Talisman distribution, Pareto efficiency and “why can’t I hold agro?”

Talisman Distribution

Pareto Efficiency and “Why can’t I hold Agro?

This is the second part of a three part essay on gear selection in the Secret World. In the last essay, I discussed the fallacy of asymmetric progression and the problems that can hinder potential of talisman in your DPS output. (Go to the First Essay in the series on gearing)

In this essay, we will look at talisman load-out configurations.  How many talisman and of what kind, should you wear based on your role?  In the next essay, we’ll look specifically at glyph signet choices.

Pareto Efficiency?

One activity familiar to any student of economics is the “guns or butter” problem.  In it, a nation is positioned as capable of making two goods, guns or butter, and has to decide on the allocation of resources to each.  The nation could produce 100% guns and thus no butter or it could produce 100% butter and thus no guns.  Or, more likely, it can produce some of each.

The logic of “how many of each” was addressed in the works of Vilfredo Pareto, providing us with the Pareto optimization curve (see link for an example).

OK, So I have Guns in TSW, but no Butter?

TSW gives us a similar two factor trade off with our talisman.  Actually, its a three factor trade-off between hitpoint gear, attack power gear and heal gear, but for many builds and roles only two factors are relevant.

  • For the typical tank, the trade-off is between +hitpoints (survivability) and +attack (threat generation via DPS).
  • For the typical healer, the trade-off is between +heal (heal throughput) and +hitpoints (survival)
  • For the typical DPS, the trade-off is between +attack and finding a way to crank +attack up to 11

Recently, theory-crafting has confirmed that there is a specific attribute allocation for talisman in TSW.  In the table below I depict these weights and the resulting value for hitpoint, attack power, and heal power.  These figures use QL10 Blue quality gear from the upper elite dungeons.  Elites have a lower tier of gear with weapons around WP 330 and a higher tier around weapons with WP 347.

Weight Hitpoint DPS Heal
Head 19% 1,945 439 495
Finger 17% 1,740 393 397
Neck 17% 1,740 393 397
Wrist 17% 1,740 393 397
Luck 10% 1,024 231 248
Waist 10% 1,024 231 248
Occult 10% 1,024 231 248
Weapon 347
Max Value 100% 10,237 2,310 2,430

This tells us that a Q10 equipped tank could go full +hitpoint gear resulting in an additional 10,237 HP and only rely on his/her weapon for DPS.  A DPS could go full ATK and pick up 2310 attack to supplement his/her weapon – leaving him/herself with 1970 hitpoints.  A healer could, likewise carry 2430 +heal to augment weapon power and similarly carry only 1970 hitpoints

Show us the Curve!

While stacking to the maximum of your primary factor might seem prudent, in practice it just doesn’t work.  Particularly for tanks and healers, if you don’t slot some second factor talisman, you will find yourself unable to actually fulfill your primary role.  Tanks will find they can’t generate enough agro to keep pace with DPS.  Healers will find the occasional blow from an add leaves them pretty dead, making healing a bit challenging.

It turns out that the trade-off in TSW produces a similar Pareto efficiency.  Assuming you are wearing the best quality gear, given your QL and echelon, your two factor values look something like this for a Tank or DPS.

Now, bear in mind, this is using the upper end blue-quality Q10 talisman.  Just starting elites, your numbers will be lower, but the concept of an efficient frontier remains the same.  For the most part, the trade-offs start to manifest as soon as threat tables become non-trivial to manage.  Bluedot did a very outstanding write-up on threat mechanics in TSW, a good read for those inclined to succeed.

Trade-offs will start sooner for some roles than for others.  Tanks, operating with reasonably efficient builds, will start finding problems holding agro perhaps as early as the Ankh (QL8).  By this point, DPS have exceeded the threat output by more than what you can accomplish with builds alone.

At this point, a tank will discover life gets easier by getting squishier.  Take off one or more +HP talisman and swap over to +ATK.  For me, it’s been my headpiece (19% of my attribute total).  I run roughly 80/20 tank/dps and that works perfectly fine.  I have no pattern of agro loss and our runs proceed smoothly.  This has carried us cleanly through every elite dungeon as well.

Healers, similarly, will find that survival matters come into play early.  As early as Darkness War (Q6), it’s not uncommon to see swarms of adds enter the battlefield (Hi2U Mayan Battle Mage) and these guys usually make a bee-line towards your healer.  Heal throughput, attentive tank-agro and on-the-spot DPS add pickup usually do the deal, but having a few more hitpoints can make life a lot easier for your healer.

For DPS, this often doesn’t manifest until elites.  Prior to that, simply managing threat reducers and staying out of environmental effects do the job.  Additionally, prior to elites, even the occasional environmental effect tends to be non-lethal.  But, with elites, you start finding that pretty much everything hits DPS for 2,000 or more damage.  Just enough to one-shot an all +ATK DPS.  Entering into elites, it’s advisable for DPS to move to between 3k and 4k health.  Conveniently, this happens at 20%-30% +HP gear.

Common Talisman Loadouts

The two most common loads have already been introduced, the 80/20 or 70/30 build.  In this format, the higher value goes to your role specific talisman (+HP for tank, +Heal for healers and +ATK for DPS) with the lower value going to your secondary factor (+ATK for tank, +HP for heal/DPS).

The logic for which to use is different for the roles.  For tanks, you want to carry the least amount of HP necessary to do your job safely.  Unused hitpoints are wasted DPS.  If the encounter is particularly threatening, go 80/20.  If not, 70/30.  If you are on farm mode for a dungeon, you might find yourself going 50/50.

For healers and DPS, the converse is true.  Wear as much primary factor gear as you can get away with.  Dead healers don’t heal and dead DPS don’t DPS.  You need enough HP to survive anticipated damage and no more.  Going into elites, that’s probably 70/30, giving you 30% HP gear to get past learning mistakes (for the whole party).  In an experienced group, and by that I mean a group who is experienced with each other, go for your maximum output. 100/0 for DPS is quite feasible in elite (although not in nightmare).  Healers will probably never hit the 100/0 level, but 80/20 and 90/10 might be possible as runs become smoother.

Turning Percentages into Talisman Choices

80/20 – Toss on a second factor headpiece and you are golden.  Alternately, two minor talisman on your second factor and the remaining gear to your primary factor works.

70/30 – Three minor talisman devoted to your second factor, the rest to your primary factor.

60/40 – A headpiece and two minors towards your second factor, everything else towards your major factor

50/50 – Your three major talisman to your second factor, everything else towards your major factor

Three Factor Talisman Loadouts

Up until now, I have focused on builds which make trade-offs between two factors (HP/ATK for tanks ATK/HP for DPS and HEAL/HP for healers).  Some builds actually need all three factors to play well.  In particular, the classic survival build (primarily soloing) needs a bit of all three factors to play well.  However, AR build healers may find some benefit from having a little of everything too.

In these cases, you need to prioritize.  What is your primary, secondary and tertiary factor.  For a survival build, I would imagine it’s DPS, HP, Heal.  For an AR build, it’s likely Heal, HP, DPS.  I don’t have a lot of data on either of these models.  While I do play survival builds when I outdoor quest, I duo with Aela, so my  +heal needs are pretty low.

I would probably start a three factor model on a 50/40/10 or 50/30/20 model.  You can use the gearing choices in the preceding section to get the talisman allocations to make this happen.  From there, tweak as needed.  Bump in 10% increments as needed, remembering that any increase in one factor has to be offset by a decrease in another.  What are you willing to give up to do more?

Putting Glyphs into Talisman

I plan on spending more time on glyphs in a subsequent essay, but I think it’s worth touching on something here.  I prefer to have my secondary factor gear be glyph-eligible gear.

While you may find dungeon drops for your primary factor where the other attributes are perfect for your role (+Evade or +Defense for me typically), it’s hard to find gear for your second factor with the right stats.  I mean, how many non-glyph 439 ATK head pieces are out there with +Defense?

So, when you are looking for the perfect loot for your second factor, look for gear that can be glyphed.  I can take a 439 ATK head piece and re-glyph it so it has the same ATK but still helps my DEF or EVA (or whatever else I need in my build).

Blue (and better) talisman and glyph kits are worth their weight in gold!

Share in top social networks!

The clothes don’t make the man, but the talisman do!

Talisman Skill Matters

Vertical Progression in the Secret World

This is the first part of a three-part essay dealing with weapons and talisman in the Secret World (part 2 is now available).  TSW is a game with a flatter than normal leveling curve and it is a game for which the relationship between attributes and appearance has been completely decoupled (attributes come from talisman/weapons while appearance comes from clothing).  Further, the talismans themselves come in three basic flavors (health, attack and healing) and at three broad echelon (green, blue and purple).  On top of that, talisman are often customizeable with glyphs from very early on and some additionally have both glyph and signet customization later in the game.

The end result is a system in which a player can really customize his or her gear to fit his or her play style.  It is a gear system which nicely complements the flexibility of the deck system in the game.  However, just as the deck system itself lends itself to unplayable and substantially sub-optimal choices, so to does the gear system leave you plenty of rope to metaphorically hang yourself.  When the game launched, we positioned our My First 60 and Blue Mountain builds as options to help players find synergistic, low-AP builds to play. It is now time to turn our attention to the relationship between gear and builds.

Poor gearing choices can make a build play bad.  A good deck becomes difficult, a sub-optimal deck becomes painful and a bad deck simply results in continuous trips to the nearest rez conduit.  Often, when the gear choices are at fault, the player won’t realize it and will instead focus on the build as the problem.  While sometimes the build is the problem (e.g. you can’t use super-gear to turn a bad build into a functional one), there are a host of gearing problems that we believe underly a number of “this [game/zone/quest] is too hard,” complaints.

I am writing this as a series of three essays.  Essay #1 will address the fallacy of asymetric vertical progression (don’t worry, I’ll explain it in a minute).  Essay #2 will address the talisman distribution conundrum.  Essay #3 will address glyphs, signets and you.

Vertical Progression in TSW

TSW is billed as a “level-less” game, but that is largely marketing distortion.  There are clearly quality levels to gear, there are clearly skill ranks needed to get into said gear, there is clearly visible stat benefit from increasing the quality level of your gear and finally there is clearly a difference in how monsters “con” as you change your skill level.

True enough, the levels as they are in TSW are flatter than those of other games.  A player in mid-game gear (Q6 and Q7) can function in much harder environments, albeit with necessary caution.  However, as anyone who has spent any time listening to general chatter in the Kingsmouth, Savage Coast or Blue Mountain regions can attest, pick-up groups will often try to form requiring gear 2+ QL over the content they wish to face.  It is clear, from this alone that over-gearing certainly makes content a good bit easier.

The game assigns you a vertical level using some form of weighted average of your talisman and weapon levels.  While we do know the weight each individual talisman contributes to the stat pool, we do not know the exact weights for  your vertical level calculation.  None the less, you can visibly see this happening as you level up your skills.  Monsters that used to be red dots become yellow, yellow becomes white, white becomes blue and blue becomes green.

While the level calculation is based on your current skills, the con system appears to work off the assumption that your skills are reasonably symmetrically distributed (e.g. a rank-4 in your weapons and talisman is a level 4 player).  Certainly in the leveling process, you get off balance as one skill gets a point while another doesn’t.  But, as long as your talisman and weapons are reasonably adjacent, your opponent con color is a reasonable expression of the difficulty.

What is Asymmetric Progression?

One of the interesting phenomenon about TSW is that there are about as many opinions of how to play as there are players.  For the most part, most of these opinions work reasonably well and the best advice is to find a working approach that fits your play style.  However, some approaches simply don’t work and others only work for people who play a very particular style and/or play that style very well.  One of those ill-fitting styles is the asymmetric advancement model.

In asymmetric advancement, the player focuses all of their skill points into a single tree (typically a weapon).  They do this until they hit Q10 with the weapon and then go back and flesh out their second weapon and talisman.  The logic of this model is that the one Q10 weapon (often a blue or purple) is “so much better,” that they have an advantage over the content they face, thus making subsequent content easier.

This turns out to be a fallacy.  It’s a playable one, but for many players it will actually make the game harder rather than easier.  There are two reasons why this is fallacious thinking: (1) the actual weapon power contribution of your weapon and (2) the skewing of the con system.  It turns out that the entire approach is based on a mistaken attribution that the benefit comes from the QL when it actually came from the echelon.  I will address each of these in-order.

The relationship between weapon power and combat power

In the figure below, I have mapped out my combat power based on my weapon and three pieces of attack power gear (a head piece, a major talisman and a minor talisman).  For reference, the gear depicted is blue echelon coming from elite end dungeons.  The weapon is a 347 weapon power sword, the head is a 439 attack power head piece, the ring is a 343 attack power major talisman and the belt is a 255 attack power minor talisman.

I have graphed my compbined weapon and attack power on the horizontal axis and my combat power on the vertical axis.  Combat power is the attribute used to determine my damage per swing and it is influenced directly by attack power.  The first data point is for a weapon only.  The second data point includes the weapon and head piece and combat power jumps significantly.  The third data point is weapon, head piece and major talisman with approximately an 80 point jump in combat power.  The final data point is with the three pieces and an additional minor talisman, yielding a much smaller jump.

Players who push to Q10 with a weapon are looking at the left most data point.  The weapon adds more combat power than any single piece of gear.  However, with the exception of the head piece, the other talisman come in triplets.  Bumping up your major talisman one skill point gives you three potential gear pieces which would substantially exceed the value of the weapon’s contribution.

There are benefits from pushing your weapon forward.  You do get a bit more +hit (as a hidden passive) and I would wager that some of the combat power contribution of your weapon comes from the skill ranking itself.  Or, at the very least, weapon power gets more combat power per point to justify the purchase of weapon skill.

But, if a player pushes themselves to Q10 weapon while still sitting at Q4 talisman, they are really leaving a lot of attack power on the table.  They spent 37 skill points getting to Q10 weapon (form Q4) which would have just as easily put them in all Q5/Q6 talisman (and both weapons).  I know it sounds like the Q10/Q4 isn’t that far off (and is potentially better), but the net effect is that the player messes up how mobs con, and does so in a way that harms the player.

Messing up the con system

Recall that mobs difficulty con is a weighted average of your weapon/talisman skills.  It does not appear to be effected by the actual gear, but rather by the skill investment itself.  Using my example above, a Q10/Q4 player is about the same as a balanced Q5.5 player.  I would imagine their cons are relatively similar too (the con system most likely uses the total skill point investment as its basis).

 Our Q10/Q4 player has a weapon kicking out Q10 damage.  Unfortunately, they have the hitpoints, complementary attack power and heal throughput of a Q4 player.  From our chart above, their true damage potential is a good bit lower than  it could be, their health is far below where it should be, and their healing can’t keep up with the damage they will face.  The asymmetric player breaks the con system, and does so in a way unfavorable to themselves.

 Is the trade-off worth it?  Not so much it turns out.  Using data from TSW-Loot, we can construct a figure of weapon attack power progression.  To keep things consistent with my gear, I only used blue quality gear from the normal dungeons in TSW.  On the horizontal axis is the quality level of the gear (Polaris Q3, Hell Rising Q5, Darkness War Q6, Ankh Q8 and Elite Q10, I left out Hell Fallen the Q9 but it should plot nicely along the trend line too).

So, our hypothetical Q10/Q4 player appears to be benefiting, right?  The relative attack power from a Q4 blue to a Q10 blue seems to be about 2.5x greater.  Clearly asymmetry wins?  Well, no.  While the weapon power more than doubles, but combat power is what matters.  Remember, that 347 attack power bought exactly 144 points of combat power.  Even if it turned out that the Q4 weapons provided half the weapon power the Q10, our first figure demonstrates that your combat power really amps up from your talisman.

 The player Q10/Q4 player could have made up a good bit of the ground from their super weapon simply by acquiring a few attack power talisman with a more balanced progression.  By doing so, they would also help to make sure they have the right HP level, the right heal level (if a healer) and just as importantly, the right glyph level (although you can twink glyphs, but that’s going to be in essay 3).  On top of that, a balanced Q5/Q6 player is going to have also made up some of the weapon power gap simply from the Q5+ weapon they would have on over the Q4 the rest of their gear is compared to.

 It’s really the echelon that matters

Our hypothetical Q10/Q4 player is seeking an edge, good for him, everyone is.  But, the player is looking in all the wrong places.  Rather than heading out and picking up a single Q10 and atrophying all other skills in the process.  Any advantage from this path is minor and offset by the disadvantages to the con system.  Unfortunately, the Q10/Q4 player is mistakenly attributing the benefit of their Q10 weapon to the Q-level when it mostly is coming from the blue (or purple) echelon of the gear.

Earlier, I noted that my weapon is a 347 blue weapon power item from an elite dungeon.  For reference, a green Q10 weapon is 258 attack power.   Using the chart above, a green Q10 item is roughly equal to the weapons out of Darkness War, a Q6 dungeon!  The best bang for the buck comes from doing the dungeons at the right QL.

 Run Polaris as a Q2/Q3 wearing greens.  The gear will be an upgrade and will let you easily handle Q5 Hell Rising.  Hell Rising gear sets you up nicely for Darkness War.  As we saw in the proceeding paragraph, a good set of Darkness War gear is equivalent to green Q10 gear.

 On top of that, you are picking up blue quality talisman.  And, to make things even better, you can equip all those items for much less skill investment than the Q10/Q4 player.

 You wind up breaking the conning system in your favor, if you do the dungeons.

Closing thoughts

There are a number of perspectives on how to spend your skill points.  I see some advocate “reasonably balanced, but let your weapon stay about +2 above you.”  That’s only a bit asymmetric and it is possible that the benefits of this offset the disadvantages.  Aela will tell you to stay reasonably balanced, but buy up your weapon first.  That will keep your gear in line with your weapon always giving you a slight edge.  This does seem to make some sense since the weapon contributes more combat power than any other single piece of gear.

 I advocate remaining balanced and being pragmatic with your point expenditures.  If you have two items that are one skill level out of reach (a weapon and a head piece) and you can only invest skill points to get one, go with the weapon.  But, if you are sitting on a pair of majors or three minors, invest in the talisman.  Two majors is better than one weapon upgrade.  Three minors should be too.

 In the end, all of the approaches will work.  They are all reasonably balanced and avoid the problems of extreme asymmetry.

Share in top social networks!