These forums have been archived and are now read-only.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

Level 5 invention skills

Author
mark3mog
Wrath of Mjolnir
#1 - 2013-07-09 11:28:53 UTC
JF prices are pretty nicely priced so I thought I'd build a few. Buy the tech1 BPC and try inventing the T2 i.e. the jump frieghter. Now there are risks with invention and you chance in losing everything but the higher you train the more chance of success and as inventing a jump frieghter takes quite some investment I thought it was probably best to train to level 5 in the invention skills. As i was trying to invent an Anshar BP the invention skills go like this:

Gallentean Starship Engineering 29 Days

Gallente Encryption Methods 29 Days

Mechanical Engineering 29 Days

Total 87 Days

Rough training times with top implants and tuned training attributes.

Once I trained all the skills I put 5 tries in the lab and was hoping at least 3 would take. All 5 succeeded and I thought "great, well worth the training". However the next time I tried to invent I had zero successes, everything was exactly the same, same station, same try total, everything. Can't be right so I raised a petition and was told "sorry but there is a risk", I pointed out that I had top skills and this shouldn't happen but everything was working as it should and I'd just been unlucky.

Now I've tried inventing at a lower skill and I have never had this happen before. My friends have done the same and have never had a total failure. Now I am just going to have to lump it but I'd thought I'd spread the word so others don't waste so much training time. My advise would be to stop at level 4 and not bother training level 5.
Elena Thiesant
The Scope
Gallente Federation
#2 - 2013-07-09 11:55:08 UTC  |  Edited by: Elena Thiesant
Random is random. Level 5 skills give you a slightly better chance, that's all. There's no guarantee of success.

Inventing an Anshar with maxed out skills and the +3ME decryptor = 41.58%
Inventing an Anshaw with all level 4 skills and the +3ME decryptor = 39.81%

You had 5 succeed then 5 fail so that's a 50% successes which means you're above the expected average. Be grateful.
The chances of success are independent, one failing does not mean that the next has a greater chance of success. Over time and invention attempts, the success rate will tend towards the expected chance.

p.s. I had 20 cruiser inventions in a row fail last week. It happens.
Raban Sardipan
Imperial Academy
Amarr Empire
#3 - 2013-07-09 12:55:16 UTC
It's about 5.5 times more likely you will fail 5/5 than succeed 5/5 with the probabilities posted above (quick napkin math), and close to a 7% chance to fail 5 in a row. So... yah, random working as intended.

Anecdotally, with all 4s I failed 10/10 on some module or other last week (too many inventions to remember, but unusual enough that I remember the occurrence). Rare, but it happens.
Steve Ronuken
Fuzzwork Enterprises
Vote Steve Ronuken for CSM
#4 - 2013-07-09 14:05:30 UTC
Level 5 skills have more of an effect when you're using a meta module with the invention. Not good for ships, of course.

But it can make a huge difference with modules with cheap meta4s available. 8% improvement (or so)

Woo! CSM XI!

Fuzzwork Enterprises

Twitter: @fuzzysteve on Twitter

Ginger Barbarella
#5 - 2013-07-09 15:23:53 UTC
Nothing bugs me more than the usual "random is random" replies here, but let's face it, that's the way it works with Invention. I've been tracking every invention job I've done for the last 18 months or so (I've missed one or two, but still), and my overall success rate across the board (modules, rigs, low-end ships, etc *except* for Reverse Engineering) is right now about 49.9%. Reverse Engineering is a flat 50%. Those numbers include one alt with 5's on Mech Engr & Elec Engr, a few others with all 4's, and using base items for a few select types of stuff (launchers, projectiles, hybrids, and AB/MWDs).

"Blow it all on Quafe and strippers." --- Sorlac

Scarlett LaBlanc
University of Caille
Gallente Federation
#6 - 2013-07-10 00:11:43 UTC
mark3mog wrote:
My advise would be to stop at level 4 and not bother training level 5.



OP, you have learned one of the common accepted truths of invention. The level V skills just don't make much sense for the few extra percent of success they provide. It's unfortunate that this wisdom cost you so much training time.

Perhaps your thread will help someone else avoid the same outcome.
Tau Cabalander
Retirement Retreat
Working Stiffs
#7 - 2013-07-10 00:25:53 UTC
I emptied the market of Charon BPC for about 3 months straight at one point.

I failed inventing every single time for one of those months. I don't want to think about how many billions that was.

I eventually came to my senses. I don't build JF anymore.
RubyPorto
RubysRhymes
#8 - 2013-07-10 01:05:37 UTC
With all 5s and a +10% decryptor (that gives you a -1/-1 BPC), the probability of success on any given attempt is .416

That means the probability of each set of successes out of 5 attempts are as follows:
0/5: .07
1/5: .24
2/5: .34
3/5: .25
4/5: .09
5/5: .012

0/5 is about 6 times more likely than 5/5.


I do want to not that you weren't in a hurry to petition with an "is something wrong" when you benefited from an unlikely set of results.

"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

mark3mog
Wrath of Mjolnir
#9 - 2013-07-10 08:40:47 UTC
I didn't petition straight away as I was annoyed but knew the risks. I spoke to a friend and mentioned it. He said that can't be right and you should raise a petition as there may have been a problem. I saw his point. I knew that there wasn't a massive percentage increase in success but 60 days training wasted. Hopefully, others can learn from this.
Logical Chaos
Very Italian People
The Initiative.
#10 - 2013-07-10 09:58:12 UTC
I hear onyxia deep breaths more often since the last patch!
Sir SmashAlot
The League of Extraordinary Opportunists
Intergalactic Conservation Movement
#11 - 2013-07-10 15:30:37 UTC  |  Edited by: Sir SmashAlot
Chance has no memory. Sadly people find "patterns". After 100+ invention runs your success rate will converge with whatever the invention formula says.
Joan Greywind
The Lazy Crabs
#12 - 2013-07-11 04:25:27 UTC  |  Edited by: Joan Greywind
mark3mog wrote:
I didn't petition straight away as I was annoyed but knew the risks. I spoke to a friend and mentioned it. He said that can't be right and you should raise a petition as there may have been a problem. I saw his point. I knew that there wasn't a massive percentage increase in success but 60 days training wasted. Hopefully, others can learn from this.


Well you got your explanation, but you still wanted to "petition". Random is Random is the perfect answer and it explains everything. If you can't understand that, that is your problem, not CCP's. Dont encourage people not to train skills based on your stupidity. You didn't seem to know the "risks" because you clearly didn't understand them.

You have to think about like this; is 60 days (maybe equivalent to a billion in plex) worth the 2 percent increase on on my BPC?? If the 2 percent in saving is more than a billion then go for it, if not then you shouldn't. And remember this, it is only an "excpected" 2% increase in chance, could be more could be less, but over the long run it should be close to a 2% increase.

"Petition", what a scrub.
ExcalibursTemplar
CANZUK
#13 - 2013-07-11 13:11:55 UTC
So training invention skills to L5 is a waste if time ?

I know nothing about invention but i'm looking to get into it.
Steve Ronuken
Fuzzwork Enterprises
Vote Steve Ronuken for CSM
#14 - 2013-07-11 13:49:54 UTC
ExcalibursTemplar wrote:
So training invention skills to L5 is a waste if time ?

I know nothing about invention but i'm looking to get into it.



Not a waste. but live most L5 skills, the pay back is relatively low.

Woo! CSM XI!

Fuzzwork Enterprises

Twitter: @fuzzysteve on Twitter

ExcalibursTemplar
CANZUK
#15 - 2013-07-11 14:20:20 UTC
Steve Ronuken wrote:
ExcalibursTemplar wrote:
So training invention skills to L5 is a waste if time ?

I know nothing about invention but i'm looking to get into it.



Not a waste. but live most L5 skills, the pay back is relatively low.


kk well thats at least given me something to think about, thanks.