There are three main items that can be used to affect your invention jobs to varying degrees. This is where those are covered. Items can be used to boost the chance of success; decryptors and blueprint copies are what determine how good your tech 2 blueprint copy comes out.
The quality of the item used affects how much it adds to the job. The higher the meta level of the item the better the chance it adds. Tech 2 goods have a meta level of 5 and factional/officer's meta level is 5+. Since the meta level is above the meta level of what you are inventing (5 for tech 2 goods) they cannot be used. No matter how the job turns out this item is destroyed, so make sure it is worth it. How much this adds to the chance of success is still being determined and is currently unknown and subject to change. For an example of meta levels lets look at the Ballistic Control System I. You may notice how the quality of the item when placed in order of quality, better bonus's being better, they line up with their meta levels. Some parts may not sync up with this, but I doubt it is intentional. Update: Meta level is now listed in the blueprint creatively in the field called 'Meta Level'
| Item | Meta Level |
| Ballistic Control System I | 0 |
| Cross Linked Bolt Array I | 1 |
| Muon Coil Bolt Array I | 2 |
| Multiphasic Bolt Array I | 3 |
| 'Pandemonium' I Ballistic Enhancement | 4 |
| Ballistic Control System II - Cannot be used on invention jobs! | 5 |
Items that only have a few versions, like shield hardeners still start their meta level at one and count up to the highest item. Ships have only one version so they always have a meta level of one. But one is better than zero.
Decryptors are one time use items that modify the stats of invented blueprints in certain ways. They vary greatly in cost, some being only a few hundred thousand, others upwards of 300 million. They affect four stats: the chance of a successful invention, the final material efficiency (ME), the final production efficiency (PE) and the number of runs you can get from the end result. The base ME and PE of any invention job is -4, the following show the modification to that number. So a job done with the War Strategon will come out with ME: -7 PE: -2.
The chance is the only thing that is different from any other bpc. It is exactly like it sounds, the higher the better. The best a decryptor can do is add 30%; the worst is make it 60% less likely. Once again, these are destroyed no matter the outcome of the job. The following is a chart the decryptors in game. Note they are located under the respective pirate faction on the market place: Amarr is Bloodraider, Caldari is Gurista, Gallente is Serpentis and Minmatar is Angel. Please note that these may change as the devs tweak things, I will do my best to keep this table accurate however.
Updated for Trinity with help from Aykido
| Name | Chance | ME | PE | Runs |
| War Strategon | 1.8 | -1 | 2 | 4 |
| Classic Doctrine | 1.2 | 2 | 5 | 1 |
| Formation Layout | 1.1 | 3 | 3 | 0 |
| Sacred Manifesto | 1.0 | 1 | 4 | 2 |
| Circular Logic | 0.6 | -2 | 1 | 9 |
| Name | Chance | ME | PE | Runs |
| Installation Guide | 1.8 | -1 | 2 | 4 |
| Prototype Diagram | 1.2 | 2 | 5 | 1 |
| Tuning Instructions | 1.1 | 3 | 3 | 0 |
| User Manual | 1.0 | 1 | 4 | 2 |
| Alignment Chart | 0.6 | -2 | 1 | 9 |
| Name | Chance | ME | PE | Runs |
| Stolen Formulas | 1.8 | -1 | 2 | 4 |
| Test Reports | 1.2 | 2 | 5 | 1 |
| Collision Measurements | 1.1 | 3 | 3 | 0 |
| Engagement Plan | 1.0 | 1 | 4 | 2 |
| Symbiotic Figures | 0.6 | -2 | 1 | 9 |
| Name | Chance | ME | PE | Runs |
| Assembly Instructions | 1.8 | -1 | 2 | 4 |
| Advanced Theories | 1.2 | 2 | 5 | 1 |
| Calibration Data | 1.1 | 3 | 3 | 0 |
| Operation Handbook | 1.0 | 1 | 4 | 2 |
| Circuitry Schematics | 0.6 | -2 | 1 | 9 |
The final stats of the invented blueprint are determined in the following ways. Note the ME and PE are not in any way, shape or form related to what the original tech 1 bpc had. Final number of runs is affected only by the number of runs that the original bpc had and the decryptor bonus. At this point there is no randomness involved; if you perform multiple jobs with identical parts the results will always turn out the same, at least if they all succeed. Although hints from CCP indicate this may not always be the case. Thanks to Qual for posting these formulas.
ME = -4 + Decryptor bonusmax(...) means take the largest value of those and use that. So it will either use 1 or whatever the result of the Round.Down(...) formula returns. This is a bit nasty, so lets use an example to demonstrate. If you hadn't guessed it, the Ballistics Control System 1 will be used. The bpc is a 1000 run (max for this part) with ME and PE at 0 (not like it matters). We are also using an Installation Guide decryptor.
ME = -4 + -3 = -7 PE = -4 + 2 = -2 Runs = max( 1; Round.Down( ( 1000 / 10 ) * 1000 ) + 4)So we get a bpc with the following stats ME: -7, PE: -2 and 14 runs, not too bad. Ok compared to normal building the stats are terrible, but that can't be helped. At least we can make 14 of them.
The following is the max number of runs that a bpc can hold currently. I copied this from the forums so it may or may not be 100% accurate. When making your own bpc's for invention always make sure to max the number of runs with the 999999... trick. Trick no longer needed, just build the 'Production Limit' for max runs.
| Blueprint type | Max runs |
| Frigates | 30 |
| Destroyers | 20 |
| Cruisers | 15 |
| Industrials | 15 |
| Battlecruisers | 15 |
| Battleships | 10 |
| Mining barges | 10 |
| Rigs | 1000 |
| Light missiles | 500 |
| Heavy missiles | 750 |
| Ammo | 1500 |
| Charges | 1500 |
| Probes | 1000 |
| Most modules | 300 |
| Some modules | 1000 |
| Cloaks, etc | 100 |
This is how a negative stat on a blueprint affects your production time. You take your base speed and multiply it the constant. So if an item takes 2 hours to build with PE of 0 (what it lists normally) and the actual PE is -4 it now takes 3 hours per item you build.
| PE | Multiply base speed by |
| 0 | 1.1 |
| -1 | 1.2 |
| -2 | 1.3 |
| -3 | 1.4 |
| -4 | 1.5 |
| -5 | 1.6 |
| -6 | 1.7 |
| -7 | 1.8 |
| -8 | 1.9 |
Wastage is how much minerals are wasted (duh). The more wastage the more it will cost you. 70% wastage means that instead of 100 tritanum you use 170 units, trit is just an example it works the same for all build materials needed. The formula for wastage on negative ME apears to be the following (blame Qual if this is wrong :P):
Waste = (-ME + 1)/10The following table lists the ME, wastage and the constant to multiply the base mineral use by
| ME | Wastage | Constant |
| 0 | 0.1 | 1.1 |
| -1 | 0.2 | 1.2 |
| -2 | 0.3 | 1.3 |
| -3 | 0.4 | 1.4 |
| -4 | 0.5 | 1.5 |
| -5 | 0.6 | 1.6 |
| -6 | 0.7 | 1.7 |
| -7 | 0.8 | 1.8 |
| -8 | 0.9 | 1.9 |
So you know all about invention now, have run enough jobs to be comfortable with it and are looking for a way to reduce overhead. Then you start to wonder how all these other people get datacores. Look no further, the answers are here. Most of this info is taken from Jai Centarium's guide on the forums here (not IGB safe) so kudos to him for that.
Datacores, in short, are gotten by purchasing them from research agents, a specific type of agent. They have special standing requirements as well so we best take a look. To use a research agent you must meet all the regular requirements for a generic agent of their level, as well as have your corp or personal standing no more than 2.0 less than the requirement. An example will explain this better than words I think.
Example 1: High corp and high personal standing:
Example 2: High corp and low personal standing:
There are two other possibilities but these are the most likely. In the previous example you would need to get your personal standing to 4 or more to use the research agent.
In addition to having more stringent standing requirements than regular agents, they also require skills. Luckily for you these just so happen to be the same science skills you need for invention (not the Encryption skills). The quality of the agent and what fields they can research affect these, so your best bet is to check the info on the agent you want to use and make sure your skills are up to snuff. Also you are limited to one (1) research agent at a time doing work for you, unless you train the skill 'Research Project Management' which gives you an additional agent per level, to a max of 6. The skill tags in at 40 mil, so it is up to you to determine if it is worth it or not. (Much like everything in Eve or rl for that matter)
So now you have found a research agent and have the standing to use them, now what? You talk to them silly. You choose the field of research that you want to go into, most agents have 3 different ones, and start research. Then you wait. And wait some more. Your earn research points, more commonly called RP, at a given rate. How fast you ask? Well here is the formula.
RpPerDay = ((AgentSkill + YourSkill)^2 * (1 + (EffectiveQuality / 100))) * AreaBonus
Let's look at an example:
That's all very well and good you say, but I still don't know how to get datacores. Well it is simple, after all this anyway, you buy them from your agent. The cost is 50 RP multiplied by the same Area Bonus that you used to figure out how many RP you were getting. That's right, when buying datacores the Area Bonus affects nothing whatsoever. If you don't believe me look at this simple math:
(3*n)/(150) = (3*n)(3*50) = (3/3)*(n/50) = n/50
Why CCP decided to waste precious server time on unneccesary math is beyond me, guess they like bigger numbers for more important sounding things.
If you are thinking that can't be all there is to getting datacores you are wrong. Sort of. There are tricks to increasing the number of RP you get per day. Trick 1: Your agent will sometimes hit a "snag". This isn't nearly as bad as it seems, it is really just a mission. If you do the mission you double the number of RP you get for that day. Sometimes they are bugged and you never get the email until you talk to the agent in person. Trick 2: Train social skills. Negotiation boosts EffectiveQuality by quite a bit so very worth it if you are serious. Connections helps as well, but it is more useful fo getting to newer and better agents.
One more thing, make sure you cash out all your RP into datacores before you quit research with an agent. If you don't all your RP are gone FOREVER! That would suck, so make sure it doesn't happen to you. Once you have a good agent (or agents) just sit back and watch the RP roll in. Go by every so often to do the missions and collect datacores. Try and pick a good agent because you will likely be working with them for a while