stab protection: 7

Decay from prancer (wiki claims 100% stab): 353 BVS
Decay from sabakuma (wiki claims 100% stab): 353 BVS

7 hp/0.353 pecs = 19.83 damage/pec decay for stab

cut protection: 7

Decay from caperon (wiki claims 100% cut): 353 BVS

7 hp/0.353 pecs = 19.83 damage/pec decay for cut

At least for dragon, C/S/I protection all costs the same.


C/S/I protection: 21

Decay from atrox (33/33/33): 1931 BVS

21/1.931 pecs = 10.875 damage/pec decay for C/S/I
 
Last edited:
There is no collumn for it on wiki, is because all armors decay the same. You can write a guide, and post it on wiki in the guide section. I was planning to do that somwhere in the future. I also tested the decay as function of the TT value. And tried to figure out what durability does. My conclusion: The only factors that matter are protection and TT value.

When you create a decay collumn on wiki, you might give people the impression that some armors decay more and some decay less, for other reasons than the protection they offer. So it will just be more confusing.
 
It seems the more is found out, the more confusing it gets :S.

I always was under the impression that each damage type decay on its own so to speak. But it seem now we can add all damages and then calculate total decay. It does explain some irregular testresults I had.

And I also cant figure out a relation between durability and the decay. However, armors with high durability seem to decay abit less. Although its hardly noticable.
Click to expand...

Well the last thing you had to say here was this. :laugh:

I didn't ever see that 6th order equation posted, did you? If you have an equation that correctly calculates decay for each damage type and for combined damages like I/C/S then it should be added to the wiki as a column to calculate the cost for each armor. Otherwise i think a separate entry for measured values is appropriate. It's information that people might find useful in planning their hunting kit. Put in a disclaimer that all armor follows the same trend, whatever, the point is that this is information and information is what the wiki is for.

I wouldn't have done the separate and combined tests if i'd seen anything that allowed me to calculate them. I also wanted to make sure there weren't any nasty surprises here like cut on gremlin... I'm still going to test burn+penetration to make sure they are reasonable, although they're probably impossible to test independently.
 
Doer said:
Well the last thing you had to say here was this. :laugh:

I didn't ever see that 6th order equation posted, did you? If you have an equation that correctly calculates decay for each damage type and for combined damages like I/C/S then it should be added to the wiki as a column to calculate the cost for each armor. Otherwise i think a separate entry for measured values is appropriate. It's information that people might find useful in planning their hunting kit. Put in a disclaimer that all armor follows the same trend, whatever, the point is that this is information and information is what the wiki is for.

I wouldn't have done the separate and combined tests if i'd seen anything that allowed me to calculate them. I also wanted to make sure there weren't any nasty surprises here like cut on gremlin... I'm still going to test burn+penetration to make sure they are reasonable, although they're probably impossible to test independently.
Click to expand...

Yes i know the formula, dont have it by hand though. I also know why gremlin is different with cut.

Here is a shot from the research i did together with ivol:


[br]Click to enlarge[/br]

The minimal decay of an armor is total protection*10 in mpec.
 
Last edited:
Nice, that looks like a pretty good fit, but it's a bit hard to decipher. Are you planning on posting a description of your findings?
 
Doer said:
Nice, that looks like a pretty good fit, but it's a bit hard to decipher. Are you planning on posting a description of your findings?
Click to expand...

Lol, my thought exactly. But if you turn it upside down, it makes more sense...
 
I wanted to make a post about it, but some things still needed to be tested, and i kinda lost the interest ;)

Short explenation: the horizontal dashed lines are the minimal decay of an armor, no matter how little damage you do. This minimal decay value, is total protection of the armor (so all protection types added) times 10 (in mpec).

In case of gremlin, this is 670 mpec (0.670pec).

When you look at the curve, and you lookup the decay at 9 damage (cut of gremlin), you will find it is less then 670 mpec. So thats the reason the decay is off the curve.


Further more: Protection vs TTvalue curve is straightlined, and the decay goes down following the curve. That means that when you wear a decayed armor, you dont pay full cost in decay, but just the decay your armor is protecting. Untill you reach the minimal decay of the armor.

Hope that make it clear ;)
 
Witte said:
I wanted to make a post about it, but some things still needed to be tested, and i kinda lost the interest ;)

Short explenation: the horizontal dashed lines are the minimal decay of an armor, no matter how little damage you do. This minimal decay value, is total protection of the armor (so all protection types added) times 10 (in mpec).

In case of gremlin, this is 670 mpec (0.670pec).

When you look at the curve, and you lookup the decay at 9 damage (cut of gremlin), you will find it is less then 670 mpec. So thats the reason the decay is off the curve.


Further more: Protection vs TTvalue curve is straightlined, and the decay goes down following the curve. That means that when you wear a decayed armor, you dont pay full cost in decay, but just the decay your armor is protecting. Untill you reach the minimal decay of the armor.

Hope that make it clear ;)
Click to expand...

Pretty clear, but what about damage? I feel that at 75% TT my armor protects about 50% of the damage it should.
 
mrproper said:
Pretty clear, but what about damage? I feel that at 75% TT my armor protects about 50% of the damage it should.
Click to expand...

We only tested that with pixie, and the result was that at 50% TT, the protection is also 50%.

And looking at the decay vs TT curve, this makes allot sense.
 
Witte said:
We only tested that with pixie, and the result was that at 50% TT, the protection is also 50%.

And looking at the decay vs TT curve, this makes allot sense.
Click to expand...

The it must be those lucky mobs hitting me with full damage through and armor that would absorb it all to 1.0,.... 3 times in a row... when my armor was at 75% TT.
 
mrproper said:
The it must be those lucky mobs hitting me with full damage through and armor that would absorb it all to 1.0,.... 3 times in a row... when my armor was at 75% TT.
Click to expand...


There might still be some influence from durability somewhere. Maybe at some point, when the armor runs out of durability, it will stop protecting all together.

Although, with the armors we tested (mainly pixie and kobold), we did not find a point where all protection was lost. And the protection was in case of pixie exactly the same % of the TT value. With kobold we did not test this in particular, but from the other tests it seemed to also apply to this armor.

So if you can give us a situation that is relativly easy to reproduce, where an armor has lost all its protection, I would be glad to hear it. The more info we know the better ;).
 
Witte said:
Yes i know the formula, dont have it by hand though. I also know why gremlin is different with cut.

Here is a shot from the research i did together with ivol:


[br]Click to enlarge[/br]

The minimal decay of an armor is total protection*10 in mpec.
Click to expand...

Why exactly did you respesent a straight line as a curve?
 
jsm said:
Why exactly did you respesent a straight line as a curve?
Click to expand...

The decay vs TT is a curve, and is shown as a curve, the minimal decay is a straight line, and is shown as a straight line. But the graph is just prototype and not perfected. Maybe will do that sometime in the future.
 
Witte said:
The decay vs TT is a curve, and is shown as a curve, the minimal decay is a straight line, and is shown as a straight line. But the graph is just prototype and not perfected. Maybe will do that sometime in the future.
Click to expand...
It's not a curve. It only appears so because the Y axis goes up by powers, not linearly :)

If you put it in a linear graph, and it still curves significantly, then I am wrong. And you win a medal :p
 
Last edited:
jsm said:
It's not a curve. It only appears so because the Y axis goes up by powers, not linearly :)
Click to expand...


If you use a non logarithmic y axis, you still get a second grade curve. Like i said its just inbetween work. But is still a 2nd grade function, so a curve.
 
Has anyone compared the decay of 100 argo young hits on rascal vs. shadow? :p
 
You must log in or register to reply here.
Back
Top