Fix What Ain't Broke, Please

My dear old dad always said, "Don't fix what ain't broke." What he meant was: don't waste time changing something that is working adequately. And certainly, the invention process in EVE Online isn't broken. It's a well-established mechanic that players have used for years (since the Revelations expansion in 2006) to generate Tech II blueprint copies. It has a few internal quirks and inconsistencies, but once players get some experience with it, invention becomes fairly routine and easy to manage.

For the last three years, I've generated a large proportion of my in-game funds from Tech II module manufacturing. In fact, I admit that inventing, making and selling Tech II items has made me relatively rich in New Eden - wealthy enough to support an expensive ship collecting habit. After mastering invention, I no longer worry too much about plunking down a billion ISK or more, every once in a while, for some shiny faction hull.

In fact, I got so comfortable with invention that I created a formal syllabus and taught classes about it at EVE University, and re-purposed that content into an online guide.

And so, it was with some trepidation that I read the latest dev blog from CCP Ytterbium about plans to change the invention process. After the massive overhaul to manufacturing in the Crius expansion, I have been experimenting to determine if a solo high-sec industrialist can still make a decent income - and so far, the results have been encouraging. However, since almost all of my manufacturing output is Tech II items, which have been quite profitable, I was fearful that CCP Games was planning to eviscerate my favorite money-making activity.

No Worries

Fortunately, my fears appear to be unfounded. The proposed changes to invention are definitive improvements on the process, and continue Crius' theme of making industry more accessible to the capsuleer masses.


First, it appears that CCP Games' initial efforts to open industry as an option to more players has been successful. Since the release of the Crius expansion, players' use of all aspects of industrial production have increased very significantly. Clearly, simplifying the user interface and streamlining industrial processes have garnered the attention of large numbers of EVE Online players.

The proposed invention changes continue in that same vein, by eliminating inconsistencies while at the same time introducing more variability in outcomes. For example, both Reverse Engineering, used to make Tech III blueprints for strategic cruiser components, and Invention, used to make Tech II blueprints for various modules and ships, will soon follow the same general principles and process, making both easier to understand and apply.

After reading the dev blog, I found myself nodding like an overenthusiastic bobble-head doll. In fact, I wish CCP Games would go ahead and implement all of the proposed changes in the upcoming expansion, Oceanus. Instead, they are being annoyingly prudent and sensible by waiting for player feedback before incorporating the planned invention changes into the game.

Pleasant Surprises

I am surprised, but also somewhat relieved, that passive generation of datacores from R&D agents is not addressed in the dev blog text. I recall that CCP Soundwave called for the eventual elimination of all passive activities from the game in 2012, including datacore farming, and I'd expected that to happen in the upcoming revisions. The dev blog says nothing about this, however, so perhaps CCP Games no longer worries about it, since CCP Soundwave has since left the company. Or, perhaps CCP still has some ideas hidden up their sleeve, and their true plans for R&D agents are yet to be revealed - leaving something for conspiracy theorists to chew on until the official roll-out of the planned changes.

I like the balancing of different types of datacores used for invention jobs. I've always kept more of my R&D agents focused on Electronic Engineering because I used those 'cores more frequently. After the proposed update, that won't be true any more, as there will be more need to use a wider variety of datacores in invention jobs. That means I'll need to change some of my R&D agents' focus on different science disciplines, but that is a relatively minor one-time inconvenience.

Eliminating data interfaces completely is a welcome change. They add nothing but needless complexity and expense. I recall how I bought dozens of these when I first started experimenting with invention, as I'd misunderstood the process and thought they were consumed like every other item - definitely one of my more bone-headed decisions. It took quite a while to sell that stockpile of excess interfaces in the market, after I realized you only needed to have one of each type to start every invention job. Future inventors won't ever have that opportunity to repeat my mistake - this is a good thing.

Eliminating the use of meta items from the invention process to increase the chance of success is also a good idea. Like data interfaces, they seemed a nonsensical wrinkle in the process, used infrequently by ultra-experts trying to optimize outcomes. Replacing these with the use of Teams makes much more sense, and is consistent with how manufacturing now works since the Crius expansion. However, the dev blog does not elaborate on how invention Teams will work, exactly - it only says it will be "just like it works right now for other industry activities". Later, it says, "Having Teams [will] affect Invention [Material Efficiency] and [Time Efficiency] inputs, which can shave Datacore costs down." On the face of it, this sounds like a good thing, but I'll be reading the forums looking for more precise details on how Teams will affect invention outcomes.

The changes to probability of invention success will concern some large-scale industrialists, but I am not overly bothered by what has been proposed. While the revised formula means a generally lower chance of base success, the compensating factors of Teams, revised decryptors, and the wider diversity of job outcomes more than make up the difference, in my opinion. I love having a spectrum of invention job outcomes, from exceptional success to critical failure, as opposed to the current boring and totally binary results we receive today. I also like how a portion of datacores are returned for some failure outcomes, or how ME and TE can be improved with some of the higher-ranking successes. However, it's not clear to me how datacores might be returned for smaller invention job failures. For example, if my Hobgoblin drone invention job, which uses only two datacores, gets a "terrible failure", which provides a return of 10% of datacores used, does that mean I get one datacore back (rounding up), or none at all (rounding down)? Again, I'll be reading the forums closely to get more details.

The wider variability of invention job outcomes is a welcome change to what was previously a totally binary process.


My absolute favorite line in the entire dev blog is this:

[W]e recently changed Invention to only consume one run of a Tech I blueprint copy instead of eating the whole stack. The next logical course of action is to allow players to start invention with multiple runs (that each consume one run of the input blueprint copy) so they don’t have to launch them manually every hour, which is what we’re going to do with this set of changes.
— "Lighting the Invention Bulb" dev blog, 11.09.2014 14:17 by CCP Ytterbium

If this was the only thing that CCP Games changed in the invention process, I would be a very happy industrialist indeed. The most irritating thing about invention is the need to log in repeatedly to deliver completed jobs and initiate new ones. This singular change - the ability to start automatically repeated invention job runs - eliminates that hassle completely, and improves efficiency dramatically. This is not just a good thing - it is a wonderful thing - and it warrants a sincere "thank you" to CCP for including it.

Fixing What Ain't Broke

It's not often that I read a dev blog in which I like everything that CCP Games proposes, but this is certainly one of those times. The changes proposed to invention and reverse engineering make sense, will be easier to learn and use, provide more interesting outcomes, and greatly reduce the tedium of the current mechanics. What's not to like?

Sorry, Dad - it seems that fixing what ain't broke does make sense sometimes. Who knew?

Fly safe! o7


Talking about Industry

A quick bulletin: I'll be joining DJ Wiggles' show on EVE Radio on Sunday, July 13, at 18:00 EVE Time to talk about the coming Crius industry changes, and how CCP Games is rolling it out.  It should be an interesting discussion.

DJ Wiggles of EVE Radio

DJ Wiggles of EVE Radio

Last April, DJ Wiggles reluctantly predicted the end of EVE Online within three years, citing management issues at CCP Games. He has been a tough critic on the company's decisions, especially concerning the ongoing development of EVE Online.

He and I have been exchanging tweets recently, as we both worry about the success of the upcoming Crius expansion on July 22nd. This Sunday, we'll chat more about that and speculate on what CCP Games should do to keep the EVE Online player community happy. We may even put on our tinfoil hats and explore a few wild conspiracy theories, just for fun.

Join us on Sunday!

Fly safe! o7

Crius Industry: Competing with Wal-Mart

As planned, I dropped from EVE University yesterday. It feels weird, being on my own in New Eden. I was used to being able to strike up a casual conversation in corp chat anytime I wanted. Now  I see only myself in corp chat - and perhaps one of my alts if I'm multi-boxing. And talking to oneself through alter egos is unsatisfying - trust me on this.

Getting Ready for Crius Industry

I have begun planning for my solo high-sec industry experiment. The first order of business was to select a system that would be suitable for research and manufacturing jobs. Once Crius is live on July 22nd, costs will scale relative to the number of jobs running in a system - the more jobs, the higher the cost. A previous dev blog said this cost could be up to 15 percent higher than current costs in an NPC station, but I am curious to see how it actually plays out in practice.

I've considered two general lines of thinking about where to place one's operations in order to minimize production costs after the Crius update. One idea is to operate in a remote system with no station facilities, and maximizing the stacking of POS structures (and therefore, use large towers). This should minimize build costs due to lower system industrial activity, as well as maximize advantages of lower costs from the use of multiple POS structures of the same type.

How many players will be stacking the same types of structures in their POS, to minimize production costs?

How many players will be stacking the same types of structures in their POS, to minimize production costs?

The other approach is to base operations in systems with multiple NPC station production facilities, in hopes that the higher activity scaling costs will be offset by a more frequent use of teams, which improve efficiencies in a system, but whose services must be procured by being the highest bidder.

I suspect that serious high-sec manufacturers will opt for the first approach, and specialize production of certain items by using large towers stacked with the same structure types to reap maximum cost reduction bonuses. Using this approach, each would be specialized to a particular function (e.g., copying/invention) or to single item production (modules, ships, drones, etc.) Specialization of each POS will be the key to success.

The downside to this approach is the big cost to keep those large towers fueled. And to maximize profits, they will have to be kept operating constantly. This will not appeal to players who prefer more casual solo or small group playstyles.

Alternatively, using NPC station facilities and/or smaller POS towers in "general purpose" configurations - running a variety of different structures - looks to me to be much less efficient under the Crius changes, and therefore much less profitable. That is, unless they are based in a system which wins a high number of bids for efficiency-raising teams. I am very uncertain how feasible this approach will be, as I expect that competition for teams will be fierce. However, that will depend on the available supply, which won't be known for certain until Crius is actually deployed. If teams are plentiful and relatively easy to hire, then this method could work.

Competing with Wal-Mart

Is this the future of industry in high security space?

Is this the future of industry in high security space?

Based on my initial calculations, however, things are looking bleak for casual, small scale manufacturers in high-sec. It's like the small independent shop owner in town trying to compete with the new giant Wal-Mart down the street. There's no way for small generalists to win consistently against the larger-scale and more efficient operations that will be used by serious specialists.

Even knowing this, I am going to start my experiment with a small POS tower with three research labs, and will rely on NPC station facilities for manufacturing. That is a common configuration used by many casual solo industry players in high-sec, and one I relied upon for many months when I first got into Tech II module manufacturing. I am eager to see how viable this old approach will be in Crius - and what kind of profit margins it will produce.

It's going to be interesting to see if there is still room for the small independent general producer in EVE Online after July 22nd, or if New Eden will become dominated by the large-scale Wal-Mart style efficiency specialists. If I were a betting man, I'd say that the latter is more likely. We shall see soon enough.

Fly safe! o7