Pets and minions aren’t well-regarded by the player base. Rolling dice has more consistency than the AI that drives them. As a result, minion build effectiveness is either nonexistent due to glitches, or blatantly overpowered as a player does functionally nothing while reaping ridiculous reward*.
*This includes turret engineers in sPvP, with the note that turret AI is actually really good. Engie attacks target, turrets attack target if in range. So the random aspect doesn’t really exist.
Justifiably, most players avoid pets, minions, and anything else that relies on AI to be effective. Except, rangers can’t avoid their pets. Sticking one on passive is cutting out a lot of a ranger’s DPS out of the equation. And permanently stowing one in favor of a damage boost doesn’t exist.
I’m going to delve into the exact why’s for pets being so bad, and yet so unavoidable. Then I’ll look at a short list of solutions to make a pet a ranger’s best friend, rather than their worst burden.
The first and most obvious way pets are bad is how braindead the AI acts when confronted with obstacles. And by obstacles, I mean slight changes in terrain, enemies that don’t stand still, and being told to use an active skill mid-attack chain. Expect a pet BSOD when cliffs (or large bumps), teleporting enemies, or being interrupted occur.
Sad and unavoidable truth is that pets act just like any other AI in the game: badly. Players don’t mind as quickly when it’s the enemies, but when it’s a major source of their damage and utility as a class, stupidity is not an option.
To make matters worse, pets have just as much self-preservation as necromancer minions, and a brutal cooldown if allowed to die. Playing against a boss with metric tons of AoE circles (i.e., every single boss in recent memory)? Expect your pet to act like Iron Man and collapse like a straw man. Again, the same AI that drives enemies drives pets.
To create a unified list of the gaffs the AI creates in pets, here I attempt a (likely incomplete) list:
- Active skills are still delayed in their activation, even after an update to cause them to interrupt current action in favor of the active skill.
- Cliffs and abrupt changes in terrain psych a pet out. A player just drops down, a pet is bound by the same “cliff ignorance” as enemies…and paths the same way.
- Pets do not track moving enemies well. Rather than attacking on the move with the consistency of a player, they “miss” attacks they are in range for.
- Pets do not change targets well. Compared to the amazing responsiveness of turrets (instant target switch on next attack), pets feel the need to finish attacking one thing rather than jumping to their master’s desires.
- Pets do not heed AoE circles. This kills their viability (literally) in most high-level content as it revolves around positioning and avoiding danger zones. This is only going to get worse in the positioning-focused Heart of Thorns*.
*Based on the wyvern fight and the designers’ comments on it.
(Anything I did miss, or got wrong from my relatively limited time on ranger, please let me know so I can update)
Simplified for Your Inconvenience
There is very, very little direct control a player has over their pet, either in combat or out. Either it’s doing nothing (passive) or it’s attempting to attack your current target (guard). Its active skill is a button push away, but as already covered it’s poorly suited for clutch skill usage due to delay.
Add in the ability to attack current target (allowing the ranger to switch targets until the current one dies), a simple Return to Me, and swapping the pet out for another one, and that’s it.
Attacks are automatic, pathing is automatic, positioning is automatic, the decision to keep attacking when the ranger switches target (short of F-key usage) is automatic. For simplicity’s sake, there is no way to control the minutia.
Learning curve-wise, this is good. Having a pet you have to control in addition to yourself beyond a few simple commands would be an absolute nightmare to learn. Even engineers and elementalists would tremble at the thought of having to handle a complementary, but separate set of skills.
But it also means that pets (and rangers) live and die on the strength of that simplicity-driven automation. Which as already stated, is terrible.
“Brown Bear, I Choose You!”
Rangers have the easiest stereotype to remember in the game: bearbow. Someone who sits at the very edge of the fight using Point Blank Shot at the worst moments, while letting a hearty bear tank all the damage. Because in easy open world PvE, that’s all that need be done pretty much everywhere.
Dig a bit further, and there’s a deeper problem at play: there’s dozens of pets, and only a few worth bringing out to pounce on the enemy. There’s a niche for many families, but due to the limited situations present in the game (and limited unique traits of pets), many are left without a point. Some are charmed just for “100% completion”, and never used because they’re impractical.
Stuck Between a Warrior and a Thief
This plethora of problems begs the question “why not just let us get rid of pets if we don’t want that option? Why not just buff our damage and call it good?” Because without a pet, rangers are either a heartier thief or a more finesse-based warrior.
On the first count, rangers don’t drop like flies considering their limited stealth and evasion compared to a thief. They can take a few hits, get up in an enemy’s face and mess up their day pretty good if they want to. And they’re just as susceptible to hard CC…and hard-pressed to beat a thief at avoidance.
On the second, rangers can’t wade into the midst of a big fight and tell people who’s boss. But from the fringes of the fight, firing away with bows or dancing in and out with melee weapons, rangers are excellent. That presumes that they’re allowed to do the pew-pew. They don’t have the same ability to shrug off or simply outlast heavy focus.
Without a pet, there is no ranger identity as the game is built. More than just being the profession-specific mechanic, a pet prevents the ranger from falling into a “similar but worse” version of thieves and warriors. It’s both a blessing and a curse, but as the game is presently structured, it’s unavoidable.
Fix the AI
With rumor of game AI being improved (backed up by bringing Dave Mark of Intrinsic Algorithm on as a consultant), there’s hope of pet AI bringing brought along for the ride.
But pets are more complex than simply enemies. To be a proper companion, they need to react at their master’s behest, as well as cherish their role as living beings. So simply adding potential better AI from enemies to pets is not enough, all the problems above need to be addressed as well.
Allow More Complex Control
Maybe the AI issues aren’t directly surmountable, or take even more work. In that case, allow players to take more control of pets if they choose to. For the most part, this would deal with movement and positioning to avoid the problem of playing two disconnected skill bars:
- Take a page out of Guild Wars and implement “rally flags”. These set a pet to a specific location (or maybe the area around it to keep from adding to the targeting problems). Repurposing the tech used for banners could possibly implement this.
- Create a “close guard” behavior. Pets stay within melee range of the owner like the “passive” behavior, but attack if there are targets within that range. This allows pet owners to force good positioning by their own positioning. It won’t help for closely-spaced AoE circles, but it’s better than a pet being oblivious to how hot lava really is.
Rebalance Pets to Have More Options
Some of the pets fit good and useful niches, but most do not. Similar to how skill and trait rebalancing is needed to remove uninteresting or near-worthless options, pets need some rebalancing so rangers aren’t stuck with a half-dozen good options and thirty or so animals best suited for roleplaying depth.
I do not have enough experience with ranger to make any specific suggestions in this vein, though.
Consider Rooting Ranger Identity in Nature, rather than Pets
The toughest way to “fix pets” would be to change the overall unique identity of rangers away from pets and toward nature itself. All of the ranger’s skills (especially signets and spirits) are nature-themed, just like Guild Wars (where pets were optional, not mandated).
So why not go whole hog on rangers as nature-minded wielders of martial weapons? This might cleave too close to the Druid concept of the upcoming specialization, but taking the focus off of “You have a pet! That doesn’t work that great!” reduces the outright requirement thematically that a pet be out and attacking. This gives the leverage to make a pet and the additional utility it brings (through tanking, the active skill, movement, etc.) an option, rather than a necessity.
That said, making pets an option (and buffing rangers’ abilities in some ways to compensate for the missing DPS provided by pets) would be difficult to balance.
Thus, I think a pet-less ranger is best suited to a new specialization rather than something available to all rangers at the expense of nothing. Recall that specializations replace existing mechanics with new ones. So removing the pet entirely in favor of a “spirit shaman” with passive bonuses is perfect for that approach.*
*And with the new skills a specialization would add, it would address/negate the problem of the skill set not being different enough from warriors and thieves.
When it comes down to it, rangers are stuck with their pets, at least until a specialization comes along to remove them in favor of something else. This wouldn’t be that bad, if their AI made them useful companions, or failing that players had the control to keep them acting less dumb. Add in that most pets are strictly inferior options, and the one thing rangers can’t do without is the one thing dragging them down.
The AI needs to be addressed, but barring that more positioning control should be added. And having 45 pets and only a handful of viable options needs to be addressed. Rangers need companions, not chains.