Killmail Archivist

EVE Online theorycrafting and history

Autocannons, Artillery, and Balancing Fitting Requirements

Part three of my series on tracking is being temporarily delayed — some early readers gave me great feedback that I’m incorporating into the upcoming post, mostly in the form of some added graphs and animations.  In the meantime, I’ve got a bit of a rant.

It’s a well-accepted fact in the greater Eve community that the Tech-3 Tactical Destroyers are in need of rebalancing, with the Svipul and Confessor in particular being called out particularly overpowered and dominant in small ship combat.  I strongly agree with this; however, you’ll probably notice that most (if not all) of the really egregious Svipul fits are autocannon fits. Artillery Svipuls are rarely seen outside of specialized instalocking gatecamp fits, or medium-sized themed fleets.  Why is this? The artillery uses a lot of power grid and CPU to fit, so most artillery Svipuls are comparatively lacking in tank or damage.

You don’t see the same disparity in the other Tactical Destroyers; Hecates are equally popular in both blaster and railgun forms, as are pulse and beam Confessors, and rocket/LML Jackdaws.

Also, strongly favoring autocannons isn’t a unique trait of the Svipul.  In fact, the vast majority of projectile ships have this problem; autocannon fits generally have ample grid/CPU for taking extra tank or option modules, while artillery fits tend to be barebones.

  • The Cynabal can take nearly anything it wants if it fits autocannons; you can fit a full rack of 425mm ACs, one (or two) prop mods, a neut, plenty of tank, and all the damage mods you can fit. Artillery fits, on the other hand, are relatively tight. The Rupture, Stabber, and Vagabond are in a similar situation. (The Muninn, arguably, is the only projectile cruiser that actually has sufficient grid to take a full rack of artillery by default and still have ample room for tank/prop mod.)
  • Machariels with ACs can fit nearly anything in their option high, mids, and lows; you can’t even fit a full rack of T2 1400mm artillery on it without a fitting implant or mod, which is why most artymach fits use meta-4 turrets (or 1200mm) instead.  The Maelstrom and Tempest aren’t much better; on both ships, a full rack of 1400mm will use their base grid nearly perfectly, and a fitting mod is required to put on any other essential mod such as propulsion or tank.  A similar situation exists with the Hurricane.  (The Tempest Fleet Issue can do a full rack of artillery admirably, although it’s a tight fit.)
  • Artillery Rifters and Jaguars are mostly a bad joke.  The Wolf does well enough, but needs both an MAPC and a CPU to fit both artillery and prop mod.  The Firetail stands out as a good artillery ship, though.

This is exacerbated by the relatively small set of options at each ship tier:

  • Hybrid turrets get three options each for both short-range and long-range options, at each tier size; for example, Electron/Ion/Neutron blasters and 75/125/150mm railguns.
  • Projectile turrets consistently get three short-range options and two long-range options — for example, D180/220/425mm ACs and 650/720mm artillery at the Medium size.
  • Laser turrets vary: 3 short-range and 2 long-range for small turrets, 2 short and 3 long for medium, and 3 short / 3 long for large turrets.

Artillery has the highest mean cost in terms of CPU/grid per point of DPS, while AC has one of the lowest mean costs. (One reason why the 125mm autocannon is the defacto BNC gun on support ships.) However, CCP doesn’t want to release ships that are essentially incapable of fitting artillery. So, most projectile ships are given the bare minimum amount of grid/CPU needed to shoehorn in a rack of arty.  That same amount of grid/CPU ends up being a gross excess if you fit ACs to the same ship.

So, is it possible to make a Svipul that can be viable with an artillery fit without handing AC users a cornucopia of multiple MSEs, neuts, and damage mods?

An easy solution would be the one used by the Firetail: Cut the number of turret hardpoints in half, and add a double-damage bonus to the hull to keep the DPS in the same area.  The player T3D focus group proposed “Firetail-izing” the Svipul at one point, and it’s a decent suggestion… but I suspect that a better (but harder) solution exists.  Balance could be improved across the board for projectile ships by revisiting fitting requirements for turrets, eliminating the need for such hacks in the first place.

(It’d also be an opportunity to rebalance some of the rarely used turrets in the game. When’s the last time you saw a laser boat with Quad Light Beam Lasers?)

Turret Mechanics (Part 2) – Transversal vs Angular Velocity

(You might want to read Part One first.)

Imagine that you’re sitting still, at zero m/s.  You have three targets nearby:

  • A Rifter (red) is orbiting you at 500m.  They’ve got their AB on, and are keeping up a steady 1200m/s speed in that tight orbit.
  • An artillery Wolf (blue) is orbiting you at 20km with MWD.  They’re moving at 3600m/s.
  • A Jaguar (gray) is burning directly at you at 2500m/s with MWD on.


Let’s assume, for now, that you had artillery with infinite range.  The damage you do is determined entirely by tracking.  Which one do you shoot?

When you customize the set of columns that Eve puts in your overview panel, there’s four options for you to choose from:

  • Basic velocity — Shows how fast they’re moving in their direction, in meters per second.  Ignores anything you do.
  • Transversal velocity — Take your direction, and project the target’s movement in that same direction.  Take that part of their movement that’s parallel to your direction, subtract it from your speed, and display the difference.  Measured in meters/second.
  • Radial velocity — The rate at which they’re approaching or retreating from your ship.  Draw a direct line from your ship to their ship, and measure how fast it changes length; ignore the line’s direction.  Measured in meters/second.
  • Angular velocity — the rate at which they change angle to you.  Draw a direct line from your ship to their ship, and measure how fast it sweeps around your ship; ignore the line’s length.  Measured in radians/second.

Notice that angular velocity is different from the other three in the units it uses — in fact, it’s the exact same units that are used to measure the tracking speed of guns!

Returning to our example:

  • Because we’re not moving, all of the ships have the same transversal velocity as their normal velocity, meaning that the Rifter has the slowest transversal, and the Wolf has the highest.
  • If you have radial velocity on overview, the Rifter and Wolf have zero radial velocity (because they’re orbiting you at a fixed distance), while the Jaguar is approaching you at 2500m/s.
  • If you have angular velocity on overview, the directly approaching Jaguar has zero angle, and is easiest to hit.  The Wolf may be moving quickly, but it’s doing so from a great distance, so your turrets only need to move a pokey 0.18 radians/second to track it.  The Rifter, meanwhile, is moving at a zippy 2.4 radians/second.

Transversal is the most familiar to Eve veterans, because it was the only thing available in the early days of Eve.  In theory, it provides easy feedback to how quickly something’s moving away from you — if it’s low, then you’re moving parallel to a target, and should shoot.  However, transversal often lies to you, especially when you’re moving slowly or looking at distant targets; it only becomes consistently useful when you’re skirmishing with another moving target at a moderate distance.

If you’re trying to figure out what you can hit on the battlefield, angular velocity gives you a much more accurate view of the field, and it directly corresponds to the tracking of your guns.  If I know my artillery has a tracking value of 0.013 (radians/sec), then I know that I’ll be able to track and hit anyone on my overview who has an angular velocity less than 0.013, assuming they’re in my optimal range. [1]

In practice, I actually tweak my overview based on what I’m flying.  For artillery and other slow-cycling weapons, or for very slow ships, I use angular velocity.  When I’m flying Taranises and other ships with fast tracking and high speed, I use radial velocity, since that tells me how quickly something is approaching me.  I almost never have transversal on my overview anymore; the only time transversal is notably useful is if I’m trying to fly parallel to someone at a distance, and that rarely happens in modern Eve where kiting tactics are common and speed are essential.

How is this relevant to turret damage?  Remember from Part One: turret damage starts by calculating a chance-to-hit for your guns, and then making a dice roll to determine if you hit (and what the damage is).  Angular velocity and tracking go hand-in-hand in calculating that chance-to-hit.  We’ll discuss how it’s used in Part Three.

1. CCP, if you’re reading this: It’d be significantly either to use both the tracking speed in Show-Info panes, and the Angular Velocity overview column, if you multiplied both numbers by 1000.  Call it milliradians/sec.

1 Comment

Turret Mechanics (Part 1) – Roll For Initiative(dot)

The easiest way that I’ve found to explain turret mechanics is a throwback that should make immediate sense to tabletop RPG players:

Every time you activate a turret, the Eve engine calculates a chance-to-hit percentage. You then roll a 100-sided die (D100), and use the following rules:

  • If you roll a 1, you get a crit (wrecking blow) — a guaranteed hit for triple damage. [1]
  • If you roll any other number less than or equal to your chance-to-hit, you hit.
  • If you roll a number greater than your chance-to-hit, you miss.

When you hit on any number other than a 1, you take your roll, add 50, and treat the sum as a percentage of your turret’s base damage.  So, assuming that you’re shooting a stationary target at optimal range (100% chance to hit), each roll of the dice can produce a hit ranging from 50% of base damage (a glancing blow) to 150% of base damage (a penetrating/smashing shot).

As your chance-to-hit starts dropping, not only do you start having rolls that miss, but your highest quality shots are the first ones to get converted to misses.  For example, lets say that you have a 75% chance to hit a target; that means you hit on a roll of 1-75, and miss on a roll of 75-00.  That means that you will never hit for more than 125% of your base damage; you cannot get penetrating shots.  As it drops off, not only do you have more chances to do no damage on a shot, but your shots that hit will be of increasingly lower quality.

This is the main way that damage drops off with range, and it’s particularly important for alpha-strike doctrines; fighting in falloff doesn’t only reduce your fleet-wide average damage per cycle due to misses, but it dramatically depresses individual volley damage as well.  Likewise for shooting targets with tiny signature radii.

The other crucial part of turret mechanics is how we actually compute the chance-to-hit percentage.  That’s complex enough to be its own topic, and it’ll be in the next post.

1. Wrecking shots aren’t actually a guaranteed hit — you can only get a wrecking shot if you have a non-zero chance to hit, even if it’s very small. Imagine that you’re in an artillery Tornado with Quake loaded, and an AB Taranis is orbiting you at 500m; you have a near-zero chance to hit it, due to tracking, but it’s still a tiny fraction above zero.  Thus, wrecking shots are possible.  Now, imagine that the same Taranis is 225km away; you have a strictly zero chance to hit with EMP, even if both you and it are completely stationary.  You will never hit it, even if you roll a one.

Oversized Afterburners: Pros and Cons

(Yes, I’m back to writing!)

When fitting a ship in Eve, you generally pick a propulsion module that matches the size of your ship — frigates and destroyers use 1MN afterburners and microwarpdrives, cruisers and battlecruisers use 10MN modules, and battleships use 100MN modules. With one exception [1], you should never use a prop mod that’s smaller than intended for the hull.

However, there are occasionally good reasons to use a propulsion module that’s larger than intended, especially afterburners. (It’s particularly popular right now to use 10MN afterburners on the Tech-3 destroyers, the Svipul and Confessor.) Why would you do this?


First, an oversized afterburner allows you to move very fast, as long as you don’t have to make any turns. This can be surprising to newer players; if you look at the stats for a 10MN afterburner, it lists a 135% maximum velocity increase, compared to the 500% increase on a 1MN microwarpdrive. However, they have different thrust amounts!

Screenshot of 1MN MWD and 10MN AB.

The actual top speed given by a prop mod is actually proportional to both the maximum velocity bonus, and the thrust generated:

Vmax = ship base speed * (prop mod boost% * (prop mod thrust / ship mass))

So, for most frigates and destroyers, overheating a 10MN AB will give you a top speed equivalent to a non-overheated 1MN MWD! [2]

Secondly, using an AB allows you to ignore warp scramblers. Warp scramblers have two functions: they prevent a target from warping away, and they disable any MWDs on the target ship, reducing their ability to maneuver.

One way to keep mobile while scrammed is to fit both a MWD and an AB, and switch to the AB once you’re scrambled. (This is commonly called a “dual-prop” fit.) However, a properly sized AB won’t give you the same speed as the MWD — plus, it takes up a second mid slot, which most ships don’t have many of. Another solution is to use an oversized AB as your sole prop mod, which gives you speed similar to an MWD in a single mid slot.

Keeping mobile while scrammed is particularly useful for the Svipul; it doesn’t get a turret range bonus, so it almost always operates at ranges of 15km or less. (Confessors get a range bonus while in Sharpshooter mode, giving them a little more flexibility.)

Finally, ABs of any size do not have the “sig bloom” of an active MWD. If two ships with an MWD and a 10MN AB are moving at the same speed, the AB ship will take far less damage from missiles, and be hit less often (and for less amounts) by turrets. Thus, oversized ABs can act as a form of damage mitigation.


However, there’s also some significant downsides to using an oversized AB:

  • Every propulsion module has significant mass, which is added to your ship, similar to an armor plate. The 10MN AB is nearly four times as massive as an entire Svipul! While an oversized AB allows you to move at high speed in a straight line, it robs you of your agility. You have a massive increase in inertia when you use an oversized propulsion mod; you can’t turn on a dime, or hold a tight orbit, or sling-shot a long range tackler.
  • Due to the increased mass, oversized afterburners also produce less acceleration than a MWD. Even though they have similar top speeds in a straight line, an oversized AB will take 15-20 seconds to reach that speed, compared to 3-5 seconds for a properly sized MWD. This is particularly painful if you jump into a gatecamp and need to reapproach the gate.
  • An oversized afterburner requires nearly three times the power grid of a properly-sized MWD to online. This usually means compromising some other aspect of your fit: giving up tank, firepower, or utility modules.

One of these downsides limits the number of ships you can use this on. Because you can’t turn quickly, it’s nearly impossible to manage your angular velocity in a ship with an oversized prop mod; so, you generally don’t want to use them on turret ships. Save oversized afterburners for ships with bonuses to drones or missiles.

Also, this technique isn’t limited to frigates and destroyers; there are a number of cruisers that perform well with 100MN ABs fitted instead of 10MN MWDs. (The main candidates here are drone cruisers such as the Gila and Ishtar, although it can work well for missile cruisers as well.)

Update: Added a mention on sig bloom mechanics, per some smart eyes on Reddit.

1: Supercapital ships will often carry an X-type 100MN MWD. It won’t actually allow them to move quickly in a straight line, due to the monumental mass of their ship; however, they can use it to speed up alignment times when warping to celestials, or to use it as an E-brake to prevent being bumped out of a POS.

2: This equation also explains why undersized propulsion mods generally don’t work well; the thrust that they produce is overwhelmed by the larger mass of the ship.

Fixing the Nestor

The Nestor has been a problematic ship practically since its creation; it’s the least used of all the faction battleships.  Even the Barghest, despite being added mere months ago, has more kills and more losses on zKillboard than the Nestor.  CCP recently announced a small change to the Nestor, adding refit capabilities to it; however, I don’t think that this will solve the Nestor’s issues.  Its popularity stems from two problems with it — a smattering of mismatched bonuses, and horrible fitting problems.  Let’s break them down.

First, the Nestor’s bonuses imply that it should be useful at pretty much everything, when most Eve players learn at a very early age to favor specialized ships over jacks of all trades.  Running down them:

  • 4% bonus to all armor resistances per level of Amarr Battleship
  • 10% bonus to drone hit points and damage per level of Gallente Battleship
  • 50% bonus to Large Energy Turret optimal range
  • 50% bonus to Remote Armor Repairer strength
  • 200% bonus to Remote Armor Repairer range
  • 50% bonus to exploration probe strength
  • +10 bonus to Relic and Data Analyzer strength
  • Finally, the Nestor has an absurdly low mass, allowing it to go through wormholes with low impact.

This hull tries to do everything.  It can be a dedicated logistics boat, a split logistics/drones RRBS platform, a laser gank boat, a drone boat; it can even be an exploration platform.  (The last of these is the most laughable — nobody sane is going to use a battleship in data/relic sites.  In lowsec/nullsec, you’re risking an easily-probed billion-isk hull in that site; in high-sec, someone else is going to finish off the site while you’re slowly motoring between cans.)

The few successful fits for an Nestor rely on ignoring almost all its bonuses in favor of one or two — either it’s a brick-tank brawler that ignores the remote rep bonuses, or it’s a plus-sized brother of the Oneiros, for use in high-alpha-strike environments (incursions and C3/C4 wormholes).

However, even once you’ve picked a role to focus on, the Nestor’s anemic fitting stats and relatively sub-par bonuses bite you:

  • The short remote-rep range bonus is risky, as is the capacitor needs.  A 25km range is a bit short for logistics use, and the Nestor’s comparatively slow speed (just under 1km/sec with MWD on) and long lock times means that you’ll need to pulse your MWD often to stay in range of the ships that you’re repping, if you don’t have them orbiting you.  And with no capacitor bonus for either MWD or reps, cap stability becomes an issue.  (Most of the logistics-oriented Nestor fits out there invest in an deadspace X-type MWD to stay stable, or are reliant on a full set of cap rechargers or multiple injectors.)
  • Drone damage output is questionable due to having only six lows, on a ship with an armor tank.  If you keep to a relatively small 4-slot tank (suitcase, two EANMs, 1600mm plate; a risky gamble for a 1B+ faction battleship) then you only have room for two Drone Damage Amplifiers.  Furthermore, while it may have the same damage bonus as an Ishtar or Dominix, it doesn’t have the tracking/optimal bonuses of those hulls.  The combination of short range and low mobility means that you’re forced to use heavy drones, or must stick to low-damage long-range sentries such as Curators.  You can compensate for some of this with using your ample six mids on ODTLs — and you’ll have no choice but to do so, really, due to the fitting issues:
  • The Nestor has absolutely abysmal grid, which particularly impacts laser builds.  With a hair over 14k grid at Engineering V, you can’t fit a MWD, a 1600mm plate, and a full set of Mega Pulse Lasers on it.  You’re a whopping 27% over, even with Advanced Weapon Upgrades V.  And that’s before you even consider putting on an injector, option highs (neuts, smartbombs, remote reps), or a second plate!  Due to this grid crunch, all laser fits for a Nestor either spend multiple rig slots on ACRs, or downgrade to Dual Heavy Pulse Lasers (terrible damage output) or use faction lasers (expensive, and no Scorch).  On top of that, you can fit a maximum of two Heat Sinks due to the same problems as DDAs.  For all these reasons, combat Nestors usually focus on drones, and fit 2-3 ODTLs in their mid slots because they have nothing else viable to stick in there — there’s not enough grid for a MJD or a second injector.  And in general, what do you do with six mids, when the ship is generally armor tanked?  It’s too slow to tackle something, it has no ewar bonuses, and it struggles to fit multiple prop mods.
  • The ship’s low base shield HP, high sig radius, and low mobility makes shield fits questionable.  T2 CDFEs and two invulns are needed to push a shield fit above 100k EHP, and two nanofibers barely get you to 1400m/s under MWD.  In comparison, a single 1600mm plate gets you above 100k EHP even before Trimarks are added.  Also, it has a gigantic signature radius; most shield tanked ships are 425m or less.  A shield-tanked Nestor with CDFEs would have a sig radius of 520m when slowboating and over 3000m when MWDing — that’s bigger than a carrier!

A jack of all trades is a master of none.  The Nestor isn’t particularly good at any of its roles, and hilariously bad once it starts to multitask.  I’ve seen effective uses of it as a logistics platform in C4 Cataclysmic Variable wormholes, but it also required a fairly expensive fit to be useful there: A-type hardeners and an X-type MWD.

Where to go from here?  Given the opportunity, I’d change three things:

  • Remove one of its mid slots, and add a low slot.  This gives two benefits: it allows it to have competitive drone damage with an Ishtar when using a small tank, or competitive tank with smaller damage.  It eliminates “fill all your mids with cap rechargers / ODTLs” for all roles, forcing players to make meaningful choices about their mid slots.
  • Increase its base power grid to 13000.  This bumps its final grid to around 16300 — enough that you can fit a a MWD, a heavy injector, a single 1600mm plate, and a full rack of MPL turrets, but with almost no grid left for option highs/mids.  It allows you a basic ship, but you will have to make sacrifices (in the form of downgrading to DHPLs, using ACRs, or using deadspace/cosmos modules) in order to fit meaningful additions such as extra plates, MJDs, or neuts.  Again, the goal is to make the ship viable at a few roles, but to not give the players everything they want; you want to force the player to make choices about how they fit this ship.
  • Abandon the exploration role bonuses, replacing them with a new role bonus: a 50% reduction in capacitor usage for remote armor reps.  There’s very little reason to run data/relic sites in a Nestor, when the same LP could be used to purchase a Stratios or Asteros instead.  The player base has responded well to using the Nestor as an RRBS or plus-sized logistics platform; let’s allow this to be done without filling your mids with capacitor rechargers and dropping hundreds of millions on deadspace MWDs.  Right now, running more than two reps demands use of an injector or a full rack of cap rechargers, even before you factor in armor hardeners, MWD pulses, and other users of cap.

A final option has occured to me, but I’m not sure if it’d be too powerful: Don’t give the Nestor a covert ops cloak, but do allow it to take Black Ops bridges — or even give it a jump drive that only locks onto covert cynos.  Right now, there is only one remote rep platform that can take a Black Ops bridge, and it’s a shield logistics that’s only available in tiny numbers (the Etana).  Adding an armor option would give extra reason to use Black Ops ships as combat platforms rather than as miniature titans.

If you’re intent on making a jack-of-all-trades ship, it might as well be good at a few of them.  Until then, nobody’s spending 1B+ on the hull, and giving it yet another trade (in-space refits) isn’t going to improve matters.

History of a Ship: Echelon

(Apologies for the radio silence — a medical emergency killed most of my spare time for the last week.)

CCP likes to do limited-edition ships whenever Eve hits an anniversary, or has a major feature launch. The last few releases have mostly been reskinned versions of T1 frigates; however, even when they were unique ships, most of them either were never useful to begin with (the Primae) or ceased to be useful due to balance changes.

So, let’s talk about one that is useful: the Echelon.

Echelon hull

An Echelon was awarded to every account who logged into Eve during the release of the Incursion expansion in late 2010. This expansion added Sansha incursions to the game; prior to the expansion’s official release, CCP actors did a number of live events to drum up excitement for incursions.

During these live events, a wormhole would appear near a planet, and Sansha ships would spawn around the wormhole and “steal” citizens from the planet to turn into slaves. (This is why the first name of all Incursion rats are the name of a solar system in Eve; that’s the system they were taken from.) The wormholes were targetable, and CCP hinted that running data analyzers would have some effect on it. I assume that this was a complete lie, and CCP just wanted some people to show up in scanning ships — more targets. :)

That said, the Echelon was released with this in mind: a dedicated data-analyzer ship. It has a single mid slot, and no low/high/rig slots; it’s only capable of fitting a single module in that mid slot, the Purloined Sansha Data Analyzer. (One was handed out with each Echelon.) At the time it was released, the hacking minigame didn’t exist, and data analyzers simply had a chance per cycle to unlock the can; the Echelon’s scanner was simply average.

However, when the hacking minigame was instituted and all data analyzers were redesigned, the Echelon and its custom analyzer was included in that redesign pass, and became fantastic. In particular, until you have Hacking V trained and can use T2 Data Analyzers, the Echelon is the best ship in the game for Data Sites — you need a Covert Ops or Astero with a T2 Analyzer and two hacking rigs to equal it.

Listed as a virus strength / virus coherence pair: (bigger is better for both numbers)

  Echelon Heron
CovOps & Astero
(hacking rigs)
Hacking I 40/90 25/50 30/70
Hacking IV 40/120 25/80 30/100
Hacking V 40/130 35/110 40/130

Even a perfectly skilled hacker, with the best ship and modules in the game for it, can only tie the Echelon.

Of course, there’s a flip side: the Echelon can’t fit a probe launcher, or a propulsion mod, or any sort of tank module. So, you’ll have to probe out the site with another ship, and then switch to the Echelon to hack it. But that may be an acceptable tradeoff for quiet systems and players with low SP.

If you don’t have the Echelon, they’re quite cheap, despite being a limited-issue ship; a ship and its matching module can be found at Jita for 7.5M isk total. Quite cheap, given that the nearest comparable ships for hacking cost 20M or more each.

Update: A sharp Reddit commentor also points out that, if you have multiple Echelons sitting around, you can get a rather funny set of items by reprocessing the Purloined Analyzer.

Modules With A Story: Hellhound Drones

(Apologies for going quiet for a few days — went out with the family for the Fourth of July!)

If you asked people what the rarest drone in Eve is, most people would talk about the “Augmented” drones (fast drones with split damage types that can be built with parts from Rogue Drone cosmic anomalies), or the Gecko. Incursion runners will probably know about the Shadow, the rare Sansha fighter-bomber that can be acquired from incursions.

But there’s a truly rare drone out there that very few know about: the Hellhound.

Entity, kindly showing off some Hellhounds in a equally rare ship.

The Hellhound is one of the rare cases of an developer item in player hands.

If you download the Eve Database and examine the item types table, you’ll find many items marked as unpublished — ammo, ships, modules, and more. The primary effect of being unpublished is that the item cannot be searched for in the Market, or in an item type search in Contracts.

Most unpublished items are simply unused items and unfinished artifacts from the CCP development process: They’re an idea that CCP devs liked enough to try implementing on a test server, only to find that the item wouldn’t really work out well in live play. The idea would be canceled, and was never launched in TQ, but the remnants of it remain in the database.

There are also a few unpublished items that were once part of normal play on TQ; these items were disabled by CCP for various reasons. If you still own them, they remain in your inventory, but CCP marks them as unpublished so that they can’t be sold on the Market — mainly so that newbies don’t accidentally spend billions on a useless item. This is the case for deployable minefields [1], most of the canceled skillbooks, and some old collector’s items.

The Hellhound drone, however, qualifies for both of these. It was a developer experiment for an ultra-powerful variant of the Ogre, and was never meant to get into player hands; however, due to a typo, it was accidentally added to the drop table for a structure in a certain mission. A few days later, when players started posting on the forums about them, CCP promptly fixed that typo and removed the source of them, never to be seen again. However, they allowed players to keep the drones that had dropped up to that point.

In 2012, CCP changed their mind, and attempted to remove the Hellhound (and other unpublished-but-extant items) from the game entirely; however, they ended up adding them back due to player outcry.

I don’t know if anyone who owns a Hellhound has actually bothered using them in PvP; until this spring, they were simply a clone of the Ogre II heavy drone, with slightly higher DPS. When the Kronos expansion came and readjusted all drone damage upwards (to compensate for the nerf to the Drone Interfacing skill), the Hellhound was left out of that adjustment pass; as a result, it now has significantly lower damage than any of the normal drones in use today, making it simply yet another collectible artifact of Eve history.

1: There’ll be a post in a few weeks about minefields. They once existed, and they weren’t nearly as cool as they might sound.

Missiles in Early Eve, and Cruise Missile Kestrels

If you had the fortune (or misfortune, as it sometimes was) to play Eve shortly after launch, you’d remember that missiles used to be very, very different from today — and different from any other weapon system in the game at the time.

Missile launchers existed in similar varieties and meta-levels as today, but had names that denoted their capacity.  For example, all heavy launchers had a 50m3 capacity for ammo, and all had names starting with H-50.  (For example, “H-50 ‘Arbalest’ Heavy Launcher”.)  Likewise, you had the M-12 Standard Launcher, the A-3 Assault Launcher, S-110 Siege Launcher, and so on.  The crucial difference between those days and today is that any missile launcher could load and fire any ammunition that it could fit in its bay.

For example, a single reload of a plain M-12 Standard Launcher I could hold your choice of:

  • 40 rockets
  • 12 light missiles
  • 2 heavy missiles
  • 1 cruise missile

Regardless of what missile type was loaded in it, it’d fire every 14 seconds before skills.  The same was true for larger launchers; a meta-0 siege launcher could hold 10-20 cruise missiles / torpedoes or 40ish heavy missiles or hundreds of light missiles/rockets, but always fired every 20 seconds before skills.  In general, smaller launchers featured faster firing rates (meaning higher spike DPS) but had less flexibility and more frequent reloads.

Assault launchers were particularly different back then: they had frigate-sized fitting requirements, and were intended to be a midpoint between the standard launchers and the rocket launchers of the era.  The rocket launchers in retail launch and Castor had a tiny capacity and could only fit 3-4 rockets per reload, but had miniscule fitting requirements and fired incredibly fast; assault launchers were roughly in the same space that rocket launchers occupy, with the added bonus of being able to load 4-5 light missiles if you needed to engage at long range.

With missiles working like this — and with the fact that missiles of this era were AOE weapons with splash damage — the dominant frigate of the era was a Kestrel with four standard launchers, loaded with cruise missiles.  Loading cruise missiles meant that they could only fire off a single volley before reloading, but it allowed them to put out extremely high spike damage: A single cruise missile of the era did 300 damage to targets at its center, meaning that a single volley of four cruises was sufficient to one-shot frigates and smaller industrials. [1]  Stow a few lights in the cargohold just in case, and you were good to go!  Needless to say, cruise Kestrels were fantastic for high-sec suicide ganking as well, where they were known as “KamiKestrels.”

Eventually, during the Castor Period, missiles were refactored into the groups that we know and love today.  Some collectors still have Kestrels squirreled away with cruise missiles loaded in their launchers; however, using them in combat is considered an exploit.

1: At this time, all ships had roughly 1/3rd of the base HP that they do today, across all sizes and classes.  CCP would eventually bump up the base HP for all ships during Red Moon Rising, in the interest of making combat take longer and be less brutal; from there, most ships have slowly crept upwards in base HP over repeated rebalances.


Blog Banter: Understanding How To Fit

Kirith Kodachi’s latest Blog Banter is a topic that’s very close to my heart:

“Obviously that is a not just a bad fit, its horrific. But the guy might not know any better. We get these all the time circulating social media and corp/alliance chat. How do we educate players on fitting? This guy has been playing four months and can fly a BC, but has no idea how to fit one. What could be done to help bro’s like this?

Furthermore, what (if any) responsibility do veterans players have in finding these players and instructing them on the finer arts of ship fitting? If it exists, does it extend beyond them into teaching PvP skills, ISK making skills, market skills, social skills, life skills…

And another question you can think about is this: do purposely wrong fits, aka comedy fits or experimental fits or off-meta fits, offend you or your corp? Would you, like Rixx Javix when he was in Tuskers, face expulsion for fitting your ships differently than the accepted standard?”

Multitasking: It’s Not Just A Skill

I think a large part of this problem originates in a new player’s attitude towards their ship.  Namely, I’ve noticed that new players tend to identify with their ship in a way that older players don’t.  A new player might say “I’m a Drake pilot” or “I’m a Hulk pilot,” where a veteran pilot would just see ships as merely a tool to be used.

This is one of the reasons that we tend to see ships out there attempting to multitask — that is, to have a mix of short and long range, to have utility modules like probes and data/relic analyzers, to mix shields and armor, and so on.  These new players are attached to their ships, especially if it’s the first ship they’ve ever flown in a given class.  They’ve scrimped and saved for that ship, and they probably only have one of them; they’re going to use them for everything!  (Plus, they want to feel like they’re progressing in the game by moving to ever-larger ships.)

Having more than one ship, each fit for a different purpose, requires lots of spare isk — an amount that boggles the newbie mind.  So, instead, they have one ship, and they try to fit it to do everything they’ve seen in the game… and die horribly.

That’s the first issue: You have to teach new players to break that temptation to multitask, and help them build up an early supply of ISK that allows them to afford to specialize.  Nosy Gamer was close to this when he described older players keeping spare ships as a form of “pre-paying the death penalty“; new players get attached to specific hulls precisely because they haven’t done this, and they don’t think of a ship as a canvas with which to create utility and specialization.

Once you can get a player thinking in terms of specialization, they’ll automatically avoid some of the worst sins of poor fitting.  They’re less likely to mix weapons, less likely to dual-tank, and so on.  That little bit of mindset goes a long way.  Not all the way, though…

No, We Don’t Really Need EFT In The Game

A few of the early responders to the blog banter have proposed implementing an equivalent to an offline fitting tool (EFT/PyFa/EveHQ) directly in the game.  This would be a nice feature; however, adding it wouldn’t really solve the problem of bad fits.  After all, BattleClinic’s fitting section is full of the wreckage of terrible fits that were lovingly handcrafted in EFT!

Fitting tools, whether offline or in-game, are fantastic for comparing the performance of different fits, and for experimenting with small subtle changes before you go on a Jita shopping trip.  However, neither of them really guide you on how to create a fit; they’re just a drawing board!  When an experienced player creates a fit for a ship, there’s a mental process that they go through:

  • Picking a role for the ship
  • Determining what modules are 100% necessary for that role (i.e. “I need a scrambler, I need a MWD, I need at least 300dps”)
  • Getting those modules to fit
  • Filling in the gaps (i.e. “I have two spare lows after meeting my core roles; I’m a bit thin, so let’s add a suitcase.  I’m a bit slow, so let’s add an overdrive.  I have a spare mid; should I put in a web, a TD, or an injector?  Maybe I’ll skip on the injector, and fit a nos instead?  etc.”)

Giving new players the ability to replicate that mental process is the real meat in the center of the fitting nut.  Everything else is just tiny tweaks and mechanics.

I’d actually love to see a section of the new player tutorial that guides people through coming up with a fit for a ship, comparing modules, and fitting up the ship.  Barring that, Von Keigai has a great idea on having a fitting “checker” that scans your ship for a list of venial sins, and I’d love to see this feature added to Eve.  (I’d add one more entry to his list of sins: Players should get a big red flashing warning if they’re less than 6 months old and attempt to undock a ship with an officer module fitted.)

Alliance Doctrines

That said, any alliance worth its salt will have a doctrine — a set of pre-approved ship fits to use.  A standard doctrines have a couple purposes, such as simplifying logistics and training programs; however, the most important tool they serve is allowing an FC to create a checklist of ships to have, and be able to figure out their ability to gank and tank with just basic multiplication.  For example: “I know that our doctrine Drake does between 450 and 600 dps, depending on pilot skill.  Therefore, if we have ten Drakes in fleet, I know that I’ve got 4500-6000 dps.”

The double-edge sword of a doctrine is that players can fit up a doctrine-compliant ship and contribute to a fleet without understanding what makes that fit good (or bad).  And, unfortunately, understanding doctrines is crucial: it’s important for making new players into vets, and it’s important for keeping alliances relevant.  Doctrines get stale if they’re not constantly questioned and improved.  Also, many alliances’ doctrinal fits are great for large-fleet PvP, but are not suitable for soloing or missioning or exploration; players need fits for these roles as well.

Doctrines are a useful tool, but you have to make sure that your players understand the purpose of the doctrine, and be willing to debate on changes to them, and be capable of living outside of doctrine when they’re not in fleet.

The same mechanisms of teaching people mechanics and thought processes can be done for anything: market trading, wormhole running, optimizing mining, optimizing level 4 mission running, or even out-of-game life skills.  If your alliance isn’t full of players teaching each other, you’re in a bad alliance.

Oddball Fits

Are you kidding?  My main claim to fame is a PvP video where I exclusively fly combat-fit covert ops ships.  (Nobody expects the Spanish Inquisition… or a Buzzard with rockets, scram, web, and two TDs.)

Creativity, surprise, and the threat of the unexpected is what keeps this game interesting.  And if you fail, the only thing lost is the isk in your ship.  If your alliance truly cares about that, you’re probably in a shitty alliance.  Boredom is the universal killer of alliances in Eve; keeping your players logged in and entertained is far more important than keeping your stupid killboard green.

If you have a rank newbie who is flying bad fits because they don’t know any better, then help them.  Teach them how to fit competently, teach them how to think about ships as tools, and make the newbie into a vet.  But if you’ve got a bored veteran who wants to try killing canflippers using a tackle-fit miner, or wants to go welp dual-rep battleships into the local gate camp, or go solo roaming in faction cruisers?  Let them.  Whether they win or lose, you’re letting them build a story that they’ll tell for years, and you’re keeping them interested in Eve.

So, here’s an action item for you readers: Next time you see a killmail that’s an obvious shitfit, go ahead and laugh for a bit — and then convo the pilot.  Ask them if they understand why they’re suboptimal.  Explain.  Teach.  Help that pilot become better.  It doesn’t take long, and it means they’re more likely to keep playing the game… even if that means that you get an opportunity to kill them again.  :)


The Server Tick (or, “WTF, Why Didn’t My Point Turn On?!”)

It’s a scenario we’ve all experienced: You’ve got a fleet on a gate (or a station undock), and a hostile’s just appeared.  Everyone drops drones and assigns them to the fastest locker in the gang, waiting for “point!” to be called on voice comms.  The hostile de-cloaks, and it’s a fast ship.  Everyone starts locking, and a few of you even appear get a lock on him… but before your point activates, the hostile warps away.

Congratulations: You’re yet another victim of the 1Hz Destiny Server Tick, the processing loop that forms Tranquility’s beating heart.

What IS the Destiny Server Tick?

There’s a few different parts of the Eve Server. [1]  But, when it comes to shooting things in space, the most important part of them is Destiny.  This is the part of the Eve server/engine that simulates grids — it handles advancing the physics simulation of Eve (ships and drones and missiles flying around), and communicating things to the client (physics updates, module activations, target locking, health/status, and other visual elements).

Each solar system, on the server, has a Destiny processing loop associated with it.  At all times, this loop is running, and the code inside this loop looks something like this:

  • Pre-Tick Work:
    • Move people between grids, as they warp in/out or cross grid lines
    • Manage client actions (“I want to warp to gate X”, “I want to orbit object Y at 2km”, “I want drone X to return to my bay”)
    • Manage server actions (“Drone X just exploded”, “Object Y just spawned a missile”, etc.)
    • Send packets to clients containing everything they need to know to render the game for the next second
  • Physics (aka “Evolve”):
    • Simulate ships/drones/missiles moving around on grid.  Handle bumps/bounces.
  • Post-Tick Work:
    • Clean up, and go to sleep until the next tick.

The goal of the server is for a server tick to run exactly once per second (or, 1 Hz).  If it takes less than a second to do a tick’s work, then everything’s great, and we go to sleep for the rest of the second.  (Or do a tick for another system on the same node.)  But, when it takes longer than that, we have a hard choice to make: either put off some work until the next tick, or come up with more time to process the tick.  And that’s where Time Dilation comes in — it stretches player activity out so that we have several seconds to do one tick’s worth of work.

Does This Mean Eve Is Actually Turn-Based?

Nope!  Quite the opposite, in fact.

Destiny does a lot of work on the tick, but it can do a lot of work outside the tick as well.  When client messages (including “start locking a target” or “turn on a module”) are received by Tranquility, it typically processes them immediately, and will even tell your client the result if it can.  For example:

  • If you’re trying to jump through gate, it will mark you as “jumping through the gate” the instant that it receives the message, and will send back an acknowledgement right away that you’re jumping.  However, everyone else on the grid won’t know that you’ve started jumping until the next server tick.
  • If you activate a turret module, it’ll immediately calculate the damage, and immediately apply it to the target.  It’ll send back a packet right away, telling you, “I activated this module for you.”  However, you won’t receive information about how much damage was dealt until the next tick!  And neither will the target!  Yes, this means you can be dead for up to a second (or up to 10 seconds, in the case of severe TiDi) and not know it yet!

Also, some tasks are completely divorced from the tick and can happen at any time, subject only to the latency between you and TQ.  (For example, starting a scan with probes.)

The most important place where this affects you is a pair of closely related actions: aligning/warping, and locking a target.  In both cases, they get “rounded up” to the next nearest server tick!

Rounding Up: Warping and Locking

The biggest example of rounding happens in align times.  When you click “Warp to X” in the UI, your client immediately sends TQ a message: “Align me to celestial X, and warp me as soon as possible.”  TQ immediately acknowledges this request, and queues up a work item for the next server tick.  At the next server tick, it switches your state from Impulse Mode (normal grid flight) to Pre-Warp Mode.  Then, immediately after that, and for each server tick afterward, it’s going to do this check: “Are you above 75% of max velocity, and in the proper direction?  If so, immediately switch you to In-Warp Mode.  Otherwise, continue to move on grid as needed to align.”

Because the check for 75% is done on the tick, your align time gets effectively ’rounded up’ to the nearest tick — if EFT lists your ship’s align time as 4.5 seconds, the actual time needed to enter warp will vary between 5 and 6 seconds.  If this sounds bad, don’t worry; it’s even worse for a would-be tackler.

Imagine that you’re sitting on a gate, with a warp disruptor “primed” (i.e. pre-activated, before you have a lock).  Something jumps through the gate, and you want to lock it and turn on your point.  You have a 3.5 second lock time, and you have a 75 millisecond latency (aka a 150 ms “ping” time, since pings measure round-trip) between you and Tranquility. [2]  What needs to happen?  In order:

  • TQ processes the tick in which the target decloaks and starts aligning.  It sends a network packet to your client that tells you, “ship X just appeared on grid, and it’s aligning towards Y.”
  • 75 ms later, the “target is decloaked” packet successfully crosses the Internet, and your PC receives the packet, processes it, and the target appears on overview.
  • Your eyes and brain see the ship appear on the overview, and get your finger to click.  Let’s say this takes 150-200 ms to do. [3]
  • Your PC sends a network packet to TQ, saying “please start locking target X.”  Another 75 ms to cross the Internet!
  • TQ receives your “lock it” packet.  It immediately starts counting off lock time, queuing a “finish locking” work item to trigger in exactly 3.5 seconds.  It also immediately sends a packet back, acknowledging that you’re starting to lock.
  • 3.5 seconds later, outside of the tick loop, the work item fires.  TQ says, “Okay, mark down that they’ve marked ship X (for preventing cloak).  At the next tick, tell the client that they’ve successfully locked.”
  • At the next server tick — which could happen instantly after, or could happen up to a second later — TQ sends a packet to your client, saying, “You locked ship X; it has this much HP on shields, armor, and hull.  By the way, it’s still aligning towards Y.”  That packet takes another 75 ms to cross the Internet!  (At the same time, it’s sending a packet to the pilot of X, notifying them of a yellowbox.)
  • Your PC receives the “lock finished” packet.  It sends TQ a packet saying, “OK, activate my warp disruptor module on X.”  Yet another 75 ms to cross the Internet!
  • TQ receives your “tackle it” packet.  It immediately activates the packet, and queues the “You warp scrambled X” message to be delivered at the next tick.

That’s a huge list!  Most importantly, it requires four trips across the Internet between your PC and the server.  At a minimum, it rounds up to the next server tick — and it can be worse if you’re on a high-latency connection. From your 75 ms perspective, it took you four seconds to lock; adding an additional 100 ms of latency would add a full second to your final point activation time, making it a five-second lock!

Practical Application: Instalocking Gatecamps

These rounding effects are particularly painful when it comes to very quick-locking ships, and very quickly-aligning ships. The 1Hz tick of the server produces some nasty thresholds when it comes to gatecamps.  Given the above information, it’s pretty easy to write a tool that plays with delays in server tick and effects of latency to TQ, and simulates ships trying to lock other ships.

Imagine that we have a Keres with two sensor boosters on it, both scripted for scan resolution.  This gives the Keres an advertised lock time for of 1.2 seconds for most interceptors.  For comparison, the official align time for most interceptors is between 1.9 and 3.0 seconds, depending on fit and player SP.  In theory, the Keres should be able to catch all interceptors with plenty of time to spare.  Does it?

In practice, due to latency and server tick rounding, it doesn’t!  The threshold is two seconds; any interceptor with an align time of less than two seconds will get away from the Keres.  (At best, the Keres will appear to lock them, but the point won’t activate before the interceptor can start warp.)

So, let’s look at nano-fit interceptors.  What lock time do we need for the Keres to catch a ship that aligns in 1.9 seconds?  This answer depends a lot on your latency to TQ:

  • With a 100 ms latency to TQ, the Keres needs to be officially able to lock them in less than 0.725 seconds — achievable with three sensor boosters.  And, again, this is highly thresholded; anything below 0.725 seconds will catch everything, and anything equal to or above 0.725 seconds will miss everything (due to the point not activating in time).
  • With a 150 ms latency to TQ, the Keres needs to be able to have an official lock time of under 0.625 seconds to lock and point a 1.9 sec-align interceptor.
  • And so on.  In general, each 50 ms increase in latency to TQ requires a 100 ms (0.1 sec) reduction in lock time to compensate for it.

To put it bluntly, if you want to catch instawarping interceptors, the most important part is living in London!  That, or have a lot of remote sensor boosters — which, unfortunately, are subject to stacking penalties.

Can We Fix This?

Ironically, we can, and not by nerfing interceptor mobility.  Increasing the server tick rate to run twice per second, instead of once per second, completely eliminates most of these distortions — and, far more importantly, it makes playing Eve a much better experience for people on high-ping connections such as Australians / New Zealanders.

The server tick time in TQ is actually a compile time switch; flipping it is as easy as recompiling the server and deploying it.  (And, in fact, this has happened in isolated parts of TQ, during the “inverse TiDi” mechanic of Alliance Tournament.)  However, don’t expect it to happen any time soon.  Doubling the server tick rate means roughly doubling the CPU load for each node.  In a world where Jita regularly has TiDi kicking in just from normal player activity, bumping up the server tick rate is probably a non-starter.

Alternatives include sending “prime my module” packets to the server (to eliminate one of the roundtrips), or varying server tick frequency with load (i.e. Jita/Amarr/Rens is probably just fine with 1Hz ticks, or even 0.5Hz).

1: Can any CCP employees, past or present, ping me and explain how Michelle and Macho got their names?

2: For where I live (northwestern United States), a 150-160 ms ping to TQ is typical.  Australian players can have 300 ms or more.

3: Reaction times (for a simple ‘I saw something, immediately flex a muscle’ response) will vary between 150ms and 200ms, depending on age and mental agility.  If you have to select between two targets and pick one to tackle, your brain’s pattern recognition systems have to kick in, taking 400ms or more.