These forums have been archived and are now read-only.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

Invention Chance of Success

Author
Iron Breaker
Republic Military School
Minmatar Republic
#1 - 2013-08-14 02:34:01 UTC
Since the last update I seem to be failing to invent a LOT. Normally I would fail about half the time, but lately I seem to fail 75% of the time, and the prints that do succeed only have one run!

I was thinking of investing the time to get level 5 on my data core skills, but most of the calculators really make the 18 day training time seem like a waste... is it true that going from level 4 to level 5 only gives me a .008% improved chance of success? That just seems like a tough pill to swallow for 18 days... is it really worth it?


Thanks,
RubyPorto
RubysRhymes
#2 - 2013-08-14 02:45:11 UTC
Iron Breaker wrote:
Since the last update I seem to be failing to invent a LOT. Normally I would fail about half the time, but lately I seem to fail 75% of the time, and the prints that do succeed only have one run!


RNG is R
Number of runs and ME/PE is wholely dependent on the input BPC number of runs and decryptors.

Quote:
I was thinking of investing the time to get level 5 on my data core skills, but most of the calculators really make the 18 day training time seem like a waste... is it true that going from level 4 to level 5 only gives me a .008% improved chance of success?


That's how the skill system works.

Quote:
That just seems like a tough pill to swallow for 18 days... is it really worth it?


Up to you.

"It's easy to speak for the silent majority. They rarely object to what you put into their mouths." -Abrazzar "the risk of having your day ruined by other people is the cornerstone with which EVE was built" -CCP Solomon

Zhilia Mann
Tide Way Out Productions
#3 - 2013-08-14 03:04:18 UTC
Got any real numbers to play with? I'm willing to bet your failure rate isn't that extraordinary. But if it were, it should be pretty obvious. You'd just need a large sample size.
Tauranon
Weeesearch
CAStabouts
#4 - 2013-08-14 03:12:48 UTC
Iron Breaker wrote:
Since the last update I seem to be failing to invent a LOT. Normally I would fail about half the time, but lately I seem to fail 75% of the time, and the prints that do succeed only have one run!

I was thinking of investing the time to get level 5 on my data core skills, but most of the calculators really make the 18 day training time seem like a waste... is it true that going from level 4 to level 5 only gives me a .008% improved chance of success? That just seems like a tough pill to swallow for 18 days... is it really worth it?


Thanks,


I have never bothered. I also don't invent stuff unless there is a pretty decent profit for "expected" success rate invention.

If you train more skills to iv, rather than one skill to V (which is the trade off most of us make), then you get more items to search amongst for current market conditions offering a big profit at "expected" rate. Bear in mind that the skills are 20mil each or whatever and not all of them will be profitable purchases on that basis, ie you still probably need to investigate what items the skill gives you.



Havoc Zealot
Ministry of War
Amarr Empire
#5 - 2013-08-14 03:28:36 UTC
I go through the same problem...some sets i go from 1/15 and others i will go 15/20...All in all if you have a large sample size it should come out to be right around %50 success for modules.

as for your run amount, you need a Max run BPC to get a max run invention.

for the skill training it all comes down to opportunity which is really what manufacturing and trade is all about.

Hope that helped
Kraillach
Deep Core Mining Inc.
Caldari State
#6 - 2013-08-14 05:37:57 UTC
this is purely from personal experience...

i have had 3 chars runnin the whole process from invention to production.

now i have found that if i ran 10 slots per char comprising of invention for 2 different modules (5 invent jobs per module) i would get a success rate of between 40-60% per each job completion. 4-6 t2 bpcs.

but if i ran 10 jobs for each module (extending it over 2 days for the 6 modules) then my success rate would be 50 - 70%. 5 - 7 t2 bpcs.

i found that at least 50% of the time i would get the 7 bpcs out of the 10 runs.

so it was worth it to me to fill all lines with the one module to possibly get the extra print that would add an extra 20mil prof for the 10 items.

have all invention skills trained to 4 across the 3 chars.

Elena Thiesant
The Scope
Gallente Federation
#7 - 2013-08-14 06:56:01 UTC
Iron Breaker wrote:
Since the last update I seem to be failing to invent a LOT. Normally I would fail about half the time, but lately I seem to fail 75% of the time, and the prints that do succeed only have one run!


The old 'random is random' and 'cognitive bias' applies (you remember the unusual more than the usual). Depending on what you're inventing, 50% success rate may well be above the expected average and hence you shouldn't expect to get it all the time.

If your T2 BPCs have only one run, then it means that either you're inventing ships/rigs without a decryptor (in which case 1 run is all you'll get) or if you're inventing modules/drones/ammo then you're not using max-run T1 BPCs. The number of runs is not random, it's based on what you're inventing, the number of runs on the T1 BPC and the decryptor if any.
Casta Lyron
Alpha-Helix Biononics
#8 - 2013-08-14 08:23:03 UTC
Well, I am slightly surprised you would even do invention without researching the mechanics properly, you could end up throwing a lot of ISK right down the drain.

First of all it matters a lot what you are inventing, normal modules, for example, have a base chance of 40% while all ships will have considerably less depending on the type of ship. Second of all you need to run calculations on using decyptors and metalevel items to boost you invention chance. Even with increased chance of invention greater profitability is not ensured as you will have to account for cost of decryptors, decryptor effect on ME/PE and so on.

Skilling past 4 on datacore specific skills is really not recommended unless you have absolutely nothing else to skill up.
Steve Ronuken
Fuzzwork Enterprises
Vote Steve Ronuken for CSM
#9 - 2013-08-14 09:18:38 UTC
with module invention, no meta items, no decryptor:

level 4 skills across the board: 48.25%
level 5 encryption skill, L4 DC: 48.72%
L4 encryption, 1 L5 DC: 49.08%
L4 encryption: 2 L5 DC: 49.92%
All 5s: 50.4%
As before, but with a Meta 4 item:

level 4 skills across the board: 74.88%%
level 5 encryption skill, L4 DC: 75.6%
L4 encryption, 1 L5 DC: 79.03%
L4 encryption: 2 L5 DC: 83.2%
All 5s: 84%


If you want to play with it: https://www.fuzzwork.co.uk/invention/ is a stand alone version of the one in my blueprint calculator. It just hasn't been updated for the new decryptors. The rest of the math is still valid.

https://www.fuzzwork.co.uk/2013/05/23/manufacturing-201-tech-2/

Woo! CSM XI!

Fuzzwork Enterprises

Twitter: @fuzzysteve on Twitter

Termy Rockling
Sebiestor Tribe
Minmatar Republic
#10 - 2013-08-14 20:12:56 UTC
Havent you heard they nerf invention and other proc success rates in every mmo after each patch, its a wonder theres any success left anymore Pirate
Ireland VonVicious
Vicious Trading Company
#11 - 2013-08-14 21:23:46 UTC
Same thread over and over and over and over. Year after year after year after year. Roll
JohnPaulJones
State War Academy
Caldari State
#12 - 2013-08-15 17:17:22 UTC
Yes lvl 5 skills make a diffference. One thing to know,do not mix success chances. Do not mix high chance items with low chance items,if you deviate from this expect heavy losses. Pick an item and run with it and while you are at it,use every invention slot you have. Use another toon for other chance items.
Steve Ronuken
Fuzzwork Enterprises
Vote Steve Ronuken for CSM
#13 - 2013-08-15 17:34:15 UTC
JohnPaulJones wrote:
Yes lvl 5 skills make a diffference. One thing to know,do not mix success chances. Do not mix high chance items with low chance items,if you deviate from this expect heavy losses. Pick an item and run with it and while you are at it,use every invention slot you have. Use another toon for other chance items.



Random is random. The chance of one doesn't affect the chance of another.

Sure, patterns emerge, but people see faces in clouds and toast. We're /wired/ to look for patterns, even when those patterns don't exist.

It's a survival trait.

Woo! CSM XI!

Fuzzwork Enterprises

Twitter: @fuzzysteve on Twitter

RubyPorto
RubysRhymes
#14 - 2013-08-15 17:52:38 UTC
Steve Ronuken wrote:
Random is random. The chance of one doesn't affect the chance of another.

Sure, patterns emerge, but people see faces in clouds and toast. We're /wired/ to look for patterns, even when those patterns don't exist.

It's a survival trait.



That said, it's not terrible advice. Sticking to multiples of your invention slots and running monolithic blocks of inventions makes it a lot easier to keep track of what decryptors and datacores are in what labs.


But yeah, stare at a stucco wall for an hour or two and you'll find faces. Stare into a mirror for an hour or two, and you'll never go to the bathroom in the dark again. The human mind is a freaky thing.

"It's easy to speak for the silent majority. They rarely object to what you put into their mouths." -Abrazzar "the risk of having your day ruined by other people is the cornerstone with which EVE was built" -CCP Solomon