Doctor Archive

In a previous post http://forums.station.sony.com/swg/board/message?board.id=doctor&message.id=23114&page=2I asked if the amount of resources effected the outcome of the formula used to calculate the power/charges, etc of your crafted item.

_________________

What if you have a schematic that requires unequal amounts of resources? For example several of the CM schemes require 4 non-ferrous, 10 fungi and 15 chemical. Do you just add up the OQ (or whatever) in equal amounts or do they get weighted?

Scenario #1 - (Non-ferrous+fungi+chem)/3000 = some percent

Scenario #2 - (((non-ferrous*(4/29))+(fungi*(10/29))+chem*(15/29)))/3000 = some potentially very different percent

_________________

The answer from Zarlor was a 99% no it does not matter.

In my quest to find confirmation of this, I travelled to the other crafting professions' boards and discovered the following FAQ on the Armorsmithing board - http://forums.station.sony.com/swg/board/message?board.id=armorsmith&message.id=20122&page=1. In this post, the author posts that the amounts of the resources do indeed matter.

So, I ask again - can anyone confirm whether either answer is correct?

I don't have confirmation, but I'm still 99% sure the number of resources used makes 0 difference. I could be wrong, of course, but my basic experimentation doesn't seem to suggest a difference.

Was nada last I checked as well.

1 unit or 40,00 units doesn't matter.

Components just add numbers right to the final item (including complexity).

Complexity affects your failure rate.

Higher failure rate = less good rolls on experimentation.

Now, Higher complexity items generally have:

1.) More amterials requirred
2.) Specific materials required.

These 2 things just make the item mosre costly to make, or in the case of #2 may cap an items power based on the possible stats for a material (For Ex: plumbum req. for E-11 has a max Conductivity (66% important on blasters) of like 150.. ever.

After that, it doesn't matter much.

-T

Why don't u test it by varying the resources and comparing to your formulas you've proposed? I'd like to hear the results.

It may be different for medicine crafting, but in armorsmithing and vehicle crafting, the quantities of the required resources are taken into account when determining starting/max experimental percentages. Note that I tested whether quantity affects experimental minimums and maximums using a vehicle schematic five minutes ago and I received the following results:

To test this the theory that the quantity of resources required of a particular type affects the minimum/maximum outcomes, I made up a swoop bike. The ingrediant ratio on a swoop is 4 ferrous to 1 nonferrous (I did not make any landspeeders which have a different ingrediant ratio). The ferrous metal I used had a SR of about 923 and the nonferrous was around 660.

A. If the ingrediants are, in fact, weighted, the final percentage should have come out to be around 87% ((4*923)+(1*660))/(5), giving an AWEQ of 870, which would result in a max experimental percentage of 87%. (See discussion below for what this means).

B. If the ingrediants are, in fact, not weighted, the final percentage should have come out to be around 79% (923+660/2), giving an average experimental quality of around 790, which would result in a max experimental percentage of 79%.

My vehicles ended up having a max experimental percentage of 87%, suggesting that hypothesis A, involving weighted resource qualities, is correct, at least for vehicles. I have noticed this same result for the armor types I've tested in the past.

Part VII: The mathmatics behind armorcrafting

This section is provided as a starting point for armorcrafters to understand a bit more about how they get the results they do. Since the actual equations behind crafting have not been released by the game developers, this section may not be fully accurate. However, it does describe the best understanding we have of how the system actually works.

In order to understand how the final result is produced, you first need to understand the basic formula. Essentially, the game calculates a minimum experimental statistic. This minimum experimental statistic is a function of the weighted average experimental quality. By spending experimentation points, you can increase this experimental statistic above the minimum at a fixed rate. The degree to which you can raise the experimental statistic is capped, a limit that is also a function of the weighted average experimental quality. The final experimental stastic is then mapped to an actual statistic in the final item.

A. How the weighted average experimental quality is calculated

This section describes how the weighted average experimental quality is calculated based on resource attributes. Essentially, for each experimental statistic there is weighted average quality that depends on a)the attributes of each resource used, b)the amount of each resource required by the schematic and c) the weighting of each attribute. Depending on the number of resources and attributes, this can be very simple (for example, in the case of batteries, only one resource and one attribute) or very complex (multiple resources and multiple attributes, each with different weights). In order to simplify the explanation, base resists will be used.

When you experiment on a piece of armor, you get three options, one of which is Experimental Quality. Experimenting on experimental quality raises the base resists of the item. However, in order to know how experimentation will effect the base item, you first need to calculate the weighted average quality of the item for that particular experimental statistic.

The weighted average quality for Experimental Quality on all armor is based on the Overall Quality and Shock Resistance at a 50-50 ratio. The shock resistance of each material multiplied by the quantity used divided by the sum of the total quantites of the materials used. This will give you the average shock resistance for the item. The same process is performed for Overall Quality and a weighted average of the two (50/50 in this case) is determined. For materials that do not have one of the values (for example solid petrochem fuel), the equation simply removes that factor from the calculation.

For example, say you're making a composite helm using the following materials (we're only going to list the OQ/Shock of these theoretical materials, since that's all applies to resistances):

Intrusive Ore 800 OQ 900 Shock Solid Petrochem Fuel 950 OQ Nabooian Fiberplast 700 OQ 900 Shock Aluminum 600 OQ 800 Shock Beyrllius Copper 700 OQ 400 Shock Wooly Hide 900 OQ 600 Shock

You're looking effectively using the following total quality of material...

((((70*800) + (70*950) + (35*700) + (40*600) + (30*700) + (30*900)) / (70 + 70 + 35 + 40 + 30 + 30 )) + (((70*900) + (35*900) + (40*800) + (30*400) + (30*600)) / (70 + 36 + 40 + 30 + 30)))) / 2

In the above equations, the average OQ would be 796 while the average SR would be 729. Consequently, the weighted average material rating for this experimental attribute would be 762.5.

The full equation for finding the weighted average attribute quality is described below:

General equation:

OQ1: Material 1 Overall Quality
SR1: Material 1 Shock Resistance
n1: Number of Material 1 required in schematic

OQ2: Material 2 Overall Quality
Material 2 does not have a Shock Resistance
n2: Number of Material 2 required in schematic

OQ3: Material 3 Overall Quality
SR3: Material 3 Shock Resistance
n3: Number of Material 3 required in schematic

( ( ((OQ1*n1) + (OQ2*n2) + (OQ3*n3)) ) + ( ((SR1*n1) + (SR3*n3)) ) )
(n1+n2+n3) (n1+n3)
_______________________________________________

( 2 )

B. Experimental minimums and maximums

Experimental minimums and maximums are directly related to the average weighted quality of the resources used. Currently, the benefit gained from using each experimentation point is fixed at 7%, regardless of resource, although this will almost certainly change at some point in the future.

1. Each point of experimentation placed into the object resulted in a 7% gain, regardless of material, as was expected.

2. There is a fairly clear linear relationship between average material quality and starting/final percentages. The fit is not perfect, suggesting that there is a more complicated relationship at work, but this fit is very close (R2 = 99, n=6), which is good enough for pratical purposes. Any deviations are in the fractions of single percents and won't affect our work.

Basic Equations:
starting experimental percentage y = 0.0003x - 0.0369
max experimental percentage y = 0.001x - 0.0012
incremental gain: .07 (7%)

Ultimate equation:
x = average material quality for the relevant attribute
EM>n = number of experimentation points spent

final attribute percentage = (0.0003x - 0.0369) + (n*(.07)), but capped at 0.001x - 0.0012

Note that depending on how many experimentation points you have and the resource qualities, you may have points left over that can be spent in other areas.

C. Mapping experimental percentages to actual statistics

Here are the results of some tests that mapped percentages to actual base resistance statistics on various armor segments. Note that resource attributes are immaterial at this stage- all that matters is the final experimentation statistic. All that resource attributes do is adjust the starting point and the maximum up or down.

a. maberi armorweave segment:
11% - 1% resistance
25% - 2% resistance
53% - 3% resistance

b. composite armorweave segment
23% - 3% resistance
37% - 4% resistance
51% - 5% resistance
59% - 6% resistance>
73% - 7% resistance

Note that these are not absolute cut off points. They are limited by the starting points offered by the particular resource I was using. Using resources with different qualities will allow us to hone in on where the actual cut-off points are, which will enable us to better spend experimentation points (i.e.: not spending that last experimentation point in resistance, for example, if you know it will not put you over the cutoff to the next % of resistance). Of course, this assumes that there are discrete intervals of resistance. It is possible that there are decimals involved as well, but that they are hidden behind the scenes.

Thanks, Jaato! Maybe some more testing on Meds shuld be done, but for the most part the crafting system is the crafting system, so I would expect the same to hold true for Meds.

Well this throws a little wrinkle into things doesn't it?

I would love to try to test this myself, but I don't have enough experimentation points to maz anything out to test the limits. Any volunteers?

I'd really appreciate it if you'd go ahead and test the system I described out and see if it works for meds. Over in armorsmithing we've done a bit of testing, but there's always a lot of room for improvement

Oh,

I see.. Yeah, there is a miscomminication here.

yes, the ratio of materials to each other does play a role

16 800 OQ materials will weigh more heavily then 5 300 OQ materials.


What I though you were talking about was amount..
Where an item with 500 metal requirement, would get a better experimental percentage then a 60 metal item... while using the same metal.

I dunno where my brain was. I was answering a different question.

Be aware that this may change a bit soon, since, no matter the amount, a material with no mention of a attribute should be counted as a zero (a weighted 0 by material count), but is not currently being counted at all.

-T

There's a heavy debate over that change Traigus, the reason being that a lot of schematics call for specific resources that don't have particular attributes- for example, several armor components call for various gasses, but have experimental attributes that are based on Shock Resistance, Unit Toughness and Malleability, which gasses just don't have (some might have mall, I can't remember). The problem is, this makes it so that it is no longer even theoretically possible to reach 100% experimentation, even if you are using materials that have 1000's in all available stats. For example, if a particular schematic calls for 10 units of metal and 10 units of gas, and the relevant attribute is shock resistance, the WAEQ (Weighted Average Experimental Quality) is 500 (1000+0/2) and the maximum experimental percentage you can get is 500. It is doubtful that this change would actually go into effect, but you never know.

There's another change in the works, one that changes the benefit you gain from each experimentation point to be based on the WAEQ rather than being a fixed amount (currently 7%). The problem with this is that it, as it was initially envisioned, it would function as a complete and total nerf to everyone across the board.

It's great to see all the testing that's been done in terms of the armorsmithing, but I'm afraid things are just a little different in the medical world. In our schematics, unlike an armorsmith's, almost all of them take equal amounts of each resource required.

The ONLY schematics for a doctor that I know of that vary from this are the Stim D's and Stim E's. If any testing is to be done, it should be done on those. Perhaps, if I get some time tonight, I'll check it out.

Yeah, I guess most of ours are set up for equal amounts, which would certainly explain why I've never noticed a difference like that before.

As for the issues with the schematics, experimentation and 0-rated resources, if anyone is familiar with those issues it's Traigus. He brought up a lot of the problems with those when the Devs were considering some of those changes on TC. That's why they were pulled to begin with is because of Corres like Traigus. As far as we know they are still looking into all of that, so some of that stuff may yet resurface, but at least we've made them fully aware of the problem with introducing that 0 affects ALL of our schematics.

The word we have, is that there should be a material in game that will supply all attributes to all stats before the 0 change can possibly go in.

We have no word on which direction the devs want to go with on this... since it is important to ALL crafters.

Could end up being a pass on schematics, or it could end up being a pass on material attributes (adding and tweaking). Could be both.

Be assured, it would not go in until it can actually work... this isn't just us... it is ALL of crafting.

from talking to a lto of other Correspondents on this issue, and looking at my own schematics in a lot of professions... it seems this got borked about the time they simplified all the schematics/materials... and it is everywhere.

It is also tied to the "capped" quality materials... (All plumbum iron has crappy (capped in the 100-200 range) conductivity, even though it is required for a huge chunk in the final combaine of 4-6 blasters (which also rate 66% conductivity on the final combine).

This is why all our general meds come out better(percentage-wise) then our specific material meds. It is actually more effective for us to put a material in with NO attrubute, then it is to put in a material with a LOW attribute.

Yeah, our stuff is pretty even as materials go... 8,8 16,16 etc. usually.

-T

Message Edited by Traigus on 12-18-2003 01:34 PM