8 - Power Crafting
Power crafting is what most Traders ultimately strive for, to be able to craft that supreme space ship, or that dazzling drink, or that fearful firearm, or that prizzy probot, just to mention a few examples. This chapter has grown to by far the largest section of this guide, not because power crafting is hard but since several pieces of information are glued together.
This chapter is split into the following sections:
- Crafting Stations
- Assembly
- Experiment
- Weighed Average Result
- Resource Caps
- Expertize: Resource Refinery
- Complexity versus Time
The three colored sections include a shared-and-continued example that illustrate how one piece of information adds to the other to finally result in a capped item.
By now you know basic crafting and many items are no harder to craft than to find proper resources and assemble them, perhaps you must first craft some components. Often these simpler items are made by bulk or by recycled resources. However,
the only way to improve on a quality item is to experiment on it.
Power crafting depends on (in no particular order)
- real-life fortune
- Luck
- expertize
- resource stats
- profession skills
- entertainer buffs
- city specialization
- food & drink buffs
- RE* power-up items
- RE* modded wearables
- diner or helper-monkey
- old skill enhancing tapes
- rating of crafting equipment
. . .
*
RE = Reverse Engineering
For numerous items the final result depends on the quality components or items, whether required or not.
Crafting Stations
To be able to experiment on an item you must use a specialized crafting tool
together with a crafting station. The two must match, the tool and the station must be of the same kind. See chapter 5 for details on schematic
complexity versus crafting tools and stations.
Only when you are crafting nearby a Crafting Station and see an icon for the station ny your name label, after the window Crafting Summary you see another window that is named Finish Crafting. The usual options are: Experiment, Create Prototype, and Create Manufacture Schematic.
Public Crafting Stations are found in all NPC cities and around starports.
Private Crafting Stations are crafted and sold by Structure Traders and are placed as furniture in houses. If you can enter the room you can benefit from the station, you do not have to own neither the house nor the station.
Private Crafting Stations have a
Functionality Rating and its maximum is 45.00. Both the quality of the crafting tool and of the crafting station have an impact on the result, on a diminishing return but nevertheless.
Crafting stations do not need energy or maintenance, once obtained they do not cost you anything.
Droids with crafting modules act as mobile Private Crafting Stations for their owners. Their functionality rating is NUL, but most crafters feel that droids act as level-30 stations.
Assembly
When you add resources and components to empty slots at the assembly window, please
ignore the colored bars, the red/yellow/green colors are often way off, but rather look at...
The lower right-hand section of the window displays the quality of the highlighted resource in detail, each experimentation line by itself. Just remember that neither expertize bonus nor stats-relative-caps are accounted for at this window.
Click Assemble and the next window that is named Crafting Summary displays the random outcome; you want at least a Great Success. Quality items also display the
initial result of the assembly; if the schematic has several experimentation lines they are listed each line by itself.
With perfect resources, bonuses, and a great or amazing success, the initial result reads 29%.
The appropriate expertise bonus is applied during this assembly phase to compute the initial result and to compute the maximum possible result, which is explained later.
Assembly skill points affects the probability for a Great or Amazing Success during this assembly phase, which gives a better initial result for your product.
Sockets: the more Assembly skill points the better chance for a socket and 160 Assembly skill points warrants a socket. Sockets are very important on wearable items such as armor, clothes, and weapons, because SEAs are put into sockets, see Reverse Engineering.
It is possible to improve the Assembly skill with Pyollian Cakes, SEAs and PUPs, Manufacturing Center city specialization, and Expertize points. However, except for sockets most Traders ignore assembly points.
Experiment
The only way to improve on the initial result is to experiment on the item. This section covers the basics about this topic, defering the deeper details until next sections.
As you level up you are rewarded with
Experimentation Skill points, for example Food Experimentation Skill. Every 10 points yield one
experimentation box; these boxes are spent at the experimentation window. A master has accumulated 100 experimentation points (some Shipwright skills have 150 points) which means 10 (15) boxes to spend ---
some players say "points" for these boxes but to avoid the ambiguity about "points" versus "points" I call them boxes in this guide.
How to improve my skill?
Expertize can give
+10 points.
SEAs can give you
+24 points, only pre-NGE SEAs can give 25 points and more¹.
A helper-monkey² adds another...
...
+6 to Chefs and Architects,
...
+4 to Engineers, and
...
+2 to others.
A Diner³ gives
+6 to Chefs and Architects but it does not stack with the monkey.
Altogether it is possible to obtain 137/139/141 experimentation points which grants
maximum 13/14 boxes depending on profession.
See Savacc's post
Notes:
¹ SEAs are capped at 25 points but just in relation to boxes.
² A Helper-Monkey is a very-hard-to-achieve collection reward.
³ A Diner is a structure from TCG loot cards.
The
generic crafting skill is improved by Bespin Port and Research Center and it adds to the specialized skills way beyond any cap for experimentation boxes. These generic buffs
improve the dice roll to yield the more Amazing Successes, but they do not give you another experimentation box.
The
Luck modifier increases the chance for Critical Success, also known as Lucky Strike or Lucky Break. A Critical Success is stronger than an Amazing Success.
It is debated whether Bespin Port and Research Center specialization yield the more amazing successes or not. It is also debated whether Luck is limited to the rather rare Critical Successes or if it additionally yields the more Amazing Successes. However, several trusted Traders insist that the generic buffs and that the more Luck yields the more Amazing Successes. Fact is, though, that nobody has put forward any solid statistical data, neither for yay nor nay, and SOE has remained mysteriously silent. Until robust data is published somewhere you may trust whoever, or your gut feelings
Run an Experiment --- some schematics have just one experimentation line. Other schematics have several lines with different meanings, often one is more important than the other and you must decide whichever. Select a line and spend one or more boxes at a time and click Run Experiment; repeat until you are out of boxes.
Each experiment is a dice roll and yields a random outcome in the range Critical Failure to Critical Success; this outcome adds to the initial result.
Each box that you spent on a particular dice roll gives:
Code: Select all
Critical Failure | -14.00
Failure | -7.00
Moderate Failure | ??????
Moderate Success | 1.05
Success | 1.75
Good Success | 3.50
Great Success | 7.00
Amazing Success | 8.05
Critical Success | 8.4
: you spent 3 boxes and got a Good Success; 3 * 3.50 = 11.50 which accumulates to the initial result for the line you experimented on.
Example 2: you spent 3 boxes and got an Amazing Success; 3 * 8.05 = 24.15 --- Yay!!!
A normally buffed Master who experiments in a research center seldom faces worse than Great and Amazing Successes.
Weighed Average Result
This section is the first of three that together cover the most important knowledge a Trader in SWG will need. Right after the introduction there is an example which is augmented to in the following sections, each adding another layer of details. The example is simple enough to get the point across without any distracting bells and whistles. However, now it is time for the Trader touchstone, to separate the grain from the chaff; it is time for you to face the mysteries of Power Crafting ---
remember, this chapter is about schematics for quality items even though not always spelled out.
A schematic specifies several factors which are weighed together. All of these factors come to play, which provides us with maximum flexibility:
- per experimentation line
- one or more stats, and...
- the weights for each stat
- per resource slot
- the resource class, and...
- the amount
Each experimentation line of a schematic specifies its
required stats and for that line they determine which values of a resource to consider. For any suitable resource, the values of those stats directly affect the outcome. However, it is the mix of resources in all resource slots that counts and sets the limit for what is possible to achieve for that line, not each resource by itself.
Each resource slot contributes to the result just by its
relative weight. This means that a resource slot with just a few units does not contribute much in a heavy item, but a big resource slot contributes the more. If it happens that a resource is missing a specified stat it does not withdraw from the result for that stat in that line, except if all resources are missing that stat. Per experimentation line, there is a penalty if a stat is not met at least once by any of the resources.
Notice: it is enough to meet each required stat just once, if that is possible; the reason is explained just after example 1.
The maximum result that is possible to achieve for an experimentation line is based on the sum of the present values of the required stats when they are weighed with their relative amounts
and the experimentation weight of each stat respectively. This sum is better called the
weighed average result, here abbreviated WAR; and this is the theoretical/abstract formula:
Code: Select all
WAR = [
stat1.weight * (A.stat1 * A.amount + B.stat1 * B.amount + ...) /
/ (A.amount + B.amount + ...) +
+ stat2.weight * (A.stat2 * A.amount + B.stat2 * B.amount + ...) /
/ (A.amount + B.amount + ...) +
+ ...
] / 1000
This example displays the importance of amount-per-slot and weigh-of-stats. Assume a schematic with just one experimentation line that it is based on 60% OQ and 40% CD, and it calls for 25 units of Metal and 15 units of Chemical. We happens to have some Colat Iron with OQ=975 and CD=310, and some Lub Oil with OQ=998. First, notice that Lub Oil has no CD and consequently it does not contribute to the CD slot, however, the CD slot is already "filled once" by the iron:
Code: Select all
| 975 * 25 + 998 * 15 310 * 25 | 1
WAR = | 0.6 * ------------------- + 0.4 * -------- | * ----
| 25 + 15 25 | 1000
= (0.6 * 983.625 + 0.4 * 310) * 1/1000
= (590.175 + 124) / 1000 = 714.175 / 1000 = 0.714175
WAR = 71 %
A simpler way to solve the equation is to split it into several smaller expressions. First compute the relative weight per stat, and then compute WAR from the intermediate results:
Code: Select all
OQ_w = (975 * 25 + 998 * 15) / (25 + 15) = 983.625
CD_w = (310 * 25) / (25) = 310
WAR = 0.6 * 983.625 + 0.4 * 310 = 714.175 / 1000 = 71 %
: if it is possible, fill each stat only once. The example displays why; consider what would happen to the result if a fictitious chemical exists with OQ=998 and CD=10. It is easier to find a resource with
one awesome stat than with several. So we use the Lubricating Oil which has no CD as it is instead filled by the metal.
The very same concept applies to the amount that different resources contribute, their relative weight. In a schematic that calls for 1000 units of Ferrous Metal and 200 units of Non-Ferrous Metal the stats on the Ferrous Metal out-weighs the Non-Ferrous in importance by 5 to 1. This means, you can get away with a fair copper if you have splendid Iron, but not vice versa.
Your challenge is to puzzle together the best possible combination of your resources.
Resource Caps
Many resource classes have upper and/or lower caps on one or several stats, then the values of a resource can never ever get outside the specified range. Only the upper caps are considered in crafting.
All values of the required stats are
always adjusted for possible caps before they are applied to the formulas. Hence, this section focuses on "WAR adjusted for caps".
Remember,
Schematic Is King!!! --- just the caps of
the resource class that is called for are considered, each resource slot by itself,
not the caps of the classes that you actually fill the slots with. If the schematic calls for Metal and CD it does not matter that your Colat Iron almost caps its CD.
In formulas, for each capped stat you replace its value with the result from...
Let us revisit Example 1. First recall that Iron has an upper cap on CD at 500, but Colat Iron has an upper cap on CD at 332. Also recall that OQ is never capped on any class so we need not bother. Assume two, somewhat tweaked, clones of the previous schematic, and that we are using the same two resources.
Example 2: Assume that one tweaked schematic calls for
Iron; this schematic considers the cap for Iron, not for the Colat Iron we are using:
Use 620 rather than 310 in the formula:
Code: Select all
| 975 * 25 + 998 * 15 620 * 25 | 1
WAR = | 0.6 * ------------------- + 0.4 * -------- | * ----
| 25 + 15 25 | 1000
= (0.6 * 983.625 + 0.4 * 620) * 1/1000
= (590.175 + 248) / 1000 = 838.175 / 1000 = 0.838175
Adjusted WAR = 84 %
Assume the other tweaked schematic calls for
Colat Iron; now this schematic really considers the cap for Colat Iron:
Code: Select all
310 / 332 * 1000 = 933.73493975904
Use 933.73493975904 rather than 310 or 620 in the formula:
Code: Select all
| 975 * 25 + 998 * 15 933.73493975904 * 25 | 1
WAR = | 0.6 * ------------------- + 0.4 * -------------------- | * ----
| 25 + 15 25 | 1000
= (0.6 * 983.625 + 0.4 * 933.73493975904) * 1/1000
= (590.175 + 373.49397590361) / 1000 = 963.66897590361 / 1000 = 0.96366897590361
Adjusted WAR = 96 %
We have seen that the same two resources can yield very different results, this is determined by the schematics and which resource classes they call for: Metal, Iron, or Colat Iron respectively. These three consecutive examples chisel out the most important crafting rules for schematics and resource caps in SWG.
"JTL" resources have no caps and hence their values can surpass the cap of their parent resource class. However, they can never boost the quality outside the caps of the resource class called for by the schematic, each value is simply capped at 100.0% for that stat in that slot, which is good in itself.
See
resourcetree.xml, or
Resource Caps, or SWGAide >> Resource Info for known caps.
Expertize: Resource Refinery
The Trader Expertize has an ability to refine the resource quality up to 4 %; this improves the "WAR adjusted for caps" and not each resource by itself.
Code: Select all
Refined WAR = Adjusted WAR + Expertize Refinery
Soon players found that rather than adding 4 % this expertise adds 40 points flat. This explains why you hear about 960 as the magic limit --- yeah, 40 of 1000 is indeed 4% but 40 of 960 is 4.167% and then the worse a resource the higher the % ... whatever, we happily accept the "bonus".
Let us review example 2 and 3 and now apply the resource refinery expertize; we assume that it is maximized.
Example 2 --- Iron Code: Select all
Refined WAR = 838.175 + 40 = 878.175
or as decimals
Refined WAR = 0.838175 + 0.04 = 0.878175 = 88 %
Code: Select all
Refined WAR = 963.66897590361 + 40 = 1003.7... ==> 1000
or as decimals
Refined WAR = 0.96366897590361 + 0.04 = 1.0037... = 100 %
the refinery expertize
is not applied to each resource by itself, but just one time and to the final "WAR adjusted for caps". Hence, when we casually say that you should "look for +960-resources" that is just a general advice. In fact, any mix of resources which gives an "adjusted WAR" greater than 960 will cap that item, given a maximized resource refinery expertize and that experimenting is successful.
Complexity versus Time
Each experimentation dice roll adds one point to a schematic's default complexity. This sums up to a
final complexity that is displayed at the crafting window and it also reads at manufacture schematics. It makes no difference if just one, or if several experimentation boxes are used in a dice roll, each roll adds only
one (1) point to the final complexity.
When the item is readied the final complexity points translates to processing time:
- Manual crafting: 2 seconds per point
- Factory run: 8 seconds per point
How many boxes per dice roll?
The question is legitimate and there is no one answer because there are different aspects to consider, these are a few:
- If you need one or more amazing successes to cap an item:
One box at a time until you are happy, then many or all (to reduce the processing time).
- If the item is "cheap" and you want the quickest processing time:
As many boxes as possible each dice roll, unsuccessful results are wasted, rinse and repeat until you have happy.
- If the item's quality rating does not matter:
None (you want the quickest processing time).
There is a
higher risk for failure the more boxes that are spent in a dice roll, thus many crafters play safe. But why not just one box at a time? Partly to minimize the final complexity but also because each dice roll is in itself a risk; any outcome less than a Great Success is a severe loss. Even if a normally buffed master seldom faces failures this risk is not eradicated, just small.
