Saturday, February 13, 2016

Guest Post: Jinrai Tremaine and the search for Strontium Clathrates

Sugar Kyle has very generously allowed me to borrow her podium in order to talk to you about a proposed change for the Citadel release that I think is going to have very serious long-term consequences for all of New Eden.

Before I go into detail, a brief introduction. I imagine that most of you have no idea who I am, so: my main character in EVE is Jinrai Tremaine and my space-job is mining ice in hisec. I run one of the locust fleets you may occasionally hear solo miners complain about, as we strip ice anomalies bare before they get a chance to make much ISK. I’m actually not here to discuss my playstyle or why I do what I do, I only bring it up to establish my credentials - ice is a big deal for me and has been for several years, so I try and stay on top of upcoming changes that are likely to affect it, which in this case involves the upcoming Citadel release.

A few days ago, CCP published the long-awaited Structure Fitting Devblog which included, amongst other things, the reprocessing rates for the various sizes of Citadel in different areas of space. I’m actually not here to talk about those, either - the part I want to bring to your attention is actually near the end of that devblog, where Team Game of Drones mention adding in Strontium Clathrates (Stront for short) as an ingredient for regular fuel blocks, in order to ensure that there’s still demand for Stront in the distant future when Structures have wholly replaced POSes and there is no more stront timing for reinforced modes. They initially planned to require 400 Stront per 40 fuel blocks, which was later revised downwards to 200. Based on my knowledge of ice mining, that still seemed like an extremely high number, so I decided to look into the effects this new requirement would have on fuel production.

Fig 1: Refined ice products from each belt, assuming maximum local refining:

Nullsec (0.0 to -0.5)
Nullsec (-0.5 to -1.0)
Racial Isotopes
Heavy Water
Liquid Ozone
Strontium Clathrates

As you can see from the table above, regardless of where exactly one happens to be mining, any type of ice produces significantly less Stront than it does any other ice product. This is an issue, because the proposed changes to fuel blocks will make Stront the second most numerous ingredient.

As a brief digression, here is how ice/ice product/fuel block production balances out currently:
  • Highsec and Lowsec ice anomalies local production is limited by the available Liquid Ozone in their ice - they over-produce Racial Isotopes (Nitrogen, Helium, Hyrdogen and Oxygen) and Heavy Water
  • Both types of Nullsec ice anomalies are the opposite - their production is limited by available Racial Isotopes and they over-produce both Liquid Ozone and Heavy Water
  • Over-production fuels jump drives (isotopes), Jump Bridges and Beacons (Liquid Ozone) and Industrial Cores (Heavy Water) and is traded back and forth between highsec and nullsec to balance out supply.
  • Heavy Water has low value due to being over-produced everywhere and having little demand outside fuel blocks.

The addition of Stront to fuel blocks changes that entire relationship. Because there’s so little Stront produced in any belt compared to any of the other resources, Stront will become a bottleneck to fuel block production in all areas of space. Every other ice product will be over-produced compared to Stront, much as heavy water is today, which will drastically lower their value at the same time that Strontium values shoot through the ceiling. There’s also no way to avoid this by cherry-picking only stront-heavy ice where it’s available; the only way to cycle ice anomalies and respawn the stront-heavy ices is to completely clear out the entire anomaly and wait 4 hours, which means over-production of other ice products will be built into efficient stront production.

That’s obviously bad news for everyone who mines ice in either highsec, lowsec or nullsec with poor truesec, as they’ll get a lot less stront per anom and a lot more low-value over-production materials. However the real danger is simply that there is not enough Stront production across New Eden as a whole to maintain the current supply of fuel blocks, let alone meet additional demand from the construction and operation of Citadels and their rigs and modules.

Fig 2: Comparative local fuel block production caps per anomaly

Null (0.0 to -0.5)
Null (-0.5 to -1.0)
Without Stront
Liquid Ozone
Liquid Ozone
Racial Isotopes
Racial Isotopes
Fuel Blocks
With Stront
Fuel Blocks

Production change per anomaly

% Change

As you can see from this table, there’s going to be a massive reduction in fuel block production across the whole of New Eden as soon as the existing stockpiles of fuel blocks and strontium run out. If it helps to illustrate the extent of the difference: Currently, emptying a hisec anomaly would fuel a Large POS for 18 days. With the addition of a strontium requirement, that same anomaly would fuel a Large POS for 10 hours. Right now, emptying a deep nullsec anomaly would fuel that same POS for 5 months. With the stront requirement, that drops to a little over 1 month.

As it stands now, nullsec anomalies are likely significantly under-utilized, in contrast with hisec anomalies which tend to be heavily mined. However even assuming a significant shift to nullsec ice mining there’s still a very finite limit on how much can be produced per day - each anomaly takes time to be mined out and then requires 4 hours to respawn, making it effectively impossible to get more than 5 spawns per belt (some systems have multiple “belts” giving them 2 or 3 anomalies at a time) per day. With even the lowest truesec null systems facing nearly 80% less potential production than they currently have it doesn’t seem possible for them to pick up the slack that will be left by the almost complete removal of hisec ice as a viable fuel source.

This is going to hit closest to home for hisec-based ice miners like myself and those actively involved in running POSes, as well as everyone interested in setting up one of the new Citadels once they’ve launched, however it’s important to note that the knock-on effects will be felt by all of New Eden’s residents. Every unit of moongoo is extracted via POSes, large amounts of manufacturing in highsec and lowsec take place in POSes, ore compression and refining in high and lowsec take place in POSes, reactions that produce composites for tech 3 ships and turn gasses into boosters take place in POSes as do huge numbers of blueprint copy jobs. In all of these cases, the people running the POSes are going to look to pass on their increasing fuel costs to consumers further down the line, or else they’ll stop using POSes and pass on the costs that their POSes were saving them (NPC station 10% manufacturing taxes, for example) to their consumers. Either way, this will result in increased prices for most ships and items produced by players.

As a final note, all my figures for calculating the effects of these changes can be found in my Google Spreadsheet in case anyone wants to look at the numbers themselves, or double-check my calculations.


  1. Looks like CCP should check their numbers... removing 96% of the fuel production capacity of highsec and 78% of nullsec's is beyond silly.

    1. I honestly hope they do, and that the end result is a better solution that preserves value for stront in the absence of stront timers for reinforcement, without shackling all fuel block production to an incredibly narrow bottleneck. Though in all honesty, I care a whole lot more about the latter than I do the former.

  2. Thanks for writing this. In my head a red flag went up when I saw the proposal to integrate Stront into fuel blocks. I didn't know where to find the numbers to see how bad it would be. You numbers suggest it's much worse than my wildest concerns were.

    1. You're very welcome, thank you for taking the time to read it. Sugar also deserves a huge amount of credit, not just for giving me the space on her blog but for hearing me out when I first told her I guesstimated that there wouldn't be enough Stront production in New Eden to meet fuel demand (I got a red flag at that one myself), and for encouraging me to actually look at the numbers involved. That lead directly to the analysis and the post above.

      It's one of the few times where I'd actually be pleased if someone popped up and pointed out an error I'd made in my calculations that had thrown everything off by a few orders of magnitude, but I went over them several times while I was writing this because it does seem hard to believe; if there is a mistake in there, I can't find it. It really does appear to be that bad.

      On the plus side, none of this is live yet. CCP have already made one adjustment to the suggested inclusion of Stront in fuel blocks, there's no reason they can't make more. In particular, the whole purpose of this change is simply to make sure there's still some demand for Stront, not to turn it into Technetium 2.0. Obviously this suggestion goes too far, but that still means it's not doing its job, which hopefully means that once more people realise how bad this implementation is it'll go back to the drawing board.

    2. I don't see the issue here. CCP will simply alter the "new" outputs and mechanisms, assuming your numbers are correct.

      But make no mistake, they will also use this as an opportunity to hammer high sec income again. High sec players will bear the brunt as there will be more wealth distribution weighted toward the null sec cartels, by whatever mechanism CCP chooses to use to fix this potential mess.

      Just watch, wait, and remember this comment.

    3. I don't think this was thought out as deeply as you're implying; the idea of adding stront to fuel blocks wasn't mentioned in the Building Your Citadel devblog (which actually suggested making an omni-racial fuel block for Citadels which didn't involve Stront) or in the discussion thread that followed, and in the discussion for the Stucture Fitting devblog it was quickly rolled back from 400 units to 200-and-a-volume-reduction after people pointed out the logistical nightmare of hauling large quantities of strontium around.

      To me, that suggests this was a quick solution someone jumped on after realising that the eventual removal of POSes would mean no more Stront being consumed in RFs. Throwing Stront into fuel to make sure it's still in demand even makes sense, the biggest issue is simply the quantities involved.

      It's possible that CCP will still use this to take a baseball bat to the knees of ice mining in hisec after they've fixed the problem (I say "fixed" because I cannot believe that cutting all fuel block production by over 80% is intentional); naturally I hope they don't, since that's my bread and butter. But I think it's more important to acknowledge that the current numbers aren't viable for anywhere in New Eden before any of us start getting provincial over the bits we operate in.

  3. This comment has been removed by the author.

  4. It's also possible that it's "working as intended" CCP has all the data for it.

    1. I can't see how, unless they're intentionally trying to sabotage the Citadel launch - there's simply no way that New Eden can maintain the same fuel production if Strontium becomes a required ingredient for fuel blocks. Since those fuel blocks are a major manufacturing component for Citadels as well as being used to run their services, Citadel prices are going to be heavily affected by fuel block prices. That's going to serve as a fairly significant barrier for entry into Citadel ownership as those fuel block prices rise; I can't imagine that CCP want that to happen with a "Jesus feature" that's been hyped for months.

  5. I did wonder about the Stront impact. These figures are so extreme it is hard to imagine why CCP would intentionally do this. Thanks for dong the legwork

    1. You're welcome. I wish I had better numbers for how much utilization there actually is for ice in each area of space right now and/or roughly what the current fuel demand is for POSes, so I could make a more accurate prediction for how significantly this will affect fuel production.

      With that being said, I'm fairly confident that even with 100% utilization, nullsec anomalies capped at 20% of their current potential cannot make up for the 97+% drop off from hisec. I just wish I could point to actual figures and say "This proves it!".

  6. So the solution would to substitute a high proportion of Stront with Heavy Water. Which would consume an excess material?

    You did skip over the ice now available in shattered wormhole, but it is in the same ratio as true sec. So it's not a fix to the problem, just an extension of it.

    1. Or just not require as much stront per set of fuel blocks - CCP's stated goal was just to ensure demand for stront in the future when RFs no longer consume it, not to make it the new bottleneck for fuel production. That could be accomplished with, say, 4 units of Strontium per 40 fuel blocks - hisec would then over-produce strontium very slightly (137 units per belt after refining) while nullsec would over-produce significantly more, although bear in mind that's only looking at demand for fuel block production, not demand for siege/triage.

      I tried to get the numbers for shattered wormhole ice, but couldn't find a reliable source - all I gathered was that they're large and they contain all types of ice, but not the total volumes or ratios, or whether they follow the same 4 hour respawn pattern as k-space anomalies or pop up more randomly.

      Also I'll admit I made the assumption that they probably don't have a significant impact on fuel production based on the logistics - without POSes there's no refining in system and the only way to compress would be to deploy a Rorq in space, in a system that's always got a K-space static in addition to its other static connections; without that, every miner's got to jump out every time their ore bay gets full. I imagine there are some WH residents who do it just to produce fuel without having to import from K-space, but I can't see the increased risk and logistical work involved making it massively popular. If you have hard numbers, or even just a better-educated guess than my assumption, please let me know though.

    2. (disclaimer - I do not under economics)

      Allow me to be the Devil's Advocate for moment. If Stront became a bottleneck in production, then would it become the new technetium? Valued for scarcity.

      The other thing I also thought of; was that how much "float" of stront is currently held on the market? If there is a current over-supply, then a short-term high consumption might be required. (that might more over-thought than it really needs to be; I just would like my faith in CCP to be upheld in logic).

      If you are mad-keen on the WH numbers for an ice anom, I will see if I can obtain a hand count.

      oh and FYI Shattered WH do not have anchor points for POS. And the J00 (jay zero zero) series only allow frigates, destroyers and fitted HIC inside. Which means no logistics, compression or refining. So yes, miners need to do all the leg work.

      My experience only comes from day-tripping and ninja mining.

      And a huge thanks to Sugar for hosting this, and thanks Jinrai for presenting all the detail.

    3. I got a bit carried away exploring the ins and outs of strontium bottlenecks/stockpiles in general, so this is going to have to be split over 2 replies in order to fit.

      Part 1:

      There are 2 issues to consider, because "a strontium bottleneck" isn't the same thing as "200 stront/40 fuel blocks". The latter is a strontium bottleneck, but that isn't why it's a problem - it's a problem because it's a much narrower bottleneck than the existing racial isotopes/liquid ozone bottlenecks. A less restrictive Strontium bottleneck would have a significant effect on the ice production profession, but wouldn't have the same effect on fuel prices or the EVE economy as a whole. On the other hand the 200 stront/40 blocks limit is going to have massive effects on fuel prices which will knock on to the rest of New Eden as costs get passed on from POS owners to consumers.

      On Strontium-based bottlenecks in general, the biggest issue is the ice harvesting mechanics; each ice anomaly contains a finite volume of ices based on system security level. Once one type of ice is mined out the only way to get more is to mine out all the ice in the anomaly and wait 4 hours for a new one to spawn. In the context of a Stront bottleneck that means producing more strontium requires also over-producing every other ice product. There's no way to cut back production of low-value ice and produce more high-value ice, or for null to focus only on producing Stront-rich ice types only available there and leave the low-value ice production to hisec. Every area of space will over-produce isotopes, liquid ozone and heavy water in the process of maximizing Strontium production. As a comparison to Technetium, that's as if, back when it was the main bottleneck for T2 production, mining/reacting Technetium moongoo was actually producing all the other non-Tech moon products and doing so in greater quantities than could be used with the Tech.

      The other issue is that there's a big difference in strontium production between hisec and nullsec, which means any sort of Stront bottleneck is going to have different effects on each area of space. A bottleneck aimed at only slightly reducing hisec fuel production (which would be about 5 Stront per 40 fuel blocks) wouldn't put any cap on Nullsec production, which would still be limited by the availability of isotopes, whereas a bottleneck that only slightly limited nullsec's potential fuel production (a minimum of 38 stront per 40 blocks) would cut hisec's potential to less than 15% of current. Personally I think that hisec ice mining is a niche that's worth preserving;I'm biased since it's my space-job, but I also find that having a single anomaly that every miner in the system shows up at builds a community in a way that having the same number of players spread over asteroid belts doesn't. It even serves as a conflict driver - I know of one ice miner who locks down ice systems to mine via extensive use of gank alts.

      Having covered the negatives, what are the benefits? The answer is that they're actually mostly the same as the negatives; if you wanted to reduce overall fuel production in New Eden, or to reduce production in hisec specifically without harming Nullsec then a Strontium bottleneck would be one option to achieve that. One big difference between a Stront bottleneck for fuel and the Tech bottleneck for T2 is that Stront probably won't serve as a major nullsec conflict driver - ice anomalies are pretty evenly spread across all of nullsec, rather than just one area, so it's not possible to establish a monopoly. I don't know if null alliances might go to war to just acquire more Stront, but at the very least they'd need a lot more ice miners to improve production in the areas they already hold before they started looking to take over more.


    4. Part 2:

      On the subject of Strontium stockpiles, I don't know how much exists. On the one hand, most ice production has occurs in hisec, with Stront as a by-product. Given how poor hisec ice is for Stront production, that suggests there aren't huge piles of it lying around, which is supported by market volume - Jita only sees about 2 mil units traded per day, some of which is likely stockpiles changing hands rather than fresh production. On the other hand Stront has relatively high volume and low value, so it's possible there are stockpiles that have never been on the market because they weren't worth the cost to move there - I've still got nearly all the Stront I've produced for that reason. It's also a fair bet that most cap-heavy alliances are sitting on stockpiles, either as alliance-level assets or just in the hangars of capital pilots, since it fuels siege and triage modules. I don't know how much that's likely to consist of, but at the rate of 1,500 strontium per hour of T2 siege / 900 per hour of T2 triage per capital it's not impossible that those stockpiles run into the tens or hundreds of millions for larger groups.

      Burning off those stockpiles would be one reason to set the stront bottleneck as tight as has been proposed then relax it later, but I can't see what benefit that offers given the reasoning for these changes is just to maintain Stront demand after POSes are eventually phased out. It's also probably not the best way to do it, since it's going to require burning through the stockpiles of POS fuel before it really starts in on the Stront stockpiles, not to mention that there's a big question mark over whether or not null alliances would be willing to convert their Siege/Triage strontium reserves into fuel blocks at all.

      Finally, thank you for the offer of w-space ice numbers, but I could get those via Singularity. The info I'd really like to know is how the anomalies themselves work - do they have the same "get x anomalies per system at a time, when an anomaly has been cleared it respawns 4 hours later" mechanic as k-space, or do they spawn on a less predictable basis more like combat anomalies? While I could get that from SiSi too, I'm not quite curious enough to get my entire mining fleet into a w-space ice anom, spend several hours emptying it and then watch to see if/when it respawns. I'd also love rough numbers for total ice production in w-space, just like I'd love those same numbers for each area of k-space, but I'm pretty sure CCP are the only ones who have those and they're not sharing.

  7. This comment has been removed by the author.

  8. In wormholes, I suspect each citadel will replace multiple pos, so fuel block consumption will have a net decrease. Not sure how that will trickle through the larger economy.

    Currently pos security is bad so many towers dot an inhabitted system as individual shield passwords is vital. The citadel removes that concern. Most likely several citadels will be desired to increase the level of comitment required to evict however that number will likely be far less than what is deployed now.

    Also, in the past strontium has been of such low value that when moving wormholes it is nearly always trashed rather than hauled out. So clearly there is some room for a price increase. The unmentioned effect is that with a fuel block consumption decrease there will be a PI value decrease. That might be of more concern as the current value is low enough that many do not bother even with the strong planets which dot many wormhole systems.

    - Kynric

    1. Citadels won't require fuel to simply have the basic structure running, only to provide services, so I'd imagine that WH residents will only have one "active" citadel per WH consuming fuel - any others added to make them harder to evict will still provide storage, but no market, reprocessing or other services. That will also cut down on ISK costs for fitting, since they won't need rigs in that case either.

      On the other hand, Citadels can't run reactions, or at least won't have that functionality when first released, so if WH corps are running polymer reactions for T3 production then they'll still need at least some POSes to keep doing that. Though I have to admit I have no idea if that's common in WHs, or if the residents just export the raw materials to K-space where they get reacted in lowsec.

      As for Strontium value, that comes down a lot to ice mechanics; the only way to respawn an ice anomaly is to completely mine it out, there isn't the option of just cherrypicking the best bits and waiting 'til downtime as there is with asteroid belts (or carefully balancing industrial indexes to de-spawn/re-spawn ihub belts every other day). Which means that ice products basically exist as either bottlenecks for production or as over-production, without much in-between. Right now both Liquid Ozone and the 4 Racial Isotopes exist as bottlenecks in different areas of space, hence why they're the two most valuable ice products (compared to how many are produced per belt) and heavy water is always over production (and it has comparatively low value as a result).

      Strontium is harder to pin down because currently its use is situational - you want a pile of it when setting up a new POS and you need a stockpile for siege/triage, but for most players those aren't things they do every day (wormhole capital escalations being the one exception that springs to mind) so there's no way to give a specific "We need X amount per day". Adding strontium to fuel blocks will change that, which isn't necessarily a bad thing - as you've noticed, it's not exactly the most valuable commodity in New Eden right now - it's adding it as a requirement at such a level that it not only becomes the new bottleneck but limits production to a fraction of what it used to be that is going to cause problems.

    2. Hmm because community, why would you have to do this alone? I would be game for a mining fleet test in sisi, with shattered WH. If it was 1) CCP sanctioned and 2) occurred between Sat/Sun for timezone coverage.

      Although CSM summit, ask for the question to be tabled. Do shattered WH follow the same rules as k-space?

      Getting in on sisi is not a problem. The /moveme command typed in chat will allow transfer to both Thera and two shattered WH. All three contain one ice anom. The /moveme gets around the restrictions, meaning you could access with a rorqual and exhumers. The sleepers are tough however. The ice is in a gentle arc with stretches north and south for 380km. As a guide 23 chunks of gledius ice ranging 55 to 65 units per might give you some of scale involved.

  9. This is making huge presumptions that the fuel use of all the new structures in the game (including Citadels) is going to be the same as the fuel use of all the POS now.

    Personally, I can see my fuel-block usage going down by 100% once Citadels are available - as all I want to do is store ships and stuff. Therefore I'm not going to need any fuel.

    There are plenty of POS that are just being used as a safe place to run to, somewhere to store a cap ship etc. I really think that we're going to see demand for all materials currently used in POS fuel completely crash as Citadels come online.

    1. Actually I'm mostly looking at the short term (for a certain value of "short" - it's been nearly a year since Citadels were announced as being the first class of Structures to be added and they're not on TQ yet; I don't think it's unreasonable to assume it'll be a year or more before POSes are removed) where Citadels are co-existing with POSes because there's a lot of functionality they won't provide, at least on initial release. While you're right that people won't need fuel upkeep if they just want an armed storage depot they will need fuel for refining/reprocessing and POSes will still need to be used (and fuelled) for moon mining, reactions, blueprint research/copying and manufacturing. Fuel demand isn't going to dry up overnight, nor is it likely to drop off by 80% or more, which is what this Strontium bottleneck is going to do to fuel block production.

      Also, bear in mind that every Citadel (and future Structure) and all the high/medium/low/service slot modules will require a fuel blocks to construct in the first place, even if they never consume any; that one-off cost won't equal the long-term demand for running a POS, but it's still going to fuel (heh) some demand.

    2. In which case, CCP needs to do nothing now, there's still going to be some Stront demand for at least some time.

      Once the POS/Structure balance is switching more towards the new structures, then CCP should have enough data to work out where the Stront demand is going to fall out, and then they can make an informed decision as to how to change the recipe.

    3. That's exactly the kind of reactive approach I'd be far happier to see CCP take; my main complaint has been that they're doing it the other way around - they've decided that Stront demand WILL be a problem in the future, so they've pre-emptively thrown out some numbers for a "fix" which will actually wind up breaking a bunch of things in the here-and-now.

  10. Well it seems CCP is going to adopt 'wait and see' approach seeing their latest reply.

    Structure skills and the component changes are listed to be introduced for March. Disregard our previous reply, we will also have the fuel block changes in March to see how consumption evolves before Citadel release so we can quickly iterate on it should we need to.'

    1. Yep, I'm kind of at a loss - the given reason for the change in the first place was simply to ensure demand for Strontium in the future when it's no longer being consumed by reinforcement.

      It's pretty obvious that these changes go far, far above and beyond that - we can certainly discuss whether the additional effects are good or bad, but we can't deny they exist. But apparently they're still going through, barring any more last minute adjustments.

    2. The only way to justify CCP's thinking is probably to burn through the surplus stront that is currently in the market

  11. good on exactly what you'll discover inside the add-on. best kodi sports addons whether it was a Box Office hit as well as based on exactly nice.