test content
Logo
What is the Arc Client?
Install Arc
Crafting stars math: getting to 5/5
Archived Post
Posts: 1,156,071 Arc User
I've been doing some basic experimenting on how crafting stars work, and wanted to put what I had out for more detailed investigation and/or dev explanation.
Background:
When you use a crafting table in Research mode, you can drag appropriate items from your inventory into the research window to deconstruct them. You can get crafting skill-ups, ordinary material components of various rarities, and/or special components known variously as Insight (Arms), Profundity (Mysticism), or Epiphany (Science). There is one grade for each crafting level in each major crafting profession. We don't currently have a good generic name for these; it would be helpful if the Devs would let us know what they're called internally.
These special components (or, if you prefer, currency) are the only way to purchase a significant number of blueprints; currently all conversions, all power replacers, some mid- and high-level consumables, and some high-level items in most other categories require these as all or part of the blueprint purchase price. AFAIK, they are not required to build any items, merely to acquire the blueprints.
As these are Bound items, currently the only known way to get these these is to deconstruct items of your crafting profession at a crafting table of your profession. Each experiment you run (single press of the Experiment button) is given an overall rating, from 0 to 5 out of 5 (displayed as 0/5, 1/5, ... 5/5); this is displayed both numerically and as a bar graph at the bottom of the table window.
This overall rating controls the likelihood of getting special components and which level, how much your skill can increase, and possibly other things. As has been confirmed by the devs elsewhere, to have the best chance of a special component drop of your level, you need to be running 5/5 experiments (fill the bar graph). At higher levels, if you want to get lower level special components, your best chance is from doing experiments at one star grade level below 5/5 for each level below your current the desired special components are.
Observations:
Each item you can deconstruct has a hidden numerical value. This value is approximated by a number of stars, from 0 to 5, which can only be seen when the item is in the research window. If you watch the bar graph at the bottom of the window carefully, however, it can be seen that not all items of the same star value move the bar graph the same amount. It also becomes clear that the stars are not representations of linear quantities; a 3-star is more valuable than three 1-stars. I'm going to refer to the semi-hidden detailed value as "points" for now; if the devs want to let us know what they're actually called to aid communication, that would be helpful.
I took some time to run a number of experiments with different combinations of items I had in my inventory. Note that because the stars themselves are approximations, rounding boundary conditions are not sufficiently constrained.
1+1 = 2/5
1+1+1 = 2/5
1+1+1+1 = 3/5
2+1 = 2/5
2+1+1 = 3/5
2+2+1+1 = 3/5
2+2+1+1+0 = 3/5
2+2+1+1+1 = 4/5
2+2+1+1+1+0 = 4/5
2+2 = 3/5
2+2+2 = 3/5
2+2+2+2 = 4/5
2+2+2+1 = 4/5
2+2+2+1+0 = 4/5
2+2+2+2+1 = 4/5
2+2+2+2+1+1 = 4/5
2+2+2+2+1+1+1 = 4/5
2+2+2+2+1+1+1+1 = 5/5
The simplest explanation for the above is that each star (whether as an input on an item or in the output quality of the experiment) is worth / costs about twice what the previous one was. It is not clear from my data whether 0 star items are truly worth 0, or just a very small amount. Note that these values may be relative to your current crafting and/or character level; good items from lower levels may be worth less.
For the mathematically inclined, we'll refer to the star value as S and the point value as P. Depending on exactly how their system works, for S>=1, either P = 2^(S-1) or P = 2^S. Correspondingly, either S = Log2(P) or S = Log2(P)+1. (Realistically, it's probably a look-up table or case statement.) It's entirely possible that the geometric basis is shifted even further; 5/5 might be 255, for instance, with others shifted appropriately.
To approximate all this in your head, it's easiest to use the S = Log2(P) + 1 version, which means the following values:
Stars..Points
==========
0............0
1............1
2............2
3............4
4............8
5...........16
Add up the point value of all the items, and compare it to the above. If it falls in between two values, it's not yet clear how rounding works; either more experimentation or a Dev explanation will be needed. From limited data, it seems to be doing a round-nearest, but whether on the linear or the geometric scale and with what rounding arrangement is not clear. In practical terms, if you're going for 5/5, it appears that 11 points is probably not enough, 13 points should be solid, and anything in between depends on fine details we can't predict yet.
Further experimentation along these lines would be a good idea, and some Dev insight into how close this comes to what's actually going on would be very helpful. The closer we can come to describing what's really happening, the easier it will be going forward to submit explicit bug reports, and comment intelligently on any crafting changes or developments.
Background:
When you use a crafting table in Research mode, you can drag appropriate items from your inventory into the research window to deconstruct them. You can get crafting skill-ups, ordinary material components of various rarities, and/or special components known variously as Insight (Arms), Profundity (Mysticism), or Epiphany (Science). There is one grade for each crafting level in each major crafting profession. We don't currently have a good generic name for these; it would be helpful if the Devs would let us know what they're called internally.
These special components (or, if you prefer, currency) are the only way to purchase a significant number of blueprints; currently all conversions, all power replacers, some mid- and high-level consumables, and some high-level items in most other categories require these as all or part of the blueprint purchase price. AFAIK, they are not required to build any items, merely to acquire the blueprints.
As these are Bound items, currently the only known way to get these these is to deconstruct items of your crafting profession at a crafting table of your profession. Each experiment you run (single press of the Experiment button) is given an overall rating, from 0 to 5 out of 5 (displayed as 0/5, 1/5, ... 5/5); this is displayed both numerically and as a bar graph at the bottom of the table window.
This overall rating controls the likelihood of getting special components and which level, how much your skill can increase, and possibly other things. As has been confirmed by the devs elsewhere, to have the best chance of a special component drop of your level, you need to be running 5/5 experiments (fill the bar graph). At higher levels, if you want to get lower level special components, your best chance is from doing experiments at one star grade level below 5/5 for each level below your current the desired special components are.
Observations:
Each item you can deconstruct has a hidden numerical value. This value is approximated by a number of stars, from 0 to 5, which can only be seen when the item is in the research window. If you watch the bar graph at the bottom of the window carefully, however, it can be seen that not all items of the same star value move the bar graph the same amount. It also becomes clear that the stars are not representations of linear quantities; a 3-star is more valuable than three 1-stars. I'm going to refer to the semi-hidden detailed value as "points" for now; if the devs want to let us know what they're actually called to aid communication, that would be helpful.
I took some time to run a number of experiments with different combinations of items I had in my inventory. Note that because the stars themselves are approximations, rounding boundary conditions are not sufficiently constrained.
1+1 = 2/5
1+1+1 = 2/5
1+1+1+1 = 3/5
2+1 = 2/5
2+1+1 = 3/5
2+2+1+1 = 3/5
2+2+1+1+0 = 3/5
2+2+1+1+1 = 4/5
2+2+1+1+1+0 = 4/5
2+2 = 3/5
2+2+2 = 3/5
2+2+2+2 = 4/5
2+2+2+1 = 4/5
2+2+2+1+0 = 4/5
2+2+2+2+1 = 4/5
2+2+2+2+1+1 = 4/5
2+2+2+2+1+1+1 = 4/5
2+2+2+2+1+1+1+1 = 5/5
The simplest explanation for the above is that each star (whether as an input on an item or in the output quality of the experiment) is worth / costs about twice what the previous one was. It is not clear from my data whether 0 star items are truly worth 0, or just a very small amount. Note that these values may be relative to your current crafting and/or character level; good items from lower levels may be worth less.
For the mathematically inclined, we'll refer to the star value as S and the point value as P. Depending on exactly how their system works, for S>=1, either P = 2^(S-1) or P = 2^S. Correspondingly, either S = Log2(P) or S = Log2(P)+1. (Realistically, it's probably a look-up table or case statement.) It's entirely possible that the geometric basis is shifted even further; 5/5 might be 255, for instance, with others shifted appropriately.
To approximate all this in your head, it's easiest to use the S = Log2(P) + 1 version, which means the following values:
Stars..Points
==========
0............0
1............1
2............2
3............4
4............8
5...........16
Add up the point value of all the items, and compare it to the above. If it falls in between two values, it's not yet clear how rounding works; either more experimentation or a Dev explanation will be needed. From limited data, it seems to be doing a round-nearest, but whether on the linear or the geometric scale and with what rounding arrangement is not clear. In practical terms, if you're going for 5/5, it appears that 11 points is probably not enough, 13 points should be solid, and anything in between depends on fine details we can't predict yet.
Further experimentation along these lines would be a good idea, and some Dev insight into how close this comes to what's actually going on would be very helpful. The closer we can come to describing what's really happening, the easier it will be going forward to submit explicit bug reports, and comment intelligently on any crafting changes or developments.
Post edited by Archived Post on
0
Comments
It is actually possible to get a greater return than you put into for some items. Many customizable crafting items will cost 1 or 2 dust to make, but their return is listed as 4 dust and 8 crystals. If you do 1-3 stars you will receive the complete cost back to make them. If you go to 4 or 5 stars you can actually earn back more materials researching them than it costs to make them. Getting back both dust and crystals.
the items in question were:
Colossal Artillery
Accelerated Overpower.
Reinforced Heart
Accelerated Radiant Explosive
Reinforced manus
Colossal tympanic membrane
Fulminating radiant stampede
0 star item - 0 stars
0+0 - 1 star
0+0+0 - 2 stars
0+0+0+1 - 3 stars
0+0+0+1+1 - 4 stars
0+0+0+1+1+1 - 4 stars
0+0+0+1+1+1+0 - 5 stars.
I have the screen shots from it, if you need to see for any further details. Took screenshots as I added each component.
Have to look into the items you can change the stats on though. Didn't notice anything when I was making them before.
You can trivially see that not all x star items are created equal, without needing to actually break anything down.
It doesn't give you exact amounts though, and it doesn't appear to give you exactly the same percentage depending on where you put it in line, though the total always appears to add up to the same amount...
5/5 seemed to be about 10 silver resources from 101-200, 80 silver resources from 201-300, then at 301 it suddenly skyrocketed to about 4 gold resources.
YMMV but this has seemed consistent for me in testing so far.
Which items are you using for which craft? I noticed a couple times that there were items that I made and could research them and get good stuff but that was during beta.