Jump to content
Search In
  • More options...
Find results that contain...
Find results in...
Anthrage

Experimentation Roll Result Transference

Recommended Posts

So I did some testing last night, and while I need to do more testing, I have reason to at least consider the possibility that experimentation roll results achieved when turning a raw resource - say wood - into a crafting material - a plank - can influence the subsequent results when you turn that material into a component.

 

This terribly boring video has 2 possible instances of this, just after the 27 minute mark. Planks which had failure rolls in their creation - tagged with CF for Critical Failure and F for Failure - seemed to lead to similar rolls on the item component they were used in. The only Amazing rolls which occurred also were similarly linked.

 

The sample size is very small, but it is suggestive enough to warrant further testing. Has anyone seen any signs of something similar before? If you use 3 planks which all had Amazing Experimentation rolls, does it increase the chance that the Staff Limb you make out of them for example will also see an Amazing roll?

Share this post


Link to post
Share on other sites

So I did some testing last night, and while I need to do more testing, I have reason to at least consider the possibility that experimentation roll results achieved when turning a raw resource - say wood - into a crafting material - a plank - can influence the subsequent results when you turn that material into a component.

 

This terribly boring video has 2 possible instances of this, just after the 27 minute mark. Planks which had failure rolls in their creation - tagged with CF for Critical Failure and F for Failure - seemed to lead to similar rolls on the item component they were used in. The only Amazing rolls which occurred also were similarly linked.

 

The sample size is very small, but it is suggestive enough to warrant further testing. Has anyone seen any signs of something similar before? If you use 3 planks which all had Amazing Experimentation rolls, does it increase the chance that the Staff Limb you make out of them for example will also see an Amazing roll?

Damn you!!!!! 

 

Now I am going to have to check this out ... as if I did not already have enough to track in spreadsheets  :blink:


Don't forget, the one EK that no one will judge you for looting your guilds treasury is Anhrez's Doober Shack. Where you can take those long con gains and 'simplify' them to more easily fit in your inventory. While you are unloading your hard earned winnings, swing by the Bazaar and pick up something to celebrate your genius.

LR0tCJt.png

Share this post


Link to post
Share on other sites

Anecdotally, yea it "feels" like this happens more than odds would indicate.

 

I wonder if it is some sort of hidden "source quality" modifier applied behind the scenes, and a reason for actually using that seemingly pointless ("it's a convenience") "Abort" button they have going now.

 

https://youtu.be/og_DJoG08T4?t=268

 

The best way to test this we don't have access to just yet BP's.  If we could make a run of identical metal bars for example, and then see the crafting results for making a single component (pommel) from Amazing success bars vs critical failure bars, then we would be getting somewhere.  

 

1800 ore later, we may have our answer.  

Share this post


Link to post
Share on other sites

Yes, if one could accumulate a reasonably high number of identifiable high success result experimentation roll materials, and then make a high number of components, it should show something. I will try to scale this up this weekend and see if this dynamic is real.

 

In a way it makes logical sense, and aside from the Abort button, is that have something similar with the individual slots in a multi-slot crafting interface where you use a range of quality types - a white, a green and a blue for example, can make a white, green or blue result. There is something going on behind the scenes beyond RNG and straight averaging.

Edited by Anthrage

Share this post


Link to post
Share on other sites

So I did some testing last night, and while I need to do more testing, I have reason to at least consider the possibility that experimentation roll results achieved when turning a raw resource - say wood - into a crafting material - a plank - can influence the subsequent results when you turn that material into a component.

 

This terribly boring video has 2 possible instances of this, just after the 27 minute mark. Planks which had failure rolls in their creation - tagged with CF for Critical Failure and F for Failure - seemed to lead to similar rolls on the item component they were used in. The only Amazing rolls which occurred also were similarly linked.

 

The sample size is very small, but it is suggestive enough to warrant further testing. Has anyone seen any signs of something similar before? If you use 3 planks which all had Amazing Experimentation rolls, does it increase the chance that the Staff Limb you make out of them for example will also see an Amazing roll?

 

thanks for taking the time to do this...


WAZ6Fov.png

"The cinnabar is a lie"

Share this post


Link to post
Share on other sites

Anthrage, if you don't mind a suggestion, here's how I'd conduct the experiment.

(I'd do this myself, but between Cloudbleed and a dinner party I don't expect to be playing much this weekend)

 

I'd focus on make metal bars and then transforming them into staff heels.

This is a relatively cheap item to make.

 

Each time you make a metal bar, code the item quality with a number

 

1 = critical failure

2 = failure 

3 = success

4 = moderate success

yada, yada yada

 

Record the number in column 1 of a spreadsheet

 

Immediately there after, turn the finished bar into a staff heel.

Once again, code the item quality with a number

Record this in column 2 of your spreadsheet

 

Do this as many times as you can stand and then create a jittered scatter plot charting column 1 X column 2.

Finally add a trend line to the scatter plot

 

This is probably the easiest way to see whether there is a relationship between the values...


WAZ6Fov.png

"The cinnabar is a lie"

Share this post


Link to post
Share on other sites

This feels un-testable right now. I also wouldn't bother. This is like a derivative of a derivative type level testing. The top level values aren't fixed yet. I.E: once you completed this testing, a single value change at any level and you'll have to re-figure out everything from the top again.

Share this post


Link to post
Share on other sites

This feels un-testable right now. I also wouldn't bother. This is like a derivative of a derivative type level testing. The top level values aren't fixed yet. I.E: once you completed this testing, a single value change at any level and you'll have to re-figure out everything from the top again.

 

Yes and no

 

I agree with respect to individual parameters, however, I think that there is value in understanding the fundamental model that is being used.

In particular, knowing whether crafting success is independent of the success of the individual components seems extremely valuable.


WAZ6Fov.png

"The cinnabar is a lie"

Share this post


Link to post
Share on other sites

This feels un-testable right now. I also wouldn't bother. This is like a derivative of a derivative type level testing. The top level values aren't fixed yet. I.E: once you completed this testing, a single value change at any level and you'll have to re-figure out everything from the top again.

 

I am going to agree with Narsille on this. There must be some information to be gathered from this. Also, some people just enjoy these kinds of experiments and connecting of dots.

Share this post


Link to post
Share on other sites

This is what I had planned to do next, with some minor differences. Last night I crafted 2 items, got 1 Amazing Roll on the material, then a Great on the item...again, not conclusive by any means, but the more I pay attention to this, the more it seems like there is something there.

 

Therefore, on with the mass materials making!

 

Anthrage, if you don't mind a suggestion, here's how I'd conduct the experiment.

(I'd do this myself, but between Cloudbleed and a dinner party I don't expect to be playing much this weekend)

 

I'd focus on make metal bars and then transforming them into staff heels.

This is a relatively cheap item to make.

 

Each time you make a metal bar, code the item quality with a number

 

1 = critical failure

2 = failure 

3 = success

4 = moderate success

yada, yada yada

 

Record the number in column 1 of a spreadsheet

 

Immediately there after, turn the finished bar into a staff heel.

Once again, code the item quality with a number

Record this in column 2 of your spreadsheet

 

Do this as many times as you can stand and then create a jittered scatter plot charting column 1 X column 2.

Finally add a trend line to the scatter plot

 

This is probably the easiest way to see whether there is a relationship between the values...

Share this post


Link to post
Share on other sites

This is what I had planned to do next, with some minor differences. Last night I crafted 2 items, got 1 Amazing Roll on the material, then a Great on the item...again, not conclusive by any means, but the more I pay attention to this, the more it seems like there is something there.

 

Therefore, on with the mass materials making!

 

FWIW, I harvested 27 stacks of white silver this AM.

 

I am going to try to get up to either 50 (or if I am ambitious) 75.

I'm then going to make a poorly made socks load of staff heels.

 

I'll share the raw data and any results once i am done.


WAZ6Fov.png

"The cinnabar is a lie"

Share this post


Link to post
Share on other sites

I did a preliminary test.  I unfortunately, I don't see an evidence in support of Anthrage's theory.

 

I created roughly 50 silver book clasps using white silver ore and running 3X crafting pots the whole time.

 

Each time I created a metal bar, I recorded its quality level.  I coded a critical failure as a 1, a failure as a 2, all the way up to an Amazing success which was a 7.  Immediately after creating the metal bar, I created a book clasp and, once again, coded my success level.

 

Next, I pulled all the data into R and created a scatter plot of the quality of the metal bar versus the quality of the book clasp.

(I added some jitter so the data points would be visible and added a linear trend line)

296k9ld.png

 

 

 

To my eyes, the two variables look independent of one another.

 

Here's the raw data (minus failures that didn't create a bar)

 

 

 

 

MetalBar BookClasp 7 6 7 4 7 4 7 4 7 4 7 4 7 3 7 3 7 3 7 3 6 5 6 5 6 4 6 4 6 4 6 3 6 3 6 2 5 7 5 5 5 5 5 5 5 5 5 5 5 5 5 4 5 4 5 3 5 3 5 3 5 3 5 2 5 1 5 1 4 7 4 6 4 5 4 4 4 4 4 4 4 3 4 3 4 3 4 3 4 3 4 3 2 5

Edited by narsille

WAZ6Fov.png

"The cinnabar is a lie"

Share this post


Link to post
Share on other sites

I have a question. Is that raw data in order of your attempts? If so, there's a distinct pattern of descending quality every 2 attempts.

 

Nope, I sorted it


WAZ6Fov.png

"The cinnabar is a lie"

Share this post


Link to post
Share on other sites

Even if your conclusion was different, I'd still suspect that any relationship to crafting component success and ending stage assembly was merely a coincidence (or an enormous and glaring bug in their code!). Thankfully it did not! Kudos.

Share this post


Link to post
Share on other sites

I don't disagree in principle, but there is already some code at work for a similar dynamic, with the averaging and transference of qualities through component and item creation. As anyone who has crafting a large quantity of items can attest, there is something going on there beyond the obvious. Could be a bug, could just be RNG, or perhaps there is something structured going on we do not understand the dynamics of yet. Testing is the only we we're going to find out unfortunately, unless they just come right out and tell us. :)

 

Even if your conclusion was different, I'd still suspect that any relationship to crafting component success and ending stage assembly was merely a coincidence (or an enormous and glaring bug in their code!). Thankfully it did not! Kudos.

Share this post


Link to post
Share on other sites

I spent a few hours trying to make some Perfect (As in all amazing sub-parts rolled at amazing on final craft) White Blacksmithing Armor Pieces. Unfortunately i was only able to make about 12 amazing bars after over 1000 ore (Amazing rolls on 5 pips are about 1 in 10 for me, i sacrificed a durability to roll higher easier). Testing those with some amazing reinforced leather while making a helmet 4 times got me the following results. 2 normal success and 2 Moderate success. I tested it again with a set of green amazing bars and only got a moderate on the helmet. So far it seems like each roll is independent unless i just need a bigger data sample

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...