Guild Wars 2

First, Ill be the first to admit, my understanding of how your server works is completely limited. However, Im just asking from a logic standpoint, it may be completely stupid.

Recently Jon Peters said:
“Summoned creature AI is a different can of worms that we aren’t opening for the same reason. Pets that delay F2 use isn’t some wait script we put into their skills it has to do with core AI behavior shared by all pets and creatures and how they decide tasks. Rewriting that has the risks of breaking millions of unknown things so we have up until now band aided the solution. It is something that needs addressing but won’t be addressed until we can kitten how and when we will test it”

From which I ultimately got the reason F2 is unreliable is because its connected to how all AI behaviors function. Understandable, you don’t want to mess everything up.

Now, where my question comes in:
Ranger Shouts: I cast them, the pet does them (after a cast time, yes)

So, the priority for us, overrides what they are doing? Would it be possible for the pets F2 to be actually cast by rangers? So its not longer on the pet, but is seen as more of a shout? Now, the downfall of which would be no longer being able to simultaneously channel and use pets skills, and when stunned would not be able to use pets F2(Though, you may be able to get around that). But just form a purely, would it work standpoint? Would it work?

If it would, would you ever consider using such a system?

Thank you for all that you do and for any answer.

actually, they could make it that way, however it would have to trigger an animation in both the player and pet model.

So say you want to cast fern hound healing.

The casting is done player side, and a animation triggers in the player to cry out a shout-phrase like “heal me” or something, at the same time, it triggers a “howling wolf” like animation in the pet.
The most reliable way of implementing this system is by setting it to player side only, that would reduce the amount of server resources needed to make it work.

The current system, in which the pet is somewhat controlled by the server, and any action done by the player has the same latency as when you interact with any kind of NPC (like when talking to NPC’s, you notice there is a slight delay between pressing next/continue and the action happening. That is caused by ping-delay.)

So it would work, it would just make the delay longer? So 1 second cast from the ranger? And 1 second cast from the pet?

But overall more reliable?

Also, why would the ranger need to cast as well? Would it be possible for simultaneous casting?

Hedgehog in the fog.1053:

So it would work, it would just make the delay longer? So 1 second cast from the ranger? And 1 second cast from the pet?

But overall more reliable?

Also, why would the ranger need to cast as well? Would it be possible for simultaneous casting?

no, you misunderstood.

Current system:
Server controls pet, when we press F2, the server gets a command which is added into pet “action queue”.
However, the priority of the actions you queue are “broken” so that movement actions, or attack actions overwrite/start simultaneously with the F2 action. However, F2 effect (healing/boon/condition) are applied when pet F2 animation ends.
So when another animation overwrites the pets F2 animation, the F2 does not work.

Suggested system:
The server controls the pet, however the pet F2 effect (Healing/condition/boon) are applied by the PLAYER. However, when you press F2, instead of pet executing animation (hopefully) and THEN applying effect (healing/boon/condition) the PLAYER instantly applies said effects, and the pet just does a complimentary animation so it seems like it did something in the first place!

By doing it this way, it does not matter if pet moves or not, the effect will happen, regardless of the animation succeeding or not. In the case of condition applying attacks, such as Blinding Slash, the effect must be added AFTER the attack animation, as is currently done.

EDIT: for this system to work seamlessly, there must be absolutely zero aftercast on the suggested pet F2 system

The main issue being that if you make the cast/animation client-side, issues like lag and server extrapolation throw a HUGE monkey wrench into the mix. Then you’d have instances in Player-versus-player content the pet would effectively hit the target potentially before the target gets the visual queue to dodge. Seeing as this game features dodges and whatnot, this is a pretty serious problem which would make people get hit out of nowhere, the pet being far away visually, and then the pet would appear way after striking the target due to server communications.

Honestly, without instantaneous computation/decision making AI-wise (impossible) and instantaneous communication between client→server→client→server→client, asking for anything but the current system will have massive problems.

Reality is the ranger needs a ground-up rework. ANet made the ranger from a mix of two failed classes, warden and beastmaster, and combined Archer elements with it because it fit thematically and archer could not be made into a class by itself due to its restrictiveness. The first two were failed because they also didn’t have enough variability and depth from ANet’s perspective to warrant individual classes. Thus, all three got combined.

Ranger simply needs tweaking to allow people to specialize these ways. Solo-archer, beastmaster, and woodsman/warden/nature magician are all cool archetypes that players want to play. Issue is, the traits and styles are two dispersed in the ranger tree to really allow anyone to play the way they want to.

The best way to fix the broken pet issues would be to reward players more for using BeastMastery-style play. Want pet effectiveness to go up? Like spirits, let the player run multiple pets at once for deep investment in Beast Mastery at a cost of weapon damage and grant the pets’ attacks AOE’s and player-stat scaling as to prevent whiffing and the likes. Two pets = more chances to hit, and with proper development, they’d have two separate movement paths to also help counteract possible whiffing.

For the warden builds, give them more utility and heals with the nature theme, like rooting nearby targets upon reaching low health with roots coming from the grounds, obviously playing more into healing springs, etc.

And for the Archer builds, run more solo-character DPS supported by the utility from both BM and Warden styles, to say have a pet (if brought out in combat) help soak aggro while utilizing natural roots to immobilize the enemy and strike from a distance (of course, applying a penalty to damage if pet is out like there currently is and allow the pet to compensate for it).

Those types of changes are what the ranger REALLY needs. AI will never improve until technology effectively breaks the laws of physics (never happening). Pets work in other games due to different dynamics which allow them to. Visual queues for dodging are essential to maintain, so the delay will always exist.

But, of course, ANet will likely keep using the busted pet mechanics because they’re too proud of their hodge-podge of failed classes actually being a class than solidifying the game and becoming a better studio at the expense of eating their words for the benefit of the players.

I had a build back in guild wars 1 where I carried a staff and my whole skill bar was pet attacks. Why did A net not design the ranger so a strictly beast master build is possible in this game???

danielcole.3418:

I had a build back in guild wars 1 where I carried a staff and my whole skill bar was pet attacks. Why did A net not design the ranger so a strictly beast master build is possible in this game???

It wouldn’t be much use in pvp when you can’t bodyblock with collision. The enemy would walk right through the pet and kill you (the same way BM died in GW1 because you had to sacrifice ranger survival skills for pet attacks). We already have problems with keeping pets outside of enemy killzones in PvE without crippling dps so I’m not sure how well it would have worked.

The apparently had a BM in the works but scrapped it when they decided on the version of the game that we have now.

Op sound like the skill would interrupt the ranger action. If I’m ressing or using a channeling skill use the pet skill would interrupt my current action. One of the benefits of being a ranger is you can have your pet perform one action and you do another. This sulotion sound like it will take away from that.

“The main issue being that if you make the cast/animation client-side, issues like lag and server extrapolation throw a HUGE monkey wrench into the mix. Then you’d have instances in Player-versus-player content the pet would effectively hit the target potentially before the target gets the visual queue to dodge. Seeing as this game features dodges and whatnot, this is a pretty serious problem which would make people get hit out of nowhere, the pet being far away visually, and then the pet would appear way after striking the target due to server communications.”

Possibly. However, yesterday I fought a warrior and as he said fear me, I was already feared. There is already a lack of visual Qs in some regards.

@Prysin, is there anything stopping them from doing this type of system, besides the lack of pet based visual Qs? Which, as you said, if the ranger had the command, there would still be a Q.

Also, since you seem to know a bit. Is there a way to change the priority system?

@Seraphin Storm. Personally, I would take the ability to have a reliable utility over simultaneous casting of F2. Im sure some would disagree, but sometimes you really need that wolfs fear to work, and it…doesn’t.

danielcole.3418:

I had a build back in guild wars 1 where I carried a staff and my whole skill bar was pet attacks. Why did A net not design the ranger so a strictly beast master build is possible in this game???

Why didn’t they build the game so that I can use my old marksmanship/survival build without being shackled to a bug-ridden NPC? They basically alienated everyone at both ends of the spectrum.

Substance E.4852:

danielcole.3418:

I had a build back in guild wars 1 where I carried a staff and my whole skill bar was pet attacks. Why did A net not design the ranger so a strictly beast master build is possible in this game???

It wouldn’t be much use in pvp when you can’t bodyblock with collision. The enemy would walk right through the pet and kill you (the same way BM died in GW1 because you had to sacrifice ranger survival skills for pet attacks). We already have problems with keeping pets outside of enemy killzones in PvE without crippling dps so I’m not sure how well it would have worked.

The apparently had a BM in the works but scrapped it when they decided on the version of the game that we have now.

The problem is, against some enemies it’s almost impossible to get a single F2 skill off the entire fight due to the delay. Even worse, if an enemy interrupts your pet’s animation, it goes into full cooldown. That’s a very frustrating outcome when the skill should have been cast a good 5 seconds before the interruption occurred.

In looking at the F2 skills across the pets Anet seems very short-sighted for on: 1 designing the lag in response that nothing can fix. 2 Giving pet long activation animations along side the lack of responsiveness 3 Assigning the cool down that they did for the most of the effects.

Other MMOs seem to have much better systems in place for handling these types of skill queue situations. When I commanded my pets to use a skill in say, Perfect World, they used that skill. Less complex combat system and mechanics yes, but it functioned properly.

On that note, the fern hound’s F2 is still bugged. If you attempt to cast it immediately upon swapping to the hound, it will occasionally do nothing and go on full cooldown.

Also, toxic knights can dodge roll. Why can’t our pets do this?

I’ve said this many times and I’ll do it again. Please, please make actual skill effect more important than animation tied to it for pets. That one thing would simply fix all pets pathing and dealing damage to moving targets. Long pet ttack animations hurt gameplay and it’s much more important than attack animations. Like now, pet stops after launching his AA for a split second to finish animation thus he’s often just OT MOVING AND HITTING AT ALL. Sorry for caps, but I felt like it should be highlighted. Priorize pet actions over finishing animations, don’t stop the pet after attack, make him even slide on the ground towards enemy, I don’t realpy care, but these pet lags are ridiculous and you’d either have to make all pet attacks small leaps, castable on move(best solution), make pets like 50% faster.

Quarktastic.1027:

On that note, the fern hound’s F2 is still bugged. If you attempt to cast it immediately upon swapping to the hound, it will occasionally do nothing and go on full cooldown.

My Jungle Stalker does this too.

Quarktastic.1027:

Other MMOs seem to have much better systems in place for handling these types of skill queue situations. When I commanded my pets to use a skill in say, Perfect World, they used that skill. Less complex combat system and mechanics yes, but it functioned properly.

On that note, the fern hound’s F2 is still bugged. If you attempt to cast it immediately upon swapping to the hound, it will occasionally do nothing and go on full cooldown.

Also, toxic knights can dodge roll. Why can’t our pets do this?

Thing about point-and-click games like PWI/WoW is that the pet animation is nothing more than an animation. Often times the actual skill takes place and then the animation is performed client-side as it’s sending the message to the servers to make another player notice he’s being hit by it. Due to the fact that these games don’t have real interactions and whatnot between the servers, the animation appears to be more successful and smooth. If you’ve ever seen gameplay from two sides at once, often times the positioning of players as represented in each client is slightly different for both with a delay period between both of them. Again, since there’s no hitbox/dodge scanning, the ability triggers as long as at one point the target is within range and then simply scores a guaranteed hit regardless of where the target moved to.

It boils down to technological issues at the end of the day.

For those saying to make the pets automatically land their skills/land the effect before the animation, you’re crazy. Having a guaranteed-hit blind from birds and other CC abilities is totally absurd due to the fact there’d be no visual queue. Sound isn’t so much an issue because shouts don’t have a real hitbox and would give an inherent disadvantage to those who may not have hearing or have hearing problems. Visual is kind of necessary, for one must be able to see in order to play the game and use a computer to begin with.

DeceiverX.8361:

Quarktastic.1027:

Other MMOs seem to have much better systems in place for handling these types of skill queue situations. When I commanded my pets to use a skill in say, Perfect World, they used that skill. Less complex combat system and mechanics yes, but it functioned properly.

On that note, the fern hound’s F2 is still bugged. If you attempt to cast it immediately upon swapping to the hound, it will occasionally do nothing and go on full cooldown.

Also, toxic knights can dodge roll. Why can’t our pets do this?

Thing about point-and-click games like PWI/WoW is that the pet animation is nothing more than an animation. Often times the actual skill takes place and then the animation is performed client-side as it’s sending the message to the servers to make another player notice he’s being hit by it. Due to the fact that these games don’t have real interactions and whatnot between the servers, the animation appears to be more successful and smooth. If you’ve ever seen gameplay from two sides at once, often times the positioning of players as represented in each client is slightly different for both with a delay period between both of them. Again, since there’s no hitbox/dodge scanning, the ability triggers as long as at one point the target is within range and then simply scores a guaranteed hit regardless of where the target moved to.

It boils down to technological issues at the end of the day.

For those saying to make the pets automatically land their skills/land the effect before the animation, you’re crazy. Having a guaranteed-hit blind from birds and other CC abilities is totally absurd due to the fact there’d be no visual queue. Sound isn’t so much an issue because shouts don’t have a real hitbox and would give an inherent disadvantage to those who may not have hearing or have hearing problems. Visual is kind of necessary, for one must be able to see in order to play the game and use a computer to begin with.

There’s a far cry from making things instantaneous and giving people multiple seconds, in some cases, to completely avoid pet skills by simply side stepping. A person really would have to be blind to be hit by the frost, fire or reef drake skills.

Not really sure if you’re agreeing or disagreeing with me.

Seeing as the drake skills are far from just instant casting (breath lasts for a duration even when cast), I don’t see why making them faster to cast would really help tbh. Breaths are cake to dodge seeing as there’s a huge visual queue beforehand. Fixing the delay on F2 wouldn’t really help these pets except for maybe they’d get a fraction of a second on someone unprepared.

It’d improve those particular pets, but those pets really only function well to begin with when their targets are locked down. Even fighting drakes is pitifully easy due to how slow and clunky their skills are and how easy it is to evade their hits. I think that breath is more of a drake problem than a time/delay problem. ANet has human players fighting drakes as some of the first mobs they encounter due to their predictability. Pet breath should probably just behave differently so that the pet rotates with its enemy despite being rooted and performs the action with less wind-up but with a shorter, more concentrated duration as to simply make the actual skill more difficult to dodge against in PvP scenarios.

Hedgehog in the fog.1053:

“The main issue being that if you make the cast/animation client-side, issues like lag and server extrapolation throw a HUGE monkey wrench into the mix. Then you’d have instances in Player-versus-player content the pet would effectively hit the target potentially before the target gets the visual queue to dodge. Seeing as this game features dodges and whatnot, this is a pretty serious problem which would make people get hit out of nowhere, the pet being far away visually, and then the pet would appear way after striking the target due to server communications.”

Possibly. However, yesterday I fought a warrior and as he said fear me, I was already feared. There is already a lack of visual Qs in some regards.

@Prysin, is there anything stopping them from doing this type of system, besides the lack of pet based visual Qs? Which, as you said, if the ranger had the command, there would still be a Q.

Also, since you seem to know a bit. Is there a way to change the priority system?

@Seraphin Storm. Personally, I would take the ability to have a reliable utility over simultaneous casting of F2. Im sure some would disagree, but sometimes you really need that wolfs fear to work, and it…doesn’t.

yes they could implement it through a buffer system. However it would be a horrible headache to implement it.

In theory, they could buffer any AI action and delay it by 1 second. Only downside is that whenever a ranger/necro/mesmer wanna do something, their AI friends will lag 1 second behind. The reason its not smart reducing the delay is because if its too short, people with pinglag will still suffer while the rest doesnt. On the contrary, too long of a buffer, and every AI’s DPS goes down the drain.

Rewriting the priority queue is about a big of a task as redesigning the entire AI system. Most likely, they would have to add more priority groups/edit a lot of values to ensure that the skills would fire off in the right sequence at the right time.

most of the issues with pet comes down to overly slow and highly telegraphed animations. Simply put, these attacks are not seamless and does not flow very well, or at all, with the rest of the attacks.
The fact that you have between 1 and 4 second warning before the animation start is ludicrous in the first place.

The main reason for me to suggest instant EFFECT application, is because in the majority of cases, the ranger is in need of the EFFECT, not the actual DPS. This is why i say that the damage must be bound by animation while effect is instantaneous.

When every other class can through one way or another inflict instant blind/fear/stun/daze/knockdown/immobilize/chill, then why should we not be able to do so too?
Look at other classes signets. Sure its not worth spending the signet, but they CAN. It’s the principle of having an effect, on demand, with an extremely short animation/instantaneous effect being applied. The main reason to use the Raven F2 is because A: it blinds, B: it does a good deal of damage. Introducing my system would add the blind right away, and the damage (which in the majority of cases is what you need to evade) is bound by the attack animation end.
Some pets, like Hawk, Eagle, Snow Leo, Lynx and others that apply condition upon hitting, should not be altered to fit into the system, as it would be incredibly OP to have 25s bleed on-demand at any time, from just having a pet there. Such damage should be possible to block/dodge/evade.

To simplify, the change in system should prioritize non-DPS skills, and disregard DPS skills. This way, your direct DPS/pet DPS does not change, however the pet utility and ultimately, reliability would

Have you put this suggestion in Suggestions forum?

How would it work for DPS based F2s? The same as it does now? Ive also noticed my Red Moas fury is applied before the animation begins. But, the jungle stalker needs to complete the animation.

So, your saying it would be difficult to do this?

I just find it funny that they can’t get it to work properly when in WoW pets were much better to use and responded instantly to commands.

When I told my Loque’nahak to attack, it would actually rush forward, stop briefly (crouch down) to perform a special attack when in range and keep on the target like glue; always trying to attack from the back. Oh, and they had an innate resistance to AoE, which helped a bit because their AI was stupid when it came to AoE effects (just like in Guild Wars 2)

In Guild Wars it’s like my pets are constantly stoned and have no clue what they are supposed to be doing and die almost instantly from random AoE.

So I think it would be better to give them an innate 75% resistance to AoE (which could then have a trait in BM to boost it to 90%?) and allow the F2 effect to originate from the player. The pet could then just have a short animation to give the illusion that it did something. Oh, and pets should have a small AoE Cleave on their attacks so they can actually hit their target.

EDIT: Oh, and Active Signets should affect both Ranger and Pet by default. It’s pretty ridiculous that we have utility skills that only function for our pet unless we actively spend 30 trait points.

Just my 2c

Aveneo.2068:

I just find it funny that they can’t get it to work properly when in WoW pets were much better to use and responded instantly to commands.

When I told my Loque’nahak to attack, it would actually rush forward, stop briefly (crouch down) to perform a special attack when in range and keep on the target like glue; always trying to attack from the back. Oh, and they had an innate resistance to AoE, which helped a bit because their AI was stupid when it came to AoE effects (just like in Guild Wars 2)

In Guild Wars it’s like my pets are constantly stoned and have no clue what they are supposed to be doing and die almost instantly from random AoE.

So I think it would be better to give them an innate 75% resistance to AoE (which could then have a trait in BM to boost it to 90%?) and allow the F2 effect to originate from the player. The pet could then just have a short animation to give the illusion that it did something. Oh, and pets should have a small AoE Cleave on their attacks so they can actually hit their target.

EDIT: Oh, and Active Signets should affect both Ranger and Pet by default. It’s pretty ridiculous that we have utility skills that only function for our pet unless we actively spend 30 trait points.

Just my 2c

Indeed. Would much rather have it affect both users and then allow for the 30 power GM trait to allow for permastow pets with a damage boon to compensate so that pure archer/DPS rangers can do without the mechanic if they so please to.