Which Archer class to play? Some musings from the Drunken Ranger

[Cadebrennus]

Note: This is a repost from my original post on the Phoenix forums. Keep in mind that information may be updated, changed, or whatever on the server with a delay in keeping up the information here. With that in mind, read on! Also, feel free to comment and add to the discussion. Slainte!
In response to a player's question on Discord. I have reposted here for the community's review, input, and more than likely hate mail from my stalkers.

Scouts are not only the most group dependent, they are also strongest in a group due to the usefulness of the Shield (such as being able to block for another group member.)
Rangers, while fun, are the one-trick-ponies of the Archer classes. They have ranged and melee, that's it. Hunter's 2handers equal a Ranger's dual wielding capabilities for damage, but dual wielding has bonuses (such as extra weapon procs) and drawbacks (such as triggering more enemy armor procs.) The Hunter Pet makes the Hunter the master of controlling a fight at range, provided he can get his slow-ass pet to the enemy Caster or Archer in time. In addition the Pet puts out more damage than the Ranger's damage add even added to two weapons.
The range difference is minimal between the Archers. It's 2200 for Scouts, 2100 for Rangers, and 2000 for Hunters. Hunters suffer from having the fastest bows so they will not have the big critical shot numbers like Scouts and especially Rangers (who have a bow that is 0.2 slower than the Scout bows and add the Damage Add to the Critshot) but do not have to swap bows to hit cap firing speed when Rapid Firing which is important to keep enemy Casters and Archers locked down. Just think of Rapid Fire as a crew served weapon, rather than a precision damage weapon.
What the class choice really comes down to is what role do you want to play as an Archer. Pure sniper? Go Scout (a tiny bit more distance) or Ranger (a bit more damage.) Melee-heavy? Ranger or Hunter. Defensive and an incredible groupmate? Go Scout. Just keep in mind that the damage from any of the Archer classes either in Melee or in Archery is subpar. At the moment a Nightshade can put out equal or greater ranged damage-per-second with an unspecced skill line than an Archer can when comparing casting speed at 1.5 seconds vs an Archer Rapid Firing at 1.5 seconds.


Thinal's thoughts:

The difference between the Archer classes - Phoenix

"

Cadebrennus wrote: ↑
Mon 14 Jan 2019 5:48 AM
What the class choice really comes down to is what role do you want to play as an Archer. Pure sniper? Go Scout (a tiny bit more distance) or Ranger (a bit more damage.) Melee-heavy? Ranger or Hunter. Defensive and an incredible groupmate? Go Scout.

I agree that the scout is the best at defensive and "group mate," presuming Albion isn't already glutted with shield-bearing armsmen, paladins, or reavers who would otherwise do it better. In RvR, the scout's portable slam and readily-available stealth groups make it an exceptional choice.

However, I'd suggest that the Hunter is being significantly underrated here in all other respects. Each gets a specialty: scouts get slam, rangers get damage add, and hunters get pets. Well, the top dog is 32 beastcraft, but slam is a solid 42 shield, and rangers can benefit from pathfinding as high as 50 AND have an extra spec line necessary for melee combat on top of that. Even if one can demonstrate that the right combination of weapon, CD, and PF would top a hunter + dog, the math is insanely complicated. I did extensive testing and was only able to come out with a solid comparison of PF versus CD for a very strictly-designed melee-only ranger. I have no freaking clue how PF compares directly to any weapon/CD ratio, with or without potions / charges / conc buffs.

Therefore, making a high-archery hunter compromises almost nothing in melee capability. Shield slam leaves a scout with medium spec on a single 1H weapon. Few rangers manage to find the points to get archery past 35, and many don't come that far. As penetrating arrow is so silly since it does diddly shit for self-bubbles, that leaves little incentive indeed for a ranger to spec anything in between 12 (free) archery and 35 for the first, exhausting rapid fire.

Hunters have the worst crit shot? I did extensive play and testing in i50 on hunters, rangers, nightshades, and shadowblades, and I could count on one hand the number of times I managed to get off a crit shot at all in RvR. Most targets are running, especially if alone or in a small enough group to where you think you could get lucky. Usually it was tough enough to get off even a normal bowshot. There are no advance stealth skills here, so you're never going to be able to pop a true sight and see that solo stealther at enough range to crit shot him. If bows are even out, it's almost always going to be against visibles, and my hunters spent considerably more time in rapid-fire mode than in critshot mode.

I like rangers. A lot. I still play one. If this game were purely PvE, I wouldn't consider any other archer. But I'm cross-realming and slowly learning the Midgard PvE map, because my hunters tested so well in RvR. For that matter, I should probably get over my hatred of Albion, because seriously... 9 seconds of uninterrupted det-ignoring mutilation on a stealther? What the hell is wrong with me?"

My thoughts again:

The difference between the Archer classes - Phoenix

[NOTE: THIS IS ABOUT PHOENIX AND NOT ATLAS, HENCE THE TEXT ABOUT THE DW DEFENSE PENETRATION NERF] Here (I don't remember the same on live) dual wielding had its reduction of block evade nerfed from 50% to 25%. That definitely made a big difference when attacking anything with Evade or Shield in beta. I tested a Sword Hunter in PvP with a similar spec to my Ranger and he SHREDDED targets faster in melee than my Ranger. As far as I know, 2hand still reduces Parry effectiveness. Hunter 2hand = Ranger CD damage. No shit. Dummy and player tested. The pet acts like an additional offhand hit on top of the already impressive 2hand damage that hits harder than an unstyled offhand, but I didn't test the frequency of pet hits. I think it's slower than CD/DW offhand but it attacks 100% of the time as opposed to 69% or lower offhand swing so it probably evens out. The only downside I can see to the 2hander is a miss that is a bigger detriment to the Hunter than a miss is to a Ranger, and has 1 proc to the Ranger's 2 procs. Upside is that for the same amount of damage the Hunter is hitting defensive procs a lot less frequently than a dual wielding Ranger or Assassin.

Thinal's thoughts again (same page) :
"

Cadebrennus wrote: ↑
Fri 25 Jan 2019 8:42 AM
Hunter 2hand = Ranger CD damage. No shit. Dummy and player tested.

I can back that up. I did extensive testing with shadowblade specs varying axe, left axe, and critical strike and comparing damage of equal double-wields versus 2H. While I wasn't able to test every possible configuration, I never found a case where the double-wield topped the best possible 2H damage I could find. It's not exactly the same thing, as I wasn't using any double-wield haste effect (two EQUAL weapons) and I was using Garrote with a higher growth rate than any of the anytimers archers would have available, but the hunter is also going to have a higher weapon spec if only due to it being a single line versus the Ranger's two spec lines."




So there you have it! As I've said earlier, feel free to contribute thoughts and feedback, as well as more information for players who want to play some sort of Archer but aren't sure which one to play.

 

[Cadebrennus]

Some more musings, mostly in response to some overblown claims about Rangers and cries that Scouts are just plain underperforming in melee. First of all, if you only spec 21 Slash, your damage is going to suck. Period. Most Rangers spec 39 Blades, and I personally never spec anything less than 50 Blades. Even 39 Blades is a significant damage and weaponskill increase over 21 Slash. There's also a lot of ruckus about the fact that Rangers DW makes them melee gods. It doesn't. First of all, Hunters are on a higher damage table, and 2h weapon speeds plus inherent 2h damage bonus puts that damage over DW/CD, not even counting the higher damage table of Hunters. Good. Now we've got that out of the way. Moving on.

Scouts typically spec 42 Shield. Especially on a server where there's a flat 60 second stun immunity regardless of the length of the stun you use, it would be insanity to spec otherwise. (I personally like to spec around 35 Shield on my Merc on Live and Phoenix, but those servers are dead or deserve to be dead at this point). 42 Shield gives Scouts an average 53% damage reduction at RR5 against all non dual-wielding opponents. Against dual-wielding opponents that's 26% damage reduction. That's against ALL melee attacks. Keep in mind that offhand attacks are unstyled attacks, and much easier to block against.

Taking in mind an average of 150 damage from a Ranger which would include standard style damage and offhand damage and damage add for both hands, the Scout can expect to block an average of 39 damage from a Ranger per Ranger's swing, counting the dual wield block reduction rate.

Now let's look at what a Ranger can gain from Celtic Dual. Considering the average spec of 18 CD (which is now unrealistic due to the custom style change on Atlas removing stun from Hib Stealthers), that's a total of 33 CD at RR5. That's a 47% chance to swing the offhand. With an average of 35 damage offhand and +15 damage from the DA, we're looking at 50 extra damage 47% of the time, for a grand total of an extra 23 average damage per swing.

This puts the Scout 16 damage per swing ahead of the Ranger.

What about procs though? Scouts can put a reactive proc on their shield, much as my Merc did. I loved it. Rangers can get a chance to proc the offhand 47% of the time they swing, which is 15% of the time they successfully swing and connect, but also open themselves up to reactive procs (a common complaint among SBs and Zerkers, among other dual wielders).

Now, more realistically due to the custom change which nerfed CD and removed the side stun in CD for Hib Stealth, I only recommend 8 CD for the rear snare. This puts Ranger and NS offhand chance more realistically at 40%, which brings the average extra damage to 20 per swing, putting Scouts ahead at 19 damage per swing.

The only thing that Scouts need to do to be more competitive in melee IS TO SPEC MORE MELEE, same as Hunters or Rangers do. When you catch a Sniper Ranger who specs 25 Pierce/18 CD or 25 Pierce/8 CD they drop faster than Scouts do, due to worse defense. I'm not aware of any common Sniper Hunter specs, but I'm sure they'd melt just as fast as Rangers if they specced low melee too.


Here are my recommended Scout specs, all assuming RR5:

Visi Scout
50 Slash
42 Shield
40 Archery
10 Stealth

Standard Scout
39 Slash
42 Shield
38 Archery
35 Stealth


So, all that being said, why do I play a Ranger, knowing the disadvantages of DW (low defense)? Sure, PF is cool. But the real reason I started playing a Ranger and continued to play a Ranger is because my first main for a long time was a Merc. I had a lot of fun in 8v8, zerg, smallman, and solo. However, when doing smallman and especially solo on early Live I got tired of being found by the rampant radar groups. I fell in love with dual wielding because honestly, it just looked cool. Polearm did more damage, but dual wield had that cool factor. I decided to play a Stealther because it was the only way to hide from radar. I also wanted to play something next that was less one-dimensional than a Tank or Assassin. The obvious choice was Ranger, since they were the only dual-wielding Archer, which definitely hit that coolness factor button for me. There you have it. The origin of Cadebrennus, and why I love playing a Ranger. The coolness factor.

 

[Fenyn]

Here (I don't remember the same on live) dual wielding had its reduction of block evade nerfed from 50% to 25%.

Small correction: There is no phoenix-equivalent nerf of dual wield penalty here. Evade and Block will be reduced by 50%.

If you have a 30% evade chance and someone attacks you with dual wield, you will have a 15% evade chance. Similarly, Parry will be reduced by 50% for 2h attacks.

This means that Hunter will gain an edge when fighting targets that will parry often, and Ranger will pull ahead when fighting targets that evade or block frequently. The Scout gains neither of these advantages due to using a 1h/shield combo.

 

[Saroi]

This means that Hunter will gain an edge when fighting targets that will parry often, and Ranger will pull ahead when fighting targets that evade or block frequently. The Scout gains neither of these advantages due to using a 1h/shield combo.

vs. evade or small shield Hunter has the benefit of his pet counting as a second attacker or not? I am not completely sure how much it reduces it here, I think evade should be divided by the number of attackers?

Against medium or large shield the Ranger definitely has the upper hand, giving that they can block 2+ without penalty.

 

[Fenyn]

That is true, the pet would grant an extra attacker. I know parry is divided across all attackers, but the evade code does not follow the same logic. Currently it is -3% evade for each attacker beyond the first.

A hunter pet would effectively remove one Dodger level's worth of evade from the target. It would be a boon, certainly, but against very evasive targets like assassins the ranger would see a significant edge still.

 

[Wreckanishu]

That is true, the pet would grant an extra attacker. I know parry is divided across all attackers, but the evade code does not follow the same logic. Currently it is -3% evade for each attacker beyond the first.

A hunter pet would effectively remove one Dodger level's worth of evade from the target. It would be a boon, certainly, but against very evasive targets like assassins the ranger would see a significant edge still.

Do you mean that on Atlas the Hunter pet will count as an extra attacker for the purpose of defense penetration? A few months back I actually went a looking for the original comments and posts from the person who went and did the testing on live, but here is a collection of quotes and results fragments.

Wyrdd wrote:From just a preliminary test (i.e. small data set, just looking for large trends), here are the results:

Only looking at the Hunter's attack rate since we do not have a basis of comparison for the pet.

No Pet: # (%-chance of attacks reaching that function) [%-of total]
Attacks: 107 attacks
Hits: 68 hits (63.6%)
Evades: 18 (16.8%) [16.8%]
Parries: 8 (9.0%) [7.5%]
Misses: 13 (16%) [12.1%]

With Pet:
Attacks: 156
Hits: 100 (64.1%)
Evades: 23 (14.7%) [14.7%]
Parries: 9 (6.8%) [5.8%]
Misses 24 (19.4%) [15.4%]

Edit: Testing subject was a RR8L1 Valewalker with ~130 Dex and Qui. No stat buffs used. 20+18? Parry and Evade 4.

This test does not prove much, even by "preliminary" standards. It does prove that the Hunter pet will not reduce defense by half, but beyond that...
Look at the parry rate. It takes a large penalty (24.4%) with the Hunter pet attacking. Of course the margin of error for this test is the size of the moon, but even the evade rate is reduced by 12.5% with the Pet attacking.
So the only trend we see with this test is that the Hunter pet will reduce enemy defense rate by x%
Another possibility is that the Hunter pet reduces defense rate by a small but constant percentage regardless of the target's total % chance to defend.
In other words, if the Hunter pet reduce's enemy defense by a flat 2-3%, then an enemy with a 20% chance to evade would have this chance reduced to 18-17%.
And an enemy with a 10% chance to evade would have this chance reduced to 8-7%.

Uthgard • View topic - Hunter's pet and enemie's defense

Actual live testing on both live 1.65 and Uthgard showed the Hunter pet does not count as an extra attacker for the purpose of defense penetration.

Afaik, Wyrdd did the testing back in 2005 because he was convinced the pet did count as an extra attacker from his own personal experience, but his testing disproved that. Further testing from Uth did too.

 

[Wreckanishu]

To clarify, i have seen no evidence whatsoever that Hunter pet did ever count as an extra attacker in actuality. Mythic did enquire about the test results to this Wyrd fellow, but nothing more came out of it.

Patch 1.22: Targeted spells now benefit more from extra attackers. The bonus per extra attacker is now 3.75% instead of 2%, so a full group attacking a monster should give a bonus of approximately 25% to the success rate for targeted spells (as opposed to 14%). Note that pets and elementals contribute to the number of extra attackers. This means that nukers in a group should be much more effective than they used to be. Casters who are not in groups will of course not be effected by these changes.

This patch note does mention that in theory it should. But it never did and was actually proven to not do and as of yet i have not found any evidence that it did.

Whether or not that was Mythic's wish to make a conscious balance decision to not apply this gamelogic to the hunter or whether it was a bug that never was fixed is a grey area to me.

The patchnote wording is also ambiguous as it seems to qualify this change as pertaining to nukers and casters and makes no mention of other pet classes like the minstrel or the hunter. Therefore i also think taking the sentence i highlighted in there as law to extend the logic to non-caster pet classes is also somewhat debatable.

All i know is the only bit of live testing from the era we're left with is that Wyrdd testing, and Uthgard probably just replicated it for consistency. /shrug.

 

[Fenyn]

No Pet: # (%-chance of attacks reaching that function) [%-of total]
Attacks: 107 attacks
Hits: 68 hits (63.6%)
Evades: 18 (16.8%) [16.8%]
Parries: 8 (9.0%) [7.5%]
Misses: 13 (16%) [12.1%]

With Pet:
Attacks: 156
Hits: 100 (64.1%)
Evades: 23 (14.7%) [14.7%]
Parries: 9 (6.8%) [5.8%]
Misses 24 (19.4%) [15.4%]

I am willing to accept that it might not be counted as an attacker, but I'm a little confused from this source. 100-150 attacks is really quite a low sample size for my preference, so the data is a little muddy but we can still see a difference in parry and evade rates with the pet attacking vs. without. That's echoed by one of the posters in that thread:

Look at the parry rate. It takes a large penalty (24.4%) with the Hunter pet attacking. Of course the margin of error for this test is the size of the moon, but even the evade rate is reduced by 12.5% with the Pet attacking.
So the only trend we see with this test is that the Hunter pet will reduce enemy defense rate by x%
Another possibility is that the Hunter pet reduces defense rate by a small but constant percentage regardless of the target's total % chance to defend.
In other words, if the Hunter pet reduce's enemy defense by a flat 2-3%, then an enemy with a 20% chance to evade would have this chance reduced to 18-17%.
And an enemy with a 10% chance to evade would have this chance reduced to 8-7%.

The 'small but consistent percentage' theory outlined above perfectly describes the current implementation and would fit with the observed data too. It seems as though we may be within parameters? I do agree it's somewhat unclear


A supporting consideration is that with our current system, and everything I know of the DAoC combat system as we have it implemented, is that the attacker count looks only at the number of incoming attacks with no consideration for the body type hitting it. Something like this would have to be coded to intentionally exclude hunter pets from the attacker count, which feels weird to me. Again, not a great reason on its own since it has no relevancy to how Mythic laid their code out back in 1.65, but it just feels out of place in the system.

 

[Wreckanishu]

Aye it's a shame when things get lost in the sands of time and i get what you're saying re how weird it would be. But the other way to look at it may have been that before that 1.22 patch, pets did not count as extra attackers in the code. Come 1.22, Mythic goes merrily on coding the caster classes (including minstrel charm which for all intents and purposes works like the mentalist one) and they just forget to do the Hunter pets?

Dol comes round, replicate their understanding of the coding, figure that it's just easier and better coding practice to just rule all entities to count as +1 extra attacker, and we get where we are.

Or maybe i'm being nonsensical from a coding standpoint. Either way, I wouldn't put it past Mythic to do weird(er) things.

Either way, it is indeed muddy, and having played the class extensively but never done any testing, i could never notice if the pet counted as +1 so.. bleh... I think it's just helpful to know what it's going to be like on Atlas, for clarity sake and move on from there.

 

[Cadebrennus]

Small correction: There is no phoenix-equivalent nerf of dual wield penalty here. Evade and Block will be reduced by 50%.

If you have a 30% evade chance and someone attacks you with dual wield, you will have a 15% evade chance. Similarly, Parry will be reduced by 50% for 2h attacks.

This means that Hunter will gain an edge when fighting targets that will parry often, and Ranger will pull ahead when fighting targets that evade or block frequently. The Scout gains neither of these advantages due to using a 1h/shield combo.

Glad you pointed this out, but this was a quote from my post on the Phoenix forum. I'll edit my post to make sure that it's more obvious so that readers won't think I'm talking about Atlas.

Side note; I'm glad to see some more spirited discussion about Archers in general. Please keep it up everyone!

 

[Saroi]

That is true, the pet would grant an extra attacker. I know parry is divided across all attackers, but the evade code does not follow the same logic. Currently it is -3% evade for each attacker beyond the first.

A hunter pet would effectively remove one Dodger level's worth of evade from the target. It would be a boon, certainly, but against very evasive targets like assassins the ranger would see a significant edge still.

Yeah then it was parry I had in mind. 3% less evade does sound very low but hard to find true values. I found a grab Bag from Shanya back in the day but there it just says that evade gets lowered by multiple attackers and sadly(but also expected) no %.

Are there any values how much an attacker reduces the block chance? And is it all the same?

If you have 4 attackers vs. Large shield, does the 4th attacker reduce your block the same like a 2nd attacker does vs. small?

 

[Cadebrennus]

Honestly I'm not sure about his numbers, (he was a notorious troll, but genuinely seems to want to help a new player here) but there is something to consider here.

Help for new Hunter - Phoenix

"...WS raises the higher the spec is, therefore your defense penetration also raises. however it has only minimal effect maybe 1-3% from 44 to 50
but being a hunter every % counts, together with you pet you will be able to drop evade7 to like 25% from originally ~45%

Horg solo Hunter Einherjar Lone Enforcer ~42,8% solo kills
Eik solo Shadowblade Flammen Fru ~50% solo kills"

 

[Fenyn]

Yeah then it was parry I had in mind. 3% less evade does sound very low but hard to find true values. I found a grab Bag from Shanya back in the day but there it just says that evade gets lowered by multiple attackers and sadly(but also expected) no %.

Are there any values how much an attacker reduces the block chance? And is it all the same?

If you have 4 attackers vs. Large shield, does the 4th attacker reduce your block the same like a 2nd attacker does vs. small?

For blocking, your block chance is only penalized once you exceed the value allowed by your shield size. For those unaware, a small shield can block 1 attacker, a medium shield can block 2, and a large shield can block 3 attackers without penalty.

Once you exceed this number, your block chance is reduced at a rate of ShieldSizeMax / AttackerCount.
A small shield blocking a second attacker would have 1 / 2, or 50%, of their normal block chance
A large shield blocking a fourth attacker would have 3 / 4, or 75%, of their normal block chance.
A fifth attacker against a large shield would reduce it to 3 / 5 of their normal block chance.


I'll put all the parry/evade/block info into a grab bag at some point since I think many are likely curious as well.

 

[Stringer]

There are no advance stealth skills here, so you're never going to be able to pop a true sight and see that solo stealther at enough range to crit shot him.

I just yesterday nuked a NS at 2k range due to true sight, so this point isnt exactly valid

 

[Cadebrennus]

I just yesterday nuked a NS at 2k range due to true sight, so this point isnt exactly valid

Cool. If it looks like a consistent thing, then let me know. I'll adjust my post to reflect the current environment