Wednesday, June 16, 2010

NPC Orders Begone!

So we've had a fun week of Planetary interaction with many people realising that the fastest way to make money was to buy NPC goods and process them into something more valuable to sell. You could setup whole farms of advanced processors on hi-sec worlds, feed in mechanical parts and consumer electronics and then process these into robotics which could then be sold to NPC's for about a 50% gain in value. I made about 100million in this way in the past week, using a single toon running a bunch of barren & temperate planets in hi-sec. Some very fancy tier 4 components could be manufactured with ease using NPC components as input and those appeared on the market well before they would be expected, with production lines kick started by NPC products.

Anyway, if you didn't know this and were interested then tough luck, because CCP have removed almost all the NPC buy/sell orders during downtime. So now, planet interaction doesn't have the NPC market supporting it we can expect a slow evolution towards equilibrium as 'market forces' free themselves from NPC orders. Except that if you look at the volume numbers on NPC goods you'll see that people clearly were stockpiling items in anticipation of this moment, and right now in Jita there are already sellers dropping their stocks on market trying to guess at the final value.

Silicate Glass is a great example - old NPC orders were <200isk, but under the new PI production chains you make it by mixing Silicon (only found on lava planets) with Oxidizing agent (only found on gas planets) - so it's one of the hardest Tier 2 materials to make. As such those people who were buying up silicate glass are now selling it at over 1000isk a unit.

Robotics is an interesting one to watch though, because of the Buy->Process->Sell chain there are no doubt large amounts of these, or at least the components in people's hangars, but there's also a constant demand for these in POS fuel, and manufacturing either requires a single installation on rare plasma planets, or shuttling materials between other planets. So I suspect that actual production will be a lot lower initially (except in 0.0) as prices will be depressed, but as supplies get used up the prices will only rise until it becomes worth people's time to produce them.

The people with those fancy plasma planets are currently better off making Precious metals, and maybe once the prices for those drop, Enriched uranium. Enriched uranium from NPC orders is pretty pricey so it didn't have the same incentive to buy and stockpile. Precious metal prices currently push that even higher, but those can be extracted from barren planets, which are much more common, so the price for those will drop more readily.

Myself - I've got almost 20 planets being farmed right now, some in hi-sec/low-sec and some in 0.0 - I've already made my investment back and expect to continue to do well over the coming days, I have a few more ideas on where the market is going and investments have been made.

How is Smartbomb Range Calculated?

One of those dumb questions I just had to jump on SiSi and test it out - indeed from the screenshot below you can see that the target is being damaged even though the ships centers are more than 5km apart. the range of the bombs is computed based upon the gun range and so the range is boosted by the collision radius of the models. This means that the 'volume of effect' (not area of effect - that's 2d) is larger for larger ships, and pilots putting smartbombs on their capitals are getting more bang for their buck than those who use bombs on their dinky little battleships.

Tuesday, June 15, 2010

Making Planetfall, Not Planet Fail

So I jumped into planetside production with both feet and managed to get all 6 of my alts trained in production mostly only with level 3 skills - so improved Command Centers and 4 planets each, but the two 'mains' have gone and trained the skills for advanced command centers. I've made a few mistakes and realised that I needed to nuke a bunch of buildings and redesign my colonies, but each new one is getting more and more efficient.

For example I used to put the command center and launchpad at a spot roughly equidistant from all the hotspots and have everything feeding into it, but now I setup a command center on one hotspot and the launchpad on another hotspot and both of them can be used as buffers for materials prior to processing. Any other hotspots that are required may get a dedicated storage node if there's a decent chance that a 5 hour cycle would overload the link and require an upgrade - even though I aim for 23 hour cycles as the default it's nice to have a buffer for moments when a 5 hour cycle is possible.

And, it appears that there's been an annoying bug which causes cycle times to increase if they cover downtime, it ends up being more like a 36 hour cycle which I then pad out with a couple of 5 hour cycles. It's not the only bug affecting PI, materials get disappeared randomly, sometimes when I'm exporting them off planet. It's mostly annoying, since as far as I can tell everyone is dealing with the same problem, at least if someone has discovered a fix they're not telling. Regardless, the market is so young and chaotic that I expect that the effect of material losses is small compared to other factors driving the prices.
3 of my neutral alts are working together, I've found a pretty nice low-sec system very close to hi-sec - when you're going for low-sec planet harvesting remember that yield goes up as sec status goes down. So like the risk-reward graphs I laid out in my discussion of missions there's this discontinuity that makes the increase in rewards from 0.4 sec planets a whole lot less attractive. No, you really want to look for those 0.1 sec systems to get as close to the yields you expect in 0.0 as you can without having to deal with holding sov or sneaking though warp bubbles. So 2 alts have some extractor dominated setups in a couple of low-sec systems and the 3rd is focussing on building the final product with 4 barren/temperate worlds that are almost exclusively host advanced and hi-tech processors. I just import the products and the whole production chain goes to work pushing out a p4 every few hours, I'm still optimizing the whole thing, and over time I expect changes as some of the NPC goods I have in stock run out and need to be built - but that'll be a long time for some.
My other characters are all out in 0.0 furiously producing POS fuel stuffs, my alliance is going through a rough patch and I don't see a huge internal market for the P4's at this time, but, we'll always have towers to fuel. (and even if we end up leaving our current space, the new residents will have towers too).

Anyway - here's some pictures of my layouts..

One of my first - concentrating on making p2 and p3 materials, I only have 3 advanced processors so I switch them aroung right now, but I think I'm going to settle on a single chain soon. Notice how the command center just sits there like a wart on the side of the network?

So my improved design uses the command center as a secondary storage node - setting it on top of one deposit, and then the spaceport goes on another, on this gas planet I only really needed two locations so I get this layout that resembles a dumbell.

And this is one of those factory planets I mentioned, only 3 extractors but lots of advanced processors fed by imported materials. The local extractors may be nuked in the future to make way for more processors as some NPC stocks start to get used up.

And finally, this is the result of trying to place a command center when a red fleet shows up and made me panic, I put it on top of a deposit which wasn't needed to make POS fuel, so it just sits there completely isolated from the network.

Monday, June 14, 2010

Training The Perfect Alt

With the introduction of Planetary Interaction Eve has added yet another thing that alts can do very effectively for very little skill training, and when I'm talking about 'alts' I'm really talking about the spare character slots on an account that frequently go unused because you can't play or train with them at the same time as your main.

I'm guessing the training is the main thing that holds most alts back from getting seriously developed, people don't want to devote weeks away from training their main to increase the utility of their alts. At least that's what dominated my strategies and skill selections - I'd plan things ahead of time in Evemon to tune the learning plan, and try to take advantage of those times when my main was spending time in a PVP clone without implants.

Two of my alts were created under the old character creation system, they started out with roughly 800,000 sp, I didn't really train them until Apocrypha changed the rules and added the 100% training time boost, this speed boost is a huge help to creating alts focussed for their tasks and while it was nice to be able to create a Gallente pilot with Drones 5 out of the box, it was rather less useful to have Drone Navigation 5. (For those of you who joined Eve after apocrypha you can get an idea of the old character creation system from online tools like this one). Nowadays you only get about 50ksp, so you end up with 750k more skillpoints trained at double speed in the skills you choose - for most things you want alts to do this is generally going to work out better for you in the long run.

So what do you do with those alts? Well - here's a few pointers

Planetary Interaction - The new stuff, it has the advantage that once you're setup you just need to login maybe every day or so to restart extractors and pickup finished products. Training time is light, new characters come with the pre-reqs for the basic skills, and all you need is a big enough ship to deploy the command centers, It's quite possible to carry and deploy them via cargo expanded astrometrics frigates.

Market Trading - They can sit in the market hub of your choice and update orders, trading skills let you place more orders and extend your reach, but you can make cash without any real skilltime invested. Only danger is that you realise your 'alt' is making more money than your main and turn into a full time trader.

Datacore Farmer - Requires a chunk of training time to get the skills (Science 5, Engineering/Mechanic/Electronics 5, Research 5, Lab Operation 5), and a bunch of missioning to get the standing, but once you do the datacores just keep getting produced and the only thing you need to do is pay your R&D agents a visit once a month.

Research Alt - There's always such long queues at the research slots in hi-sec, but you can compensate for this somewhat by queueing up lots of work across alts. Very easy to get skills for ME/PE/Copying and if you've got a datacore farmer setup then this is included for free. Invention takes more work, with you needing 2 science skills and a racial encryption methods skill too.

Manufacturing Alt - Get production efficiency to 5 and you can grind out more stuff, and we all know more stuff = more profit.

Neutral Hauler - Need to move something in hi-sec when your corp is war-decced - train up an alt and have them move it for you, it'll take less than 12 hours to get in a hauler. I recommend an Amarrian toon for this since the Sigil only needs Amarr Industrial 1, plus for smaller stuff you can move decent amounts in a Magnate (as demonstrated by the number of obvious macros running courier missions in a magnate.

Neutral Scout - No training needed at all, just fly the noob ship to the scout system and profit. It's nice if you can get a cloaking device, or even just make a frigate unprobeable with ECCM, it's even better if you train for covert ops.

Cyno Alt - All you need to train is Electronics 5 to get access to Cynosural Field Theory then you can put up that big sign that says "I'm Here, Gank Me!" so that your friends can jump in their expensive cap ships and then dock up while you sit stuck in space until you're destroyed. The nice thing about having your alts trained for this is it's quite often that your friends will be hunting around for someone in the right place to light a cyno, so having multiple extra options will make you lots of friends.

No Strings PVP Alt - Join Red Vs Blue and have some fun, or just

So, anyway, I tend to combine most of these functions into one - Datacore Farming, PI, Hauler, Trader, Cyno Alt - and so I spent some time trying to come up with an optimal skill plan that would minimize my training time.

I mused about the best starting race for some time -
Amarr - Win lots of points for hauling and PI because the Magnate and Sigil have lots of low slots which can make them really capacious, or harder to catch with the aid of a full rack of WCS's. They're excellent for managing PI in low-sec. Shame that they don't really have any effective low SP combat ships for missioning or PVP.
Caldari - They're the best race for missioning, The AML Caracal will make standing grinding a whole lot easier, and you'll need to do lots of grinding. They also have specialist ECM ships which are a gift to Low SP PVP, except that RvB bans ECM.
Minmatar - The Rifter and Thrasher are nice for mission grinding, but with low skillpoints I can't find a cruiser which is suitable. The Mammoth is also an excellent hauler if you want to train Minmatar Industrial up to 4
Gallente - drones are nice, especially for semi-afk mission grinding, but unless you train drones 5 you're missing out on Drone Interfacing which gives a huge boost to drone effectiveness. They also have the best Hauler in the form of the Iteron V, but that requires quite a big of investment, and by the time you're there you're already thinking about making your alt a freighter pilot.

In the end I went for an Amarrian pilot, trained frigate to 3 for the Punisher and Amarr industrial 1. Then I started training Caldari Frigate to 4 so I could Get Caldari Cruiser and fly the caracal for mission grinding. The industrial skills are all rank 4 so going purely Caldari and training Caldari Industrial to 3 to get the Badger MkII is more training time than training an extra frigate skill to level 3. I made the toon Amarrian because I figured that I'd be more likely to use lasers than I would be to use railguns.

To Get a cyno alt I'd need Electronics 5, which along with Science 5 gives me access to the Electronic Engineering and Electromagnetic Physics Skills, not the best datacores, but it saves having to train Engineering or Mechanic to 5. Add in the important PI Skills to Level 3, some basic combat skills for flying the caracal (general rule - if it gives some useful bonus, or unlocks some type of module then train it at least to level 1), some social skills - negotiation will increase my datacore yield, and some trade skills - feed them all into evemon and come up with implants/learning skills that reduce the overall training time and start cooking.

It's a Pain to grind standing when you're basicly confined to courier missions or low level combat missions, but with good agent choice and some help from other alts you can speed this up immensely.

The Caracal Setup I use for level 2's requires almost no training and is stupidly overpowered for anything it encounters -

Highs: 5xAssault Missile Launchers
Medium: 10mn Afterburner
2x Large Shield Extender
2x Shield Resistance Amplifiers (change as per mission needs)
Lows: 2x Ballistic Control System
Drones: 2x Hobgoblin
Rigs: 3x Medium Core Defence Purger

You don't need to train the skills for the rigs either, just contract the ship to your main, fit them and then send it back. Same applies for the Sigil - stick cargo rigs on it and get 16k cargo capacity without actually having to train Jury Rigging or other skills.

Anyway, between 2 accounts I now have 6 characters working on planets, now my task is to optimize the planet side interaction to reduce the amount of time I need to spend managing it - but that's another article.

Tuesday, June 8, 2010

Missions Need Fixing

And I don't mean in the way that the chorus of people complaining about loot nerfs, giant courier packages, low-sec missions and level 5 bug fixes are wanting, I mean a complete re-appraisal of the mission process from mission availability, mission design and rewards. In short, something that will probably never happen, but you know I can dream and maybe something might happen.

One of the biggest problems with missions in Eve is that the rewards in the hardest missions largely come from the bounties and loot dropped by the NPC's in the missions, this causes many problems indirectly -

* While agent rewards provide some incentive to find high quality agents in 0.5 sec space they don't make much of a difference compared to the loot/bounties, so pilots still congregate in the same hubs.
* It encourages players to reject missions where the bounties and loot are perceived to be poor e.g. drone missions.
* It encourages farming of missions such as Recon (1/3) where the rat bounties can be collected daily for aweek.

All three of these serve to reduce variety and therefore increase the percieved 'grindiness' of mission running, which in turn will no doubt lead to players losing interest more quickly.

NPC Spam in Missions is counterproductive, from a realism/roleplay (delete as appropriate) perspective it just seems bizarre that a single battleship can lay waste to so many pirates and odder still when you see that these mission rats have lower bounties and less loot and salvage compared to their random asteroid belt cousins. Plus there's the simple fact that mission AI is terrible, the rats almost never switch targets to defend themselves against your drones, or target the non-tanked ships in your fleet. So, I'd like to see CCP mission designers commit to revamping missioning in general to have fewer, smarter, more deadly ships. The bounties and loot can be brought back in line with what you'd expect from non-mission versions of the rats, but I'd anticipate that the mission rewards due to this would still be significantly diminished.

Reducing these rewards leaves more room to boost direct agent rewards in the form of cash, LP and other widgets that sometimes get gifted to bold pilots in exchange for risking their ships. It also gives the rewards system much more room to adjust mission rewards and make those 'crappy drone' missions much more attractive to the dedicated mission runner. As it stands right now there are a number of factors that determine mission rewards

* The Agent's Quality - higher quality means more payout
* The Security Status Of The Agent's System - agents in lower sec systems pay more
* The Average Time Players Take to Run The Mission - the longer players take to hand it in, the bigger the rewards.

This last one is one of the few ways that mission runners indirectly affect each other, it's most obvious with the 'Recon' series of missions, where players like to farm part 1 for a week before handing it in while parts 2 and 3 are quick flythroughs, so the agent rewards for part 1 are 10 times the rewards for the other segments. So when I'm working up standings with a low skilled research alt I'm always happy to see the Recon mission offered because all those people holding on to the mission for a week are pushing up the rewards I get for zipping through it in a shuttle. While farming a mission that respawns awesome rats every day is currently justified by the monetary rewards I don't believe this to be a great gameplay experience.

The Security status scaling in theory provides some incentive to mission in low security space, but it really doesn't have any real effect. Firstly we've already established that the agent reward part is pretty small compared to the loot and bounties, but even if we corrected that with the agent rewards dominating you'd still be left with a simple linear function. Now - let's do a very Eve thing and use a spreadsheet to make some graphs of risk and reward....

Umm well ok blogger wants to reduce the image quality, but nonetheless you can see 3 graphs here - Reward - linearly scaling as we head to lower security status, risk - a step function that jumps when we leave Concord's protection, and the Risk-Reward ratio. Of course I just pulled numbers out of my head here, I'm putting the risk in low-sec as 3 times higher, which is probably only true if you're a pilot who's using scouts and directional to stay on top of things. The point is that the risk reward equation shows this massive dip the instant you step into a 0.4 sec system. So Even if we fudge the numbers and make it that the risk:reward eventually exceeds the peak in 0.5 sec space there'll still be this trough which will act as a natural potential barrier discouraging missioners in low-sec.

I don't believe that missions in Eve should be forcing pilots into low-sec, I do believe that those pilots who choose to take the risks should be rewarded in a way which is consistent with the risk.

So yes, CCP developers could take all these variables and could try to come up with new equations to take account of these parameters, they could collect all sorts of internal metrics on mission running and analyze the data to come up with a reward model that's makes players happy.

Or, and this is where my discussion really departs from Eve's existing mission design, they could let the players decide by opening mission availability up to a market.

How could that work I hear you ask?

The current model of mission offers is a pilot flies to a station talks to an agent and then can either accept or reject a mission (possibly with penalties for rejecting too often). The mission selected is generally random within a set of constraints - such as those described in the courier mission dev blog. No other players get involved in your mission offers unless you ask them, and clearly some agents are very busy with hundreds of pilots working hard in mission hubs like Motsu and Dodixie. This model's very limited choice and lack of interaction with other players means a radical change is required to introduce market forces to the missioning market.

Aha! So instead a mission market is created, or rather more like a mission Contracts system. You'll be able to pull up a list of missions being offered by Agents and choose which one you really want to do right now. The list of missions will be viewable by everyone, but only a subset will be visible to players, a subset that depends on location and standing, you won't be able to see mission offers from an agent if you wouldn't have the standing to accept it. And if you're too far away to accept a mission then there's no point in displaying an offer - devs can of course add a bunch of new skills to improve mission visibility and acceptance range so players will have some more things that need the charisma attribute.

So as a missioner you can pick the mission that's best for you, no more twiddling your thumbs while waiting for your 4 hour timer to expire. So an agent has a mission on offer and you don't like it because the rewards are too low, or the location isn't quite right or any of a million other reasons. Maybe all the other pilots think like you and nobody wants to run The Blockade in pirate central, so, what does the Agent do? Well he does what he's programmed to do, he gets more needy as time goes on, he really needs *someone* to beat back these pirates, so over time he sweetens the deal, increasing the rewards, and maybe dropping the standings requirement in case there's a less reputable pilot that's prepared to take the risk. Eventually, the risk-reward equation for a mission makes it the best choice for some pilot, and the mission gets accepted so it's removed from the pool.

The whole process works something like a reverse auction, and nicely simulates a supply vs demand market, it rewards mission runners who travel off the beaten path and perhaps take some risks. Or perhaps that low-sec mission represents no risk at all, because the player is treating Eve Online like an MMO and has cultured good relations with other residents in the region. Good relations works to reduce risk and therefore bumps the risk-reward ratio in the right direction.

Now obviously some consideration needs to be given to how missions get added to agents rosters, it would be desirable to maintain a healthy supply of missions and ensure that missions would always be available, regardless of how many players are accepting them. My proposal is to have each mission offer respawn upon acceptance but with a different agent either within the faction or corporation, so the mission never disappears it just moves. It may move to an agent that's less desirable because they're in a less popular NPC corp or less accessible part of space. But what's less desirable to one pilot is better for another, so again this rewards those players who step outside the herd, sorry, that should be sloth, which is the collective term used for bears.

One thing that would remain untouched would be the storyline missions, that mechanic doesn't really fit into this model and doesn't need to. I've also been considering the issues related to probing pilots in missions, and would be interested in re-examining the old penalty for scanning ships in deadspace, but tempering this with the introduction of the ability of dedicated salvage pilots to scan down wrecks once the deadspace has despawned.

OK train of though ends here....

Tuesday, June 1, 2010

Things Getting Borked

We're getting some clues as to what CCP might do to 'fix' the PIpocalypse recycling 'exploit' from the latest changes to Singularity. The most up to date patch has disabled the ability to reprocess POS modules and new Broken versions of all the Tier 4 Planetary products have been added to the object database. So it seems entirely likely that anyone holding the unbroken products will find them replaced as of Wednesday (when the 1.0.1 patch is due to be released). The new broken items can't be used in building POS stuff any more, but they're not completely useless, they can be refined to 10,000 units of tritanium.

Now we have a clear idea where CCP is going with this, the question is how thorough they will be in following the trail of melted POS parts. Swapping all current tier 4 products with their broken versions will be easy, and it won't be much harder to cancel all active manufacturing jobs using the new blueprints and return the input materials to the installer. But people who were on the ball fast will have built and sold completed towers on the market already, at that point I don't see CCP going after completed products. So people that got in early and didn't get in too deep have probably made a tidy profit, and the copycats who rush to follow are more likely to be left with piles of worthless broken components. Then again, those components will eventually be 'collectors items' - so if I had one I wouldn't refine them.