open All Channels
seplocked Science and Industry
blankseplocked Which components of T2 production are subject to waste?
 
This thread is older than 90 days and has been locked due to inactivity.


 
Pages: [1] 2

Author Topic

Kane Tabor
Posted - 2009.02.15 19:39:00 - [1]
 

I have a tool I use to run the numbers on my T2 BPCs so I know what materials I need to have in stock to produce them, and it's always estimating a bit low for Nanoelectrical Microprocessors for the T2 remote armor reps. It started happening a while back (forget when), so I've always just thrown in a few extra of those items into the hanger. Today I sat down to figure out why.

Here's what I came up with: all the other items that the waste calculations apply to are in category 4, ie. the basic materials (trit, mex, etc.). No other items involved in the production of the T2 modules I'm working with have waste applied to them, except for the Nanoelectrical Microprocessors (at least as far as I can tell).

Now, I can work around this, but it seems odd to me that this one T2 component uses waste and no others (as far as I can tell) -- not even the other materials in group 334 (Construction Components).

Is this a bug, or are Nanoelectrical Microprocessors special in some way?

Avalon Champion
Black Thorne Corporation
Posted - 2009.02.15 20:54:00 - [2]
 

I've noticed this on other T2 module BPC's except that it appears to be across the board for all materials except the, additional mins that are calculated to the correct ME level.

Could be a bug, or working as intended, but im not sure.

Johnathan Roark
Caldari
The Graduates
Morsus Mihi
Posted - 2009.02.16 04:01:00 - [3]
 

It has to do with rounding. Your qty isn't high enough to matter.

Lord Fitz
Project Amargosa
Posted - 2009.02.16 13:22:00 - [4]
 

Edited by: Lord Fitz on 16/02/2009 13:47:47
nvm

Akita T
Caldari Navy Volunteer Task Force
Posted - 2009.02.16 13:25:00 - [5]
 

Edited by: Akita T on 16/02/2009 13:27:28

The only two things waste is not applied to, as far as I know, are RAMs and the T1 item.
There might be more, but... no idea.
Also, rounding is to nearest. So in order to have a base 1 something needed become a 2, you need at least 50.0[...]01% waste.

P.S. Get this here, enter your stats, and check if you get the same numbers.
If you don't get the same numbers, please post your stats, numbers you get in the sheet, and numbers you get in the manufacture line quote (NOT numbers on blueprint, but on the manufacture quote itself, the only 100% reliable numbers).

Avalon Champion
Black Thorne Corporation
Posted - 2009.02.16 20:59:00 - [6]
 

I do think that the calcs are broken for the BPC. Here is one example

Mag field Stab II, BPO (ME0)

Photon Micro (5)
Mechanical Parts (5)
Minature Electronics (2)

Mag field Stab II, BPC (ME-4) 50% waste on a BPO. The build requirements per run show as

Photon Micro (5)
Mechanical Parts (5)
Minature Electronics (2)

When they Should really be

Photon Micro (8)
Mechanical Parts (8)
Minature Electronics (3)

The Trit, Pyer, Morph, and Nocx calcs are all correctly calculated with a 50% Waste factor.

I've raised a bug report with CCP to make sure this isnt a bug, it also affects other mods as well, these include MAR II's, and LAR II's.

Mr Krosis
Gallente
The humble Crew
Posted - 2009.02.19 10:31:00 - [7]
 

Being subject to waste isn't an attribute of the material, it is a property of each entry on the blueprint. Each blueprint has 2 categories of materials, raw materials and extra materials. It shows you these two categories in the summary screen when you go to build something.

Anything in the raw materials category is subject to waste, anything in the extra materials category is not. It's that simple. As a general rule Tech II materials for modules are usually considered extra, while for ships they are considered raw. This came as quite a surprise to me as someone who invented and built T2 modules before I got around to ships. I was used to no T2 waste mats and thought it was the norm. My initial vagabond build cost projections were a tad low to say the least Very Happy.

Akita T
Caldari Navy Volunteer Task Force
Posted - 2009.02.19 13:41:00 - [8]
 

Edited by: Akita T on 19/02/2009 13:56:44

So where exactly in the database dump DO you find the attribute that says wether a material is "raw" or "extra" for each specific blueprint (or item) ?
I only ask because I can't find it... or see it but don't realize it's that.

Fubar
Posted - 2009.02.20 14:36:00 - [9]
 

Originally by: Akita T

So where exactly in the database dump DO you find the attribute that says wether a material is "raw" or "extra" for each specific blueprint (or item) ?
I only ask because I can't find it... or see it but don't realize it's that.

An attribute does not exist but you can determine whether a material is raw or extra by examining table typeActivityMaterials in the data dump.

The data we need to look at is activityID 1 for the blueprintTypeID and activity 6 for the productTypeID that the blueprint in question produces.

A material is raw if the activity 6 qty is greater than or equal to the activty 1 qty.

A material is extra if activity 6 does not exist for the activity 1 typeID.

Ignore any activity 6 typeIDs that don't have an activity 1 counterpart.

Now for the fun part....

In the case of bombs we have minerals that qualify as both raw and extra. Lets look at the Electron Bomb. blueprintTypeID=27921 productTypeID=27920.
typeID=27921 activityID=1
requiredType activity quantity
Tritanium 1 1362069
Pyerite 1 623994
Mexallon 1 60231
Isogen 1 9381
Nocxium 1 2370
Zydrine 1 1419
Megacyte 1 365

typeID=27920 activityID=6
requiredType activity quantity
Tritanium 6 10000
Pyerite 6 100
Mexallon 6 10
Isogen 6 10
Nocxium 6 10
Zydrine 6 10
Megacyte 6 10

The activity 6 quantities are smaller than the activity 1 quantities. So the activity 6 quantity is classified as raw and the activity 1 quantity minus the activity 6 quantity is classifed as extra. For example...
requireType  Quantity  Mineral Type  Waste Applied?
Tritanium 10000 Raw 1
Tritanium 1352069 Extra 0
Pyerite 100 Raw 1
Pyerite 623894 Extra 0
Mexallon 10 Raw 1
Mexallon 60221 Extra 0
Isogen 10 Raw 1
Isogen 9371 Extra 0
Nocxium 10 Raw 1
Nocxium 2360 Extra 0
Zydrine 10 Raw 1
Zydrine 1409 Extra 0
Megacyte 10 Raw 1
Megacyte 355 Extra 0

And in case you don't want to compile your own data here is mine.
Fubars_Manufacturing_Data.csv

Hope this helps.

Pwett
QUANT Corp.
QUANT Hegemony
Posted - 2009.02.20 14:46:00 - [10]
 

Edited by: Pwett on 20/02/2009 14:45:58
Back to the original mag stab II problem, there is no activityID 6


select *
from typeActivityMaterials
where typeID = 10191
and ( activityID = 1 OR activityID = 6)
order by activityID

10191 1 34 2275 1 0
10191 1 35 352 1 0
10191 1 38 2 1 0
10191 1 3380 5 0 0
10191 1 3689 5 1 0
10191 1 9842 2 1 0
10191 1 9944 1 1 1
10191 1 11399 3 1 0
10191 1 11452 1 0 0
10191 1 11453 1 0 0
10191 1 11483 1 0.15 0
10191 1 11540 5 1 0

Fubar
Posted - 2009.02.20 15:04:00 - [11]
 

Originally by: Pwett
Edited by: Pwett on 20/02/2009 14:45:58
Back to the original mag stab II problem, there is no activityID 6


Sure there is no activityID 6 for id 10191 (the blueprint) but there is for id 10190 (the finished product).
select * 
from typeActivityMaterials
where (typeID = 10191 or typeID = 10190)
and ( activityID = 1 OR activityID = 6)
order by activityID

10191 1 34 2275 1 0
10191 1 35 352 1 0
10191 1 38 2 1 0
10191 1 3380 5 0 0
10191 1 3689 5 1 0
10191 1 9842 2 1 0
10191 1 9944 1 1 1
10191 1 11399 3 1 0
10191 1 11452 1 0 0
10191 1 11453 1 0 0
10191 1 11483 1 0.15 0
10191 1 11540 5 1 0
10190 6 34 9520 1 0
10190 6 35 1665 1 0
10190 6 38 2 1 0

And then how about the Large Remote Armor Repair System II. bpID=26915 product=26914
select * 
from typeActivityMaterials
where (typeID = 26915 or typeID = 26914)
and ( activityID = 1 OR activityID = 6)
order by activityID
26915 1 34 566 1 0
26915 1 35 115 1 0
26915 1 36 244 1 0
26915 1 37 30 1 0
26915 1 3380 5 0 0
26915 1 9848 8 1 0
26915 1 11359 1 1 1
26915 1 11399 5 1 0
26915 1 11442 1 0 0
26915 1 11452 1 0 0
26915 1 11475 1 0.2 0
26915 1 11539 4 1 0
26914 6 34 4452 1 0
26914 6 35 3045 1 0
26914 6 36 245 1 0
26914 6 37 30 1 0
26914 6 11399 5 1 0
26914 6 11539 4 1 0

Nanoelectrical Microprocessors are typeID 11539


Pwett
QUANT Corp.
QUANT Hegemony
Posted - 2009.02.20 15:21:00 - [12]
 

Edited by: Pwett on 20/02/2009 15:23:01

hrm, interesting. so for every item we need to look at manufacturing from the blueprint and the duplicating values from the actual product? Which of course you said; it just didn't kick in before my coffee.

Avalon Champion
Black Thorne Corporation
Posted - 2009.02.20 20:42:00 - [13]
 


To to clarify, I did raise a petition, with regard the Mag Field Stab II, and MAR/LAR II issue, and the response i got was that it was working as intended.

Which means, I've been under estimating my profit margins on most things except ships.

Meha Mott
Minmatar
Carebear Research and Produktion Agency
Posted - 2009.02.21 02:01:00 - [14]
 

Try 100MN MWD. No wast on that module saves a lot of money if you invent it. Laughing

Kiay Stryx
Gallente
Phoenix Mandate
Phoenix Virtue
Posted - 2009.02.21 03:35:00 - [15]
 

Urgh.

~Stryx

Akita T
Caldari Navy Volunteer Task Force
Posted - 2009.02.21 03:37:00 - [16]
 

Originally by: Avalon Champion
Which means, I've been under estimating my profit margins on most things except ships.

Not by very much, though Wink

Avalon Champion
Black Thorne Corporation
Posted - 2009.02.21 12:31:00 - [17]
 

Originally by: Akita T
Originally by: Avalon Champion
Which means, I've been under estimating my profit margins on most things except ships.

Not by very much, though Wink



True, 10-20K per item on most mods is probably a good estimate.

Stupid question, i know, but where do i find information on the data dumps and API calls, as im planning on putting together a Db that loads all material costs, and keeps a history so i can track changes and spot trends.

Kane Tabor
Posted - 2009.03.27 03:47:00 - [18]
 

Assuming Fubar's post (http://www.eveonline.com/ingameboard.asp?a=topic&threadID=999934&page=1#9) was correct at the time, I think the formula may have changed in Apocrypha, or else I'm just not getting it. Let's look at Trit usage for the Electron Bomb: blueprintTypeID=27921 productTypeID=27920 (Trit is typeID=34)

From the Apocrypha data export:
select * from typeActivityMaterials where typeID=27921 and activityID=1 and requiredTypeID=34
select * from typeActivityMaterials where typeID=27920 and activityID=6 and requiredTypeID=34

Input Trit is 1362069, recycle amount is 10000. By Fubar's formula, there should be waste on 10000 of it, and no waste on the other 1352069. At ME 0 (10% waste: 1000), that should yield a construction amount of 1363069. However, the real construction amount is 1498276, which is exactly the amount predicted if the waste applies to the entire input amount (ME 0: 10% waste: 136207).

So, here's the question: am I misunderstanding Fubar's formula, was it always wrong, or did Apocrypha change it?

Here's a quick list of some of the affected blueprints and the material types on them affected:
Many T2 Missiles: Rocket Fuel
Depleted Uranium XL: Trit/Pyer
Discovery/Gaze Probes: Trit/Pyerite/Mex
ECCM Projector II: Morphite
Gas Cloud Harvester I: Trit/Pyer/Mex
Gyrostabilizer II: Morphite
Heat Sink II: Morphite
Mercoxit Mining Crystal II: Nocx
Tracking Computer II: Morphite
Tracking Link I: Trit/Pyer/Mex/Iso
Warp Core Stabilizer II: Morphite

Akita T
Caldari Navy Volunteer Task Force
Posted - 2009.03.27 07:11:00 - [19]
 

Originally by: Kane Tabor
So, here's the question: am I misunderstanding Fubar's formula, was it always wrong, or did Apocrypha change it?

To be honest, that's what I'd like to know too Razz

Fubar
Posted - 2009.03.27 15:24:00 - [20]
 

Edited by: Fubar on 27/03/2009 15:26:59

Originally by: Kane Tabor
So, here's the question: am I misunderstanding Fubar's formula, was it always wrong, or did Apocrypha change it?


Basically it comes down to you can't trust the show info bill of materials.

See Linkage. The picture shows the show info window next to the quote window.

The bp does still requires 1363069 (1352069+10000+1000) of trit per batch on a ML0 bp.

Edit: Corrected the last line.

Pwett
QUANT Corp.
QUANT Hegemony
Posted - 2009.03.27 16:23:00 - [21]
 

Aye Fubar's formula is indeed still correct. Ignore show info for specific values.

Kane Tabor
Posted - 2009.03.27 17:55:00 - [22]
 

Originally by: Fubar
Basically it comes down to you can't trust the show info bill of materials.

Odd. Though I can't say I'm surprised the algorithm in one place in the game is wrong -- it's quite a bit more complicated than one would think it should be.

Thanks for the confirmation Fubar -- I had just been relying on the show info window -- I never thought to see if the quote window matched it.

Dav Varan
Posted - 2009.04.25 00:47:00 - [23]
 

Edited by: Dav Varan on 26/04/2009 23:39:15

Epitrope
The Citadel Manufacturing and Trade Corporation
Posted - 2009.04.28 04:52:00 - [24]
 

Originally by: Fubar
...you can determine whether a material is raw or extra by examining table typeActivityMaterials in the data dump.

The data we need to look at is activityID 1 for the blueprintTypeID and activity 6 for the productTypeID that the blueprint in question produces.


How do you square this with types such as the Crystalline Carbonide Armor Plate (typeID 11545, blueprintTypeID 17323)? 11545 has no entries in the typeActivityMaterials table, but I can confirm that both of the types required to build it are affected by waste. Kane Tabor's idea of basing it on category 4 is good, and perhaps better than what I was doing (individually assigning all the composites to be base instead of extra), but I'm a little bit uncomfortable with just mangling the data like that.

Fubar
Posted - 2009.04.28 11:06:00 - [25]
 

Originally by: Epitrope
How do you square this with types such as the Crystalline Carbonide Armor Plate (typeID 11545, blueprintTypeID 17323)? 11545 has no entries in the typeActivityMaterials table, but I can confirm that both of the types required to build it are affected by waste. Kane Tabor's idea of basing it on category 4 is good, and perhaps better than what I was doing (individually assigning all the composites to be base instead of extra), but I'm a little bit uncomfortable with just mangling the data like that.


Good point.

After you use the previous method you run an update query that updates the waste on the Construction Components and Advanced Capital Construction Components blueprints.

Dav Varan
Posted - 2009.04.28 13:33:00 - [26]
 

Edited by: Dav Varan on 28/04/2009 13:36:39

I found that applying the rule of Fubar works.
But only apply it if there exists at least one T.a.m.6 for the product being considered.

This stops ( component | advanced capital components | Capital components )items becoming all extra material.

Epitrope
The Citadel Manufacturing and Trade Corporation
Posted - 2009.04.29 10:03:00 - [27]
 

I looked at a few more items on the test server, and have refined the rule somewhat.

First, what I'm talking about here are blueprints whose products have no activityID=6 entries in the typeActivityMaterials table. This is to try to figure out which of the materials for these blueprints are affected by waste, because the data seems not to be in the database export.

The category=4 rule does not always work: the new T3 component blueprints require materials, but they are not affected by waste. See, for example, Electromechanical Interface Nexus

Also, it's not only composites that are left out, minerals seem to occasionally be affected too: Mobile Medium Warp Disruptor I

Based on this, I'm going to say that if a blueprint's product has no activity 6 data, all the composites (groupID=429) and all the minerals (groupID=18) are affected by waste. However, this may still be inaccurate. I'm only spot checking likely candidates, so I might have missed something, and I'm checking these on sisi (build 87385), which might have different data. Corrections welcome...

How does this look?

Shawn Gallentino
Posted - 2009.05.15 00:49:00 - [28]
 

Wow, All of this complexity and I still have questions. This one is simple though. At which point in the madness does the Assembly's material modifier take effect? Does it take effect prior to the material mat (i.e. Base Material * Assembly Mod) or is it after (Final Material * Assembly Mod).

Qoi
Exert Force
Posted - 2009.05.15 21:10:00 - [29]
 

Edited by: Qoi on 15/05/2009 21:10:00
Originally by: Shawn Gallentino
Wow, All of this complexity and I still have questions. This one is simple though. At which point in the madness does the Assembly's material modifier take effect? Does it take effect prior to the material mat (i.e. Base Material * Assembly Mod) or is it after (Final Material * Assembly Mod).


It doesn't matter?
Linkage

Epitrope
The Citadel Manufacturing and Trade Corporation
Posted - 2009.05.16 01:35:00 - [30]
 

Originally by: Qoi
Edited by: Qoi on 15/05/2009 21:10:00
It doesn't matter?
Linkage


It does matter slightly, because at the end you have to round to the nearest unit. If you round to an int before multiplying by the assembly line's bonus you'll get subtly wrong amounts. You need to include the assembly line multiplier at the same time as skill waste and ML waste:

Quote:
if (ML >= 0) {
return runs * int(0.5 + qty * (1.0
+ (.25 - .05 * PE)
+ (curItemWasteFactor/100.0) / (1.0 + ML)
+ (lineMaterialBonus - 1.0)
));
} else {
return runs * int(0.5 + qty * (1.0
+ (.25 - .05 * PE)
+ (curItemWasteFactor/100.0) * (1.0 - ML)
+ (lineMaterialBonus - 1.0)
));
}



(sorry for the appearance of that, it's all lined up prettily but there seems not to be a <pre> or equivalent for me to wrap it in)


Pages: [1] 2

This thread is older than 90 days and has been locked due to inactivity.


 


The new forums are live

Please adjust your bookmarks to https://forums.eveonline.com

These forums are archived and read-only