These forums have been archived and are now read-only.

The new forums are live and can be found at https://forums.eveonline.com/

Science & Industry

 
  • Topic is locked indefinitely.
Previous page12
 

Whats the true formula for waste now?

First post First post
Author
CCP Nullarbor
C C P
C C P Alliance
#21 - 2014-07-19 17:39:35 UTC
Qoi wrote:
required = MAX(Number of Runs, CEIL( baseAmount * Number of Runs * MaterialModifier))


Confirming this is the correct formula with MaterialModifier = 1 - (ME / 100)

Any additional modifiers due to teams / facility bonuses / whatever would be multiplied by the base amount and runs BEFORE doing a CEIL to round it out to an integer. This is a significant difference from the previous industry system, where more runs may decrease your required materials per run by a noticeable amount.

The new UI will show you all of this information live by scrolling the runs up and down, or switching between facilities / teams etc.

Or you can calculate it all yourself with a spreadsheet of course Bear

CCP Nullarbor // Senior Engineer // Team Game of Drones

Steve Ronuken
Fuzzwork Enterprises
Vote Steve Ronuken for CSM
#22 - 2014-07-19 18:16:35 UTC  |  Edited by: Steve Ronuken
Just to go over this:

There is no addition to the materials that CCP have released in the SDE (if you want them in a spreadsheet, you can grab one from the SDE directory on my site) Everything is subtractive (well, other than rapid assembly arrays.)

If you have a -5% facility, it's still worth researching to ME -10 (well, depending on the way the materials break down.) as this will lead to -14.5% materials ( 0.9 multiplier, and a 0.95 multiplier)

It's also worth knowing that rounding (it rounds up. so 1.1 becomes 2) happens at the job level, not the run level. So if you have something which uses 1.1 things, a job with a single run would use 2. a job with 10 runs would use 11.

(bah. slow to get this out, due to waiting for getting sisi updated, to confirm it.)

Woo! CSM XI!

Fuzzwork Enterprises

Twitter: @fuzzysteve on Twitter

Salpun
Global Telstar Federation Offices
Masters of Flying Objects
#23 - 2014-07-19 18:25:10 UTC
Steve Ronuken wrote:
Just to go over this:

There is no addition to the materials that CCP have released in the SDE (if you want them in a spreadsheet, you can grab one from the SDE directory on my site) Everything is subtractive (well, other than rapid assembly arrays.)

If you have a -5% facility, it's still worth researching to ME -10 (well, depending on the way the materials break down.) as this will lead to -14.5% materials ( 0.9 multiplier, and a 0.95 multiplier)

It's also worth knowing that rounding (it rounds up. so 1.1 becomes 2) happens at the job level, not the run level. So if you have something which uses 1.1 things, a job with a single run would use 2. a job with 10 runs would use 11.

(bah. slow to get this out, due to waiting for getting sisi updated, to confirm it.)

Hit up the Singularity channel if you got stuff the rest of us can help test.

If i dont know something about EVE. I check https://wiki.eveonline.com/en/wiki/ISK_The_Guide

See you around the universe.

Knowledgeminer
Oriens Vis
#24 - 2014-07-26 08:55:00 UTC
CCP Nullarbor wrote:
Qoi wrote:
required = MAX(Number of Runs, CEIL( baseAmount * Number of Runs * MaterialModifier))


Confirming this is the correct formula with MaterialModifier = 1 - (ME / 100)


According to that formula, producing 60 Large Sentry Damage Augmentor I with an ME 10 blueprint should require:

5616 Broken Drone Transceiver
6156 Burned Logic Circuit
6318 Tripped Power Circuit

right?

However, that’s not what I get. What I get is:

5617 Broken Drone Transceiver
6157 Burned Logic Circuit
6318 Tripped Power Circuit

I’ve observed similar discrepancies happen with other BPs as well. Is there an explanation for this? Is it a floating point precision issue or what?
Qoi
Exert Force
#25 - 2014-07-26 09:31:24 UTC  |  Edited by: Qoi
Knowledgeminer wrote:
CCP Nullarbor wrote:
Qoi wrote:
required = MAX(Number of Runs, CEIL( baseAmount * Number of Runs * MaterialModifier))


Confirming this is the correct formula with MaterialModifier = 1 - (ME / 100)


According to that formula, producing 60 Large Sentry Damage Augmentor I with an ME 10 blueprint should require:

5616 Broken Drone Transceiver
6156 Burned Logic Circuit
6318 Tripped Power Circuit

right?

However, that’s not what I get. What I get is:

5617 Broken Drone Transceiver
6157 Burned Logic Circuit
6318 Tripped Power Circuit

I’ve observed similar discrepancies happen with other BPs as well. Is there an explanation for this? Is it a floating point precision issue or what?

Your maths is correct. I see the same numbers and i cannot explain the discrepancy. Sounds like i will have to amend my pre-crius CCP_ROUND illustration https://c.1tau.de/public.php?service=files&t=766a35c75275b1b14a7899bcd69e50b6 UghUgh

Unfortunately i have no ME 10 Large Sentry Damage Augmentor I blueprints, can someone confirm this? :/

http://eve-industry.org

Qoi
Exert Force
#26 - 2014-07-26 09:39:16 UTC
It DEFINITELY looks like a floating point error, since CCP is rounding 5616.0 and 6156.0 to 5617 and 6157, but 6318.0 to 6318.

Did i mention that i hate floating point arithmetics Cry

http://eve-industry.org

Qoi
Exert Force
#27 - 2014-07-26 10:13:15 UTC
I was just playing around with python
Quote:

>>> math.ceil(104*0.9*60)
5617.0
>>> math.ceil(104*60*0.9)
5616.0
>>> math.ceil(114*0.9*60)
6157.0
>>> math.ceil(114*60*0.9)
6156.0
>>> math.ceil(117*0.9*60)
6318.0
>>> math.ceil(117*60*0.9)
6318.0


Hey CCP, how about changing the order of multiplication.. :D

http://eve-industry.org

CCP Nullarbor
C C P
C C P Alliance
#28 - 2014-07-27 13:31:28 UTC
Qoi wrote:
I was just playing around with python
Quote:

>>> math.ceil(104*0.9*60)
5617.0
>>> math.ceil(104*60*0.9)
5616.0
>>> math.ceil(114*0.9*60)
6157.0
>>> math.ceil(114*60*0.9)
6156.0
>>> math.ceil(117*0.9*60)
6318.0
>>> math.ceil(117*60*0.9)
6318.0


Hey CCP, how about changing the order of multiplication.. :D


Yay floating point math Bear

I will have a look at throwing in a round( quantity, 2 ) before doing the ceil, just so we can all keep our sanity.

CCP Nullarbor // Senior Engineer // Team Game of Drones

Qoi
Exert Force
#29 - 2014-07-27 23:55:25 UTC
CCP Nullarbor wrote:
Qoi wrote:
I was just playing around with python
Quote:

>>> math.ceil(104*0.9*60)
5617.0
>>> math.ceil(104*60*0.9)
5616.0
>>> math.ceil(114*0.9*60)
6157.0
>>> math.ceil(114*60*0.9)
6156.0
>>> math.ceil(117*0.9*60)
6318.0
>>> math.ceil(117*60*0.9)
6318.0


Hey CCP, how about changing the order of multiplication.. :D


Yay floating point math Bear

I will have a look at throwing in a round( quantity, 2 ) before doing the ceil, just so we can all keep our sanity.

You sir are worth your volume in Compressed Hedbergite. Big smile

http://eve-industry.org

Kniht
#30 - 2014-07-29 14:02:51 UTC  |  Edited by: Kniht
Qoi wrote:
You sir are worth your volume in Compressed Hedbergite. Big smile

Either that's a comment on someone's size relative to the (space-bloated) bodies in New Eden, or a significant under-valuation of the headaches of floating-point math.

2m3 / 0.14m3 * 65,340 ISK = 933,428 ISK
933,428 ISK * (PLEX / 796M ISK) * (15 USD / PLEX) = 1.8 cents

Gives a new meaning to putting your two cents in.

o/ fly crazy

Bronson Hughes
The Knights of the Blessed Mother of Acceleration
#31 - 2014-07-29 14:13:05 UTC
Kniht wrote:
Qoi wrote:
You sir are worth your volume in Compressed Hedbergite. Big smile

Either that's a comment on someone's size relative to the (space-bloated) bodies in New Eden, or a significant under-valuation of the headaches of floating-point math.

2m3 / 0.14m3 * 65,340 ISK = 933,428 ISK
933,428 ISK * (PLEX / 796M ISK) * (15 USD / PLEX) = 1.8 cents

Gives a new meaning to putting your two cents in.

I think you just won EvE....

Relatively Notorious By Association

My Many Misadventures

I predicted FAUXs

Qoi
Exert Force
#32 - 2014-07-29 16:11:00 UTC
Oops

http://eve-industry.org

Previous page12