Thursday, April 18, 2013

Some Closing Thoughts on PI

A few final thoughts of Planetary Interaction before I stop talking about it, at least for a while, are in order:

Characters and Accounts

Something was nagging me after finishing the PI challenge.  Just to be sure about it, I asked Mabrick if his approach was just for one character, or for an entire account.  He told me that his play-style was to focus on one character.  Take that as you will, I'll take it at face value.  But I started playing around with my alts (I have 2 accounts that are active), and found a nice string of systems that I could feed into one final production planet.  So I started building and mapping out logistics.  Turns out you make up for a lot of holes in a production line this way.

I now have 2 characters feeding a PI chain, with a third who will be added to that in a few days.  Already the profits are increasing a fair bit, more than double what I was making per day with the one character.  The logistics are also tighter, but the string of systems is helping with this.  I simply unload everything via contract into one central station, then top off the production planet as needed.  I think a new(ish) player could do well to invest about a month into an approach like this once he or she is about 6 months into the game.  I still hold that missions and ship training are going to be the best bang for my buck from the starting line, but your mileage may vary.  It wont break 500m/month, but it's enough to fund some of my other production efforts.

I already had most of the training out of the way, and by my calcs, and judicious remaps, 3 alts can be at CCU IV and Interplanetary Consolidation IV in about 36 days.  Since this is a second industrial account that also manufactures and trades, it's not a huge deal to alter that training schedule.  I doubt this will happen on my main combat account though.

I plan on posting the results of month 2 in due course.

Customs Offices

I remarked in comments that I really wish there was a way to transfer goods in-system that bypasses the space hauler step.  A commenter noted that this may be fine in HS, but it would break the risk factor in Low/Null/WH space.  I'm not sure I agree.  If you kept the POCO taxes, that would leave the infrastructure pretty much alone.  You also still have to haul out the final products you want to sell, which will invariably require a larger, slower ship.  So I'm not sold that in-system transfer is game breaking.

I would envision this system working like an upgrade to the Command Center or Launch Pad, or a new building.  You spend some extra CPU and PG, maybe pay a fuel or isk or percentage of materials cost, and get the chosen goods routed to another planet in system. This would eliminate one of the more tedious aspects of moving intermediate goods around.  In my daily PI activities, hauling seems to be the biggest chore and I would gladly make some concessions on output to get rid of it.

Logical Routing

The biggest bugbear in PI is the way routing doesn't really work.  I set up 5 new planets this week and redesigned my production planet.  Sure enough, once it went live I had more than a few broken routes.  I though I fixed them, but some Mechanical Parts keep overflowing into one of my Launchpads.  All because you have to manually enter each route, and often have to commit to an intermediary step or two in order to trick the system into validating a given route.  This is silly.

Ideally, once you put schematics into Production Facilities and connect the links, the system should be able to figure out what needs to go where.  You could even prioritize competing facilities by outputs if you had that complex a setup.  I don't buy that this would be too hard to figure out.  If I can get a Minecraft quarry and pipe mod to sort the thousands of items in that game properly with an afternoon of tinkering, the same should be doable in a non-modded game.

Passive v Solo

After a month of running PI, I will not call it passive income.  To my mind, "passive" implies a few clicks here or there and you get isk.  That is not what PI is.  PI is lots of clicks, and lots of hauling, and lots of monitoring.  If you want pure passive income, go farm datacores.  Go find a quiet market hub and do some low scale trading.  Build something and list it on the market, updating prices once a day.

PI requires time and effort, and I think many people consider it passive because it is a solo affair.  But missions and mining are also often solo activities, and one mining run or one mission takes about the same time to complete as a day's PI efforts.  To get anything out of any of those requires attention, and PI is the same.


PI setup is ridiculously obtuse.  It seems like CCP doesn't want people to be able to figure out how PI actually works, or to be able to see how changes might impact your setup.  PG and CPU are hidden behind clicks, both for your Command Center and for new structures.  Multiple "submit" steps are needed to do the simplest things.  Timers slow down the ability to quickly move things.  Due to routing, you often have to do things is one very specific way or the local union will throw an error message.  Sometimes your ECUs just wipe the settings you had for no reason at all.  The sliders don't update until you release.  Some things have to happen in space, some can happen in station.  All of these things lead to lost time, lost isk, and frustration.


Yeah, I hate buzzwords, but I want to talk about PI in the light of other aspects of the game.  PI is good when you only have about 30 minutes to play, want to get into the game, but don't want to fall into the rabbit whole of joining a roam or starting just one more mission. For me, the real beauty of PI is providing a steady source of seed money for other industry.  My main is a FW pilot at the moment.  So he tends to lose isk, and what isk he does make is tied up in modules off in a low sec station or five.  I can get that stuff out to market now and then, but it takes some forethought and some specialized ships, and a fair bit of time.  I'm risk averse with my loot, and a CovOps can't really carry much and still be nigh-impossible to catch.  But there is a silver lining.

I can use my PI seed money to purchase building materials for items that sell for far more than the build materials.  In an odd way, Mabrick's proposition was right.  The 134m I made from the first month of PI has made me roughly 500m isk in the last month, but only because I used it to fund more lucrative operations out in low sec. FW chews through a lot of ships.  Ships need modules and rigs, and I can use PI money to buy the materials to make those modules and rigs.  Throw in some careful location scouting, and I have a trade/PI/stock route that manages to move everything along the circuit in the right order to save trips and make money. It's paying for my ship loss habit, at the very least.

So my final verdict on PI would be that as a standalone mechanic, it's pretty lousy and not a lot of fun.  Combined with some planning and other industrial endeavors, PI can make a decent income.


  1. use ur orca, squad warps, and load to it just like u would a mining op.


    1. So I tried to think this through, but I'm not getting it.

      Relevant info: All my PI is done across 3 characters on one account. I do have an Orca, but when speed is the problem, Industrials are the faster solution. Because I am in HS, all the loads from 5-4 planets safely fit into one properly fitted Industrial. On one day turnarounds, I cannot overfill any of the industrials I have access to (All of them except the Iteron V).

      I could Orca on a second account, true, but I would still have to warp it to each POCO or Launch, which with Orca aligns defeats the speed purpose. For traveling between systems, I still have the align time disparity to solve. So industrials win again. In the end, I see no value to using the Orca over the industrial. I can't log in multiple characters on the same account, so I'm not sure how splitting the loads would be any more efficient.

      What am I missing here?

  2. I'd like to point out too that Mabrick's original claim was predicated on using PI goods to make Nanite Repair Paste. I don't know if you could make that work in high sec, but I am about to start doing just that in low sec, and it should be about spot on. I could also just make Robotics instead once I finish CCUV and ICV for about 460M a month. Both options are on 1 character.

  3. Yep, I made Robotics, mostly. Sorry if any of the changes I made were unclear - I was trying to test the theory, but obviously had some real and some self-imposed limitations, and tried to make them obvious so readers could draw their own conclusions about the validity of my results. You are of course free to do so!

    As for 460m/month, in HS, I would be hard pressed to comprehend how you could get that with any combination of products/skills/whatever. I've gamed out the addition of PG/CPU, and I just can't see an additional 340m coming from that increase. PC/CPU progression is linear, not exponential, and it has, in my opinion, diminishing returns for a HS planet based on the time to train v lost opportunities of other training. At least at this point, for me. A good LS system may very well hit the 460m mark, but that is in LS :).

    I'm currently rerunning this whole thing with an entire account running PI in 3 systems, CCU IV and IPC IV. Currently the profits appear to increase more than linearly, so it will be interesting to see how it turns out. I should have the results of that out sometime in May, but it get pushed back to late June. Finishing school + honeymoon might stop me from completing the second round!

    1. Congratulations on the Real life events!

      As for PI, I can't speak for HS, cause I don't do it there,for comparison sake though, on most planets, I need about 2 RCUs a planet with 5 heads each to keep 2 P2 factories running at full output. I do 4d3h cycles, and reset when I start running low on P0 goods in storage. I also turn off heads on one and turn them on for the other if I start to get too big of an imbalance between P0 goods. That means 1 lava and 1 barren planet support 1 Robotics factory. At CCU IV, most planets will have a resource map with the above setup conducive to 1 Launchpad supporting both extractors, the factories needed to produce a P2 on that planet, and an extra P3 factory or 2. If there is enough grid for 2 P3 factories, then I put both up, that way, when I bring in the P2 from the other planet, if I miss a day, the factories can "catch up". That setup does away with an entire planet devoted to being a factory, and also cuts the export/import tax for the P2 phase in half. Additionally, I haven't seen any POCOs, at least where I am, with more than a 9% rate, and I have seen many at 5%.

    2. Nice description. My current setup uses 15 planets across 3 systems and characters, all in HS. 13 of those planets harvest R0>P1. 2 planets are completely factory planets, providing up to 8 P3 production slots. I expanded to a second factory planets because I was far out producing what one could handle. I'm still waiting to see what happens with this. I may segregate the second planet into two halves, a P3 Robotics chain and a P2 Overflow chain. The 13 harvesters pull out variable amounts of materials. At this point 3 of the 4 P1s are relatively balanced, and when one spikes, I try to switch it to the rate limiter. As I am doing Robotics, this has tended to be either Chiral Structures or Precious Metals. I am usually swimming in Reactives and Toxics.

      With one character, it was a lot of pressure to always manage the 24 hour cycle and transfer stuff. Now I am playing with a 32 hour cycle, and I can build up enough excess to store a bit in station in the P3 system so a day or two away doesn't negatively impact profits.

      The biggest benefit I can find from going to CCU V would be possibly adding a second launchpad to production planets to handle the volume of materials. At this point the limiting factor in ease of use is the 10,000m3 limit of storage in a single launchpad, and the cost of placing a second in terms of PG and CPU. In non-HS, where it is more feasible to run P2 production on one planet, I can see CCU V being much more useful, given the higher density of resources. But in HS, I find the nature of resource distribution makes running P2s on one planet rather dicey. You often end up needing at least one far flung ECU, and that's a pain if you are also adding an Advanced Facility at the same time. CCU V would help, but I think I'm getting more P1s, and eventually more P3s using my current setup.

      At any rate, I think I have daily revenues up to somewhere between 16m and 25m, and it'll take a few weeks to see where it settles out. Burn Jita also threw a wrench in selling things, so I'm sitting on a huge stokcpile of materials to sell, which makes my data so far more hypothetical than actual.

    3. Yeah, I tried P1 only planets to increase production, but ultimately, I didn't think the returns were worth the effort. I felt like it required much more hauling between planets, and it makes it more likely the whole system will seize up if you miss a day.

      Also, I don't know how you handle your hauling, but I made bookmarks labeled 1 - pickup, 2 - pickup, 3 - dropoff etc, for each customs office to make bouncing around less cumbersome. If you want to get really fancy, you could place each bookmark 1500 off the CO in the direction of the next one on your list to save you from that annoying feature of getting stuck when you try to warp off (much more important in low sec then high obviously).

    4. I have it down to the point where I can use 1 indy to swing by all the planets in a system (assuming it's only 1-2 days of gathering) on each character. I drop it all in a station in system. My producer then does his usual market/pi/indy trip and picks up each load in either an indy or something bigger if needed. All this ends up in one station in my production system, and then gets portioned out to the production planets. For that last step, assuming one lanuchpad and 4 P1 materials, I take out 6578 per P1 per CO and fill 'em up, taking the P3s back. The P3s then join the larger trade cycle. Rinse, repeat. I planned out my planets to have the production systems at one end of my trade circuit, and the trade hub at the other. In this way the PI pickups are just a part of my other activities.

      As for warping, ugh. I need to do what you said, if only because warping to 0 on a CO invariably bounces you off and by the time you are done dealing with transfers you are past 2.5km. Warping to 10km takes... forever to get to the CO. I'm rather proud that my COs in my production system are roughly in order of distance from the sun, and my others are organized to go out, them hop closer to my station each time, making each warp as short as possible. But really, that's a small thing :).

  4. Yea, trust me, Mab is just Mab. I play a bit like him too… while I do have my 2 alts from the one account buy I very rarely ever use em, both are way low in SP and I have only opened one other account and that was to make chars for a post… it is inactive and prolly will never be reactivated… I have enough trubble just being Tur. =\

    Most of the guys I know have multiple accounts… and you’re right, throwing more ‘people’ at any job makes it work better.

    In-system Transfers:
    I disagree that it would be game breaking. The whole point of of that is to get people out in space in ships at every possible opportunity to encourage “Player Interaction”. There was talk of that being ‘part’ or the reason for the upcoming changes to “Resources” in EvE… For example Ring Mining is designed as ‘active’ not passive gameplay to get people in ships and out in space… so having to fly your expensive crap all over the place makes you a target, adds risk to the equation from where you get your reward and encourages more ‘Player Interaction’.

    Now I live in holes so I use cloakies for everything unless contra-indicated or FC ordered. But were I running PI in a 1.0 system in Empire I would still use a Prowler for PI transfers. And I would use the system I use now in the hole…

    I BM a spot near enough to fly to the POCOs transfer range in an acceptable amount of time and far enough out so I don’t break cloak on the POCO (and off-line with everything so I don’t land ON that cloaked up Tengu camping for PI runners…); my next alignment take me tangentially past the POCO through Transfer Range; I am only vulnerable from when I manually break cloak to when I reach 75% of max speed… I transfer products as I pass and am always ready to warp out if need be.

    Moving shit around is the way things are done for very good reasons… EvE is not supposed to be ‘easy’. Allowing in-systems transfers to be a ‘passive’ (read ‘safe’) activity breaks the basis of EvE as a PvP game.

    I dunt get you. I have never had an issue with setting up routing… and I have never had anything ‘overflow’ The way PI is designed I have no idea how you have had that happen… And yes, you have to enter each route… how else is the game going to know what to move where? You are looking for some kinda “autoroute” button… sorry but as I see it that’s just a, ‘I dun’t wanna do that much work’ nerf.

    Passive v Solo:
    I think it is passive income 'cause like skilling, it goes on whether you are there or not. And I am not sure about you but I spend like under an hour a day on PI and I am running 6 fully loaded planets. I spend maybe 25 to 30 mins remotely restarting excavation and moving prods around, then maybe 25 to 30 mins doing the PI hauler runs... boom done. One hour a day is not a very ‘active’ amount of time to spend for somewhere between 300 to maybe 500mISK a month… and that is with me missing days for Ops/RL/etc… (granted I live inna C6 wormhole…) =]

    It’s complex, because life is complex… wanna see complex? Try setting up a REAL factory… I believe, and love, that EvE was designed to be as realistic as possible in a gaming environment. I love the complexity of PI because once you learn how it works, it is a joy to tweak it till it’s running just right.

    Always remember, one man’s Ridiculously Obtuse Setup is another man’s Fascinatingly Complex Setup… =]

    Liked your take on that… my ship habit requires heavier injections of ISK though… LOL, I just bought and fitted my 4th Loki last night. So I now have my goal of 2 in the hole, 1 PvE fit & 1 PvP fit and 2 PvP fit in Hisec. If it wasn’t for Wormhole ISK, both sites and PI, I couldn’t afford the dethstyle I lead… =]

    1. On Routing... it could really be less stupid even if it isn't automatic. For example, suppose you have a P0>P2 planet, so you have 2 extractors routed in to storage, then a route out and back in for each Basic Factory, then 2 routes out and 1 in for each Advanced factory. So, the routes in aren't that big a deal, but when you have the storage selected, why on earth do you have to keep re-selecting it for every route? Why doesn't it just stay open so you can do them all at once? Or better yet, if you are routing the same product to multiple factories, why can't you hold down shift or something, and click on each one of them then apply it? Routes could also be more visible, so you don't have to click into each factory and make sure you have the right incoming and outgoing items. Like, instead of multiple outing routes being shown individually, if it was grouped, and you clicked on the group, some arrows pointing where all the routes in that group went would be pretty nice.

  5. For routing, I'll try to explain what I'm getting at via analogy. I also play a fair amount of Minecraft. On some of the maps, we use mods that provide logic routing for pipes. A common use for pipes is to take raw materials, from a quarry, and use logic built in to pipes to route materials to either storage or processing, and then further process various things into the final materials desired. In a normal quarry setup you need to sort roughly 200 potential items. Through using 3-4 pipe types, of varying cost, it is possible to set up a routing system that can handle diverse inputs, handle overflow and backflow, place intermediate buffers to balance input levels, and route all the final products to the correct place. More advanced systems I have built include in-line sorters, selective item picking and transfer, and a whole host of logic condition based routing mechanics.

    In Eve, even if you route the correct way, you can still get either uncontrollable overflow, imbalanced resource utilization, or you just lose a chunk of materials. I have a factory planet that has been routed, rerouted, then rerouted again. There are no missing or incomplete routes. Yet when I place exactly equal quantities of raw materials into the system, it provides variable outputs. When resetting ECUs, I have had multiple instances where I specifically select an ECU, set the output, set the time cycle, install the program, and submit the change, and then come back an hour later to find the ECU is inactive. There are very limited ways to deal with this, and no option to add sorting or routing functionality via more expensive links or structures. In my mind these are fundamental failures of design, programming and game logic.

    I'm not asking for a super router AI that does it all for me. I just want to be able to set up a system that has robust options for handling basic tasks. After that, I would love the ability to do something like set a build priority list that produces P3s until no longer able, and then routes the rest into a lower tier, allowing me to place mixed quantities of materials into the system. If that existed, I would not mind the time spent on the process. By comparison, an automated minecraft quarry can take the better part of a week to tweak to both work and not crash the server. I love setting those up, because they work after you invest the time. PI lets you spend the same time, and you get a paltry echo of the same thing. Complexity is enjoyable when used in the service of something, but complexity for the sake of complexity is a bad design mechanic.

    I'm not sure how to better explain it, I just want PI set up and routing to not be an exercise in frustration :).

    As for transferring goods off planet just to take it to another planet, yeah, I just find that tedious. As stated, I don't do this in non-HS space. Well, I did a long time ago, and that ended badly. Maybe a compromise would be allowing role or standings based access to COs, so that the process could be made social or market based. Let corpmates move stuff, or let me set up courier contracts. PI is one of the few areas in the game where you can't farm out hauling to a third party. The only other I can think of is POS access, in some cases.

  6. Just wanted to leave a note on the 'greatness' of your blog. I've been an active EVE reader for quite a while and had not encountered your blog before, but I must say that (almost) all of the posts are good reads.

    As for the PI, that is how I found the blog, I'm really interested in the progress that you have made so far. Just started to look into it and have to make up my mind wether or not it is something I would spent my EVE time on.

    Anyways, keep up the good work. Cheers

    1. I'm not sure I would call this blog great, but thank you for the compliment! I try to balance the goodposting and the random stuff I type so that people find something of value here if they bother to read it.

      If you have any questions about PI, I'm more than happy to chat, although I will be taking a bit of a leave soon due to honeymoon. The best way to get me is to email me at grimmash at gmail dot com.