Prediction:
Gunslinger will be an elite spec with rifle. They’ll put buffed ricochet in it. It’ll be a GM trait in the same tier as a capstone rifle trait so you have to choose between them. The other gunslinger traits will have a common theme of pistol/rifle/utility choices. People will be mad about it for a week just like feline grace/daredevil. Then everyone will forget about it after a few months.
To give thieves the ability to wield rifles would require an elite specialization being made. Making an elite specialization for one profession requires making an elite specialization for all of them. Adding new specializations requires lore, which requires an expansion, which requires one year of development after focusing on at least a year’s worth of living story updates. So maybe it’ll come back; 2+ years from now.
Can we have the secondary effect of the thief’s Concealed Defeat trait changed/overhauled/removed? The ‘Create a smoke screen when downed’ portion is horrible in PvP since it gives away the fact that you were downed and your position when you’re downed (in case you were stealthed). It also provides little protection from the myriad of ways to get around the smokescreen’s blind. These two effects weren’t tied together before the massive traits overhaul, but since then this has been a massive nerf to deception skill usage.
A suggestion from a different thread 6 months ago is that it triggers at 25% health rather than when downed. That way, the secondary effect retains it use without ruining the deception recharge portion. Regardless of whether this trait is changed to trigger at 25%, 20%, 10%, or overhauled, it’s time for this design flaw to change.
Thanks.
(edited by Roy.7405)
Ironically, the only profession I’ve seen that seems to have trouble escaping the roots are rangers. Everything else (at least in the matches I’ve played) have so many dodges, teleports, invulnerabilities, high damage auto-attacks, PBAoE, and/or random AoE hitting it that it holds them for 1-2 seconds average, with mesmers and thieves seemingly invulnerable to it most of the time. Literally all thief weapon-sets except dagger/dagger and pistol/pistol have ways to get out of this effect without destroying it; and that assumes they don’t use withdraw, steal, or one of their utilities to escape.
Constructive feedback – 2 Resets per matchup:
Friday Afternoon:
There would be a 24 hour matchup from Friday afternoon- Saturday afternoon. This would be a time where WvW is competitive, a group’s contribution matters, team sizes are as even as WvW allows, and chaotic, which is what people enjoy most about reset/weekend-WvW and is why people enjoy it so much in the first place, and why no one wants to be cut out of it or have their fun limited. All the timezones would have equal importance, so there is no timezone getting shafted. It would allow smaller servers, who are usually outnumbered during the week, to show their skill and organization for a short 24-hour period; something that is completely lost over long matchups. The end of matchup rewards would remain the same (they’re pretty bad anyways =p) to encourage participation during that 24-hour period, but the results could have less or even no impact on the server ranking calculations.Saturday Afternoon:
On Saturday afternoon, the second reset would take place and continue for the rest of the 6-day week. We all know that it would be less competitive and more about population during this second part of the matchup, but at least more folks would get to experience reset which is what ANet’s primary objective is. This second part of the matchup would have the standard rewards and would be what Server rating is based off of.
If desired, as a second part of the above (slightly more complicated), there could even be two different server ratings that are tracked: A 24-hour rating and a Week-Long rating.
The 24-hour rating could be used to determine the 24-hour matchup for Friday afternoon. Since smaller servers may be able to compete better with larger ones during these shorter time periods, there could be a little more variety in the matchups, and with proper incentives could be competitive across tiers.
The matchup determined by the Week-Long rating would be the match that is started on Saturday afternoon and would be the 6-day matchup. The matchups in the 6 day match don’t have to be (and probably wouldn’t be) the same as the 24-hour matchups.
The 24-hour match could be determined 6 days in advance (since there are 6 days in between them), while the Week-Long matchup is known a day in advance, allowing time for coordination should guilds or fight clubs wish to organize something.
(edited by Roy.7405)
Constructive feedback – 2 Resets per matchup:
Friday Afternoon:
There would be a 24 hour matchup from Friday afternoon- Saturday afternoon. This would be a time where WvW is competitive, a group’s contribution matters, team sizes are as even as WvW allows, and chaotic, which is what people enjoy most about reset/weekend-WvW and is why people enjoy it so much in the first place, and why no one wants to be cut out of it or have their fun limited. All the timezones would have equal importance, so there is no timezone getting shafted. It would allow smaller servers, who are usually outnumbered during the week, to show their skill and organization for a short 24-hour period; something that is completely lost over long matchups. The end of matchup rewards would remain the same (they’re pretty bad anyways =p) to encourage participation during that 24-hour period, but the results could have less or even no impact on the server ranking calculations.
Saturday Afternoon:
On Saturday afternoon, the second reset would take place and continue for the rest of the 6-day week. We all know that it would be less competitive and more about population during this second part of the matchup, but at least more folks would get to experience reset which is what ANet’s primary objective is. This second part of the matchup would have the standard rewards and would be what Server rating is based off of.
I hope the WvW team actually takes a very, very serious look at part of Chaba’s and some others’ suggestions, because I actually like the idea of having two different resets and think it would work a lot better than what is shaping up right now.
There would be a 24 hour matchup from Friday afternoon- Saturday afternoon. This would be a time where WvW is competitive, a group’s contribution matters, team sizes are as even as WvW allows, and chaotic, which is what people enjoy most about reset/weekend-WvW and is why people enjoy it so much in the first place, and why no one wants to be cut out of it or have their fun limited. All the timezones would have equal importance, so there is no timezone getting shafted. It would also allow smaller servers to show their skill and organization for a short 24-hour period; something that is completely lost over long matchups. The end of matchup rewards would remain the same (they’re pretty bad anyways =p) to encourage participation during that 24-hour period, but the results could have less or even no impact on the server ranking calculations.
On Saturday afternoon, the second reset would take place and continue for the rest of the 6-day week. We all know that it would be less competitive and more about population during this second part of the matchup, but at least more folks would get to experience reset which is what ANet’s primary objective is. This second part of the matchup would have the standard rewards and would be what Server rating is based off of.
As others have mentioned, the Saturday reset seems set in stone for the time being regardless of what is posted here since ANet wants more people to experience reset and due to HoT, however the above seems like a suitable middle ground once the initial HoT patches are rolled out.
(edited by Roy.7405)
Also if you’re using Windows 10 (and maybe Windows 8 as well), you can go to Settings->Network & Internet->Data usage tab->Usage details and see what programs are using the most data over the 30 days, to get an overview of what’s using up the most bandwidth on that computer.
You have control as to what utilities you want to bring in the form of predefinied roles.
In case of revenant their utility skills serve identical roles to attunements. You swap to earth to gain tanky ability skills and boons. As revenant you swap to Jalis to get tanky abilities. Not much of a difference. It wont feel right if you start causing damage in earth with fire skills, and it wont feel right if you start causing conditions in Jalis.
Regarding that last point, there are many people who’d disagree. “it wont feel right if you start causing conditions in Jalis.” Switching between Mallyx and Jalis while using mace/hammer allowed me to (ideally, the utilities need improvements) switch between two slightly different styles of condition damage play, kind of like how stats can affect the way you play the same (Weapon+Utility) combination.
- Mallyx would be a bruiser style damage dealer who is capable of staying in the thick of a fight for a moderate duration and reflecting some incoming damage back to enemies. (akin to Rabid stats)
- Jalis would be the more tanky playstyle which is capable of staying in the center of a fight for long durations, while dealing slightly less damage. (Dire stats)
- Ventari utility skills would provide a more team-defense oriented condi-user who is capable of assisting allies at the cost of less damage. (Settler’s stats)
- Shiro skills would be a more bursty style condition user who is able to deal out a higher amount of damage than normal, but can’t stay in the center of a fight for long due to a lack of defensive stats or utils. (Rampager stats)
Edit: To clarify: I used Mallyx for damage and Jalis for tanking/cc, all the while using condition damage as my main source of damage. Mallyx did more damage, but I needed to pop into Jalis quite a few times for the stability/tanking abilities.
__________________________________________________________________
On another note, I wouldn’t want legends to simply affect a weapon’s skills because the legend weapon skills probably wouldn’t suit the armor stats you chos; Jalis and Mallyx skills/conditions for mace/axe probably wouldn’t benefit equally from armor with the Condition Damage stat. It also still wouldn’t solve the core Melee vs Ranged problem. You’d still be locked into 2 playstyles like most other professions, but would critically lack the important ability to switch between melee and ranged combat during a fight.
How I arrived at the playstyle count (based on weapons and utilities):
2| Warrior | (WeaponSet1+Utils),(WeaponSet2+Utils)
2| Revenant| (WeaponSet+Utils1),(WeaponSet+Utils2) ((Currently))
4| Ele | (Fire+Utils),(Water+Utils),(Air+Utils),(Earth+Utils)
Ele and engis can arguably be higher depending on how you count conjures/kits. I’d prefer to see Rev’s in the 4 playstyle-category, as the heavy armor class, with the unique combination of:
4|Rev|(WeaponSet1+Utils1),(WeaponSet1+Utils2),(WeaponSet2+Utils1),(WeaponSet2+Utils2)
While locking each weaponset to a specific legend (ex: letting the user choose axe for Mallyx, hammer for Ventari) would solve the melee vs ranged problem, it feels like it could be unnecessarily limiting when you can have more variety; 4 playstyles instead of just 2. Regardless, how/if weapon swapping is implemented will make the difference between the perception of how hard the class is to play; the difference between a basic warrior build vs a sophisticated ele/engi build.
(edited by Roy.7405)
No offense, but I really don’t agree with anything you’ve suggested.
I’ve played the majority of the beta so far with Mace + Axe using King Jalis and Mallyx.
First of all…. you said the “elite for Mallyx is fine”…. which just hurts my soul.
Did you even use this skill, lol?
The skill literally does nothing except apply an equal amount of torment to you and enemies if you don’t already have conditions on you.
…and if you do have conditions on you, the skill doesn’t even apply enough resistance to help you at all. The non-elite necromancer skill Plauge signet is literally better than this.
I believe like you just wrote a description for many of these things just to feel like you said something without actually having experienced them.
Nah, while I don’t necessarily agree with the degree of the all the changes, he was pretty much spot on regarding noticing the issues; and if you’re in PvE then yes the Mallyx elite’s effectiveness will be greatly affected by what creatures you’re up against.
After playing 25+ matches in sPvP, about 20 using the Mace/Axe + Jalis/Mallyx combo, I was able to notice the large effect the Mallyx elite was having against certain enemies (copying 6+ conditions to all nearby foes with little condition clearing such as an engis/rangers), but the class’s overall damage output was too low to capitalize on it. Adding weapon swapping and making small changes to nearly all of the things he mentioned would have turned things around in those circumstances.
On another note, I can see Unyielding Anguish getting a recharge of 5 seconds added in the future once the other changes are made; although the field last 4 seconds anyways so it just makes it slightly less spammable. That AoE teleport reminds me of the problematic decap engi’s from a while back and I can tell that once the other needed changes are made it’ll be OP.
Regarding the mace, it seems that skill #3 may have became bugged due to the new build Friday night. Before the build, all 3 blast finishers were occurring within the 410 range mentioned in its description, so combined with the fire field from skill #2 you could get all 3 blasts to land within the fire field (9 might stacks every 8 seconds). Afterwards, it changed and now it seems that the 410 range is the distance between each blast, which could have been the intended behavior but I doubt they’d balance a skill that quickly so it’s probably a bug.
In the grand scheme of things, this profession will need to be able to weapon swap to become viable in most people’s eyes, since not being able to switch to a melee or ranged play-style is a huge balance issue that will need to be addressed. Assuming that revenants are given weapon swapping, additionally toning down some of the casting times by .25s-.5s and slightly increasing damage for the weapon skills would address a sizeable amount of the major balance issues that exist.
“Jora was inspired by Daryl Hannah, the actress who played Ayla, in The Clan of the Cave Bear”.
You can get dishonor points for quitting a match, but that doesn’t affect the matchmaking algorithm. But with enough dishonor points you cannot enter the queue at all.
The average queue time is based on a number of things. It’s quite frequent to get above average queue times if you’re playing a very common profession in that time slot or are a very good (or bad) pvper, as matchmaking will try to create a team that balances out you + your allies skill vs the other team’s skills.
http://wiki.guildwars2.com/wiki/PvP_Matchmaking_Algorithm#Matchmaking
If you have a storage character or a personal guild bank, you can store it in their too. But after a while, you’ll eventually want to consider just deleting a couple stacks. I passed that point a long time ago. =p
I just don’t think Anet wants to go this route because they think it’s a “restriction” to players with unlimited time on their hands.
I see your point with the unlimited time. Of course the leaderboard shouldn’t be determined by those with the most time because the “most free time” folks make up a small percentage of the population; although that it was the current setup does. To resolve their issue however, the following could be implemented:
Set a cap to the number of points that would affect leaderboard ranking. Points can be earned beyond that cap, but they do not affect leaderboard ranking. Everyone earns rewards (see below) based on the number of points acquired. Visually, the current leaderboards site could be modified to indicate that a player has reached the cap; or a new column could be added called “Total Points” to see their total points earned; not just the season cap points. Using a moving-cap while allowing points to spill over the season cap would be pretty complicated; a set cap is easier to implement and understand.
Since there wouldn’t be a moving cap to require players to play weekly, a new system would be needed to encourage them to play often. This can be achieved by adding a rewards system that’s active only during the season.
At the end of a season, all players are rewarded for either:
A) For every X (ex.: 25) points they have earned. Ex: Reward track boosters.
B) For each point earned. Ex: Some silver for each point.
C) Both. Rewards from (A) and (B) might have to reduced.
Note: The reward has to be distributed at the end of the season so that players can’t abuse going above and below the benchmark (25pts) for repeat rewards. Also, with every player getting at least a small reward for participating, it encourages participation throughout the season for everyone even if someone knows they could never place highly.
_________________________________________________________________
So to summarize:
There is a season cap towards the number of points that affect leaderboard ranking. Additional points can be earned, but it will not affect your leaderboard ranking. If multiple players have the same number of points, MMR is used as a tiebreaker.
At the end of the season, all players are rewarded based on the number of points earned. Reward setup can be:
A) Large rewards at benchmark values.
B) Small rewards for each point earned.
C) Both. Adjust rewards accordingly.
There are two systems that work to keep people playing throughout the season:
- Reward players (including those with lots of free time) by rewarding them for each point earned.
- Discourage players from only playing at the beginning of the season via MMR decay.
Finally, that does seem like a lot of rewards on top of the reward tracks, dailies, match win rewards, and top 250 rewards; but that’s what it takes for a competetive points-based ladder to allow infinite points without punishing anyone. The best/unique rewards are still reserved for those at the top of the ladder.
________________________________________________________
Example numbers:
For a 30 day season with a 120 point cap (which is 4 points per day average):
- Using my points per game avg of .67 and assuming that stays constant, that is 180 games in a month, or 6 games a day average.
- Using the leaderboard leader’s values (.65 pts per game), that is 184 games in a month.
My issue with the proposed leaderboard A would be that the number of points earned (and thus games played) would still be the sole determinant of the final leaderboard position. Although it would be fairer than the current one, it still leaves open the situation where a player doesn’t place solely because they played a couple of games less than another player. It also shifts the time burden from being fairly equal for everyone to being mainly on lower ranked (and likely newer) players. For me personally, during the test season I think I only played 6 ranked games, partly because I knew I wouldn’t be able to compete if the determining factor was solely time played. I suspect for many other players that was a factor too, and shifting the time burden to a different populace won’t help much.
Based on their execution, ANet’s idea with the leaderboards is that anyone has a chance to come out on top; not just tournament winners. The way they have achieved this is by having essentially time-played being the determining factor in the leaderboards, instead of MMR. That way, even players with a low MMR can still place highly if they participate a lot. Which thus achieves their second goal of getting more players participating in sPVP. With the main goal of getting more players into sPvP in mind, the current setup doesn’t work as well as it could because a lot of players (like me) will quickly realize that if they’re expected to get a massive amount of points a day just to place at #250 to qualify for a reward, then it’s just not going to happen. Hence the setup is only encouraging a small number of players (who have the most free time) to participate. Ironically, this is probably the same number of people who think they can place in the top 250 in an MMR based leaderboard. I don’t have a problem with time played being a factor, but it can’t be the determining factor.
=============
With the goal that lower MMR players can potentially place in mind, my suggestion from a previous post is:
A points based leaderboard with a moving season-cap as the season progresses, and MMR is used as the tie-breaker for players with equal points.
For instance, if the moving cap was based on 4 pts per day 28 points per week for a 28 day season:
Day 1 in season, max points you can have is 4.
Day 7 in season, max points you can have is 28.
Day 28 in season, max points you can have is 112.
By the end of a 28 day season, the max points you can have is 112. If multiple players have 112 points, MMR is used as the tie-breaker. Many players may not have time to accumulate 112 points in a season (150+ games in a month), leaving room for dedicated PvPers with lower MMR a chance to gain rewards. Also, if a player only plays on weekends, they still have an opportunity to catch up in points because they’re not locked out. Finally, since its a moving cap instead of a fixed cap, it encourages player participation throughout the season instead of just a single week of grinding at the start; and players must also play occasionally to avoid MMR decay.
EDIT
For the example, have the moving cap set at 28 points per week instead of 4 points a day; so that players don’t have to login on the last day(s) to qualify for the top spot.
(edited by Roy.7405)
Or how about a points based leaderboard with a moving season-cap as the season progresses, and MMR is used as the tie-breaker?
For instance, if the moving cap was based on 4 pts per day for a 28 day season:
Day 1 in season, max points you can have is 4.
Day 7 in season, max points you can have is 28.
Day 28 in season, max points you can have is 112.
By the end of a 28 day season, the max points you can have is 112. If multiple players have 112 points, MMR is used as the tie-breaker. Most players may not have time to accumulate 112 points in a season (150+ games in a month), leaving room for dedicated PvPers with lower MMR a chance to gain rewards. Also, if a player only plays on weekends, they still have an opportunity to catch up in points because they’re not locked out. Finally, since its a moving cap instead of a fixed cap, it encourages player participation throughout the season instead of just a single week of grinding at the start; and players must also play occasionally to avoid MMR decay.
The difficulty new players have understanding sPvP is actually a symptom of a much greater, core problem of the game; something the New Player Experience patch was supposed to address but failed (miserably) at.
Considering that almost nowhere in the game players are taught about the following (let alone their importance), I’m surprised so many stick around as it is:
Boons
Conditions
Crowd Control
Stun Breakers
Weapon Swapping
Their Class’s Profession Mechanic
Runes
Sigils
Combo Fields
Combo Finishers
Dodging
Sure, players don’t have to learn all of the above list at once, but at some point they need to be eased into learning them so that they are not annihilated when they enter sPvP playing against people who do. The problem is, they never are. I wouldn’t be surprised if most players don’t learn about any of these until they enter Dungeons, WvW, or sPvP; assuming they learn it at all. Which they don’t need to for almost all of the open world and story content, which is likely part of the problem.
Both ranked and unranked attempt to match you based partly on your skill with that profession, so either should be fine. As far as builds go, I prefer to make my own because that’s what I enjoy and I do pretty well with them. (Nearly) Every build has a weakness, so after a while it doesn’t really matter what you run, its about your experience with using and knowing its limitations.
As far as hotkeys go, I’ve tried using different combinations and the standard WASD, but I still prefer the arrow keys and the number pad for movement and skills. However, if you want to try to reduce your needed reach you could try assigning skills 6-10 to the shift key. So [Shift] + 1 would map to skill 6, etc. I didn’t like it, but maybe you will.
Goodluck.
Would it be possible to have a MMR column added to the leaderboards (i.e. how it used to sort by), even if doesn’t show the value and only shows your “Top 1000/Percentile Rank”. I’m sure there are many folks who would be more interested in that than the current sort by ladder points, which from what I see seems to favor those with the most play time almost regardless of skill. That way players can see where they match up against others skill-wise and not just time-wise.
On another note, would it also be possible to cap the number of points players can gain per ladder season. Again, at the moment it seems that simply playing a lot of games (and having a lot of time) gets you high on the leaderboards; putting in a high per-season-average cap would make it less of a grind to achieve a high rank, but still require lots of participation.
For example, with a per-season-average cap of 8 points per day:
5 days into the season, you cannot surpass 40 points.
8 days into the season, you cannot surpass 64 points.
You can get more than 8 points per day (like 30 on a single Saturday) so weekend warriors would be fine, but you cannot surpass the [“per-season average” x “days season has been active”] value. Tiebreakers are determined by MMR, and if you have a lower MMR than someone with a matching number of points then you must play and win additional matches to pass them, since that is the only way to improve your MMR. MMR would also need to include odds of victory in its calculation; otherwise players in organized teams (which are less likely to lose) would always have a higher MMR than people who queue solo.
The only thing server really matters for is WvW, which is a different form of PvP.
Take a look at the link below which has each server’s rank. Higher ranked servers tend to have more action.
http://mos.millenium.org/na/matchups/
Courtyard is fine. Any team that loses by more than 400 points is going to lose no matter what map it is. Although I think giving bonus points for killing the opposing player with the ‘most kills’/‘least deaths’ sounds interesting.
If you’re looking for more action at nearly any timezone on any day, I’d recommend a Tier 1 or Tier 2 server. If you’re looking for overwhelming victories (trust me, that gets boring quickly even if you’re used to losing) you’ll probably want to look at joining a tier 3-7 server since its not uncommon to get matched up against much weaker (or stronger) servers.
Looking at the site below may help you; it shows the current scores of all the matches and some other info you may find useful.
http://mos.millenium.org/na/matchups/
I didn’t enjoy the fight either, but part of that could be due to me doing it on my ranger and my pet was dead most of the time due to the AoE.
I think the wording and signaling on what to do could be a better. It took me a while to figure out to do for the facet of darkness boss fights as I ended up completing the 1st part before the boss pretty quickly and not knowing what happened.
Specifically, it’s especially confusing when you look at the boss abilities that says ~“Steals boons from the player”. I instantly thought it was talking about swiftness, retaliation, etc (i.e. what the rest of the game teaches you and call a boon) and not some instance specific ‘boon’. Perhaps boon can be replaced with ‘enchantment/buff’ or some other unofficial term that doesn’t conflict with the game’s established terminology. Secondly, the Fragility effect is not really a boon in the first place; becoming fragile and lowering your health sounds a lot more like a hex/debuff.
Finally, I’d suggest relooking at how the vortexes are presented. If the vortexes were presented as non-hostile until the ‘boon’ was brought to them (since they can’t be killed via damage before then anyways), I think that would go a long way to helping clear up the confusion. Or perhaps replace their white AoE rings with the standard orange?
Indeed, the move toward 5v5 in hotjoin was to help people better understand how to play tourneys properly.
I’ve heard this stated by many people before and after 8v8’s removal, but strongly doubt that this was ANet’s primary goal. I strongly believe that they removed 8v8:
- To increase the number of people in SoloQ and TeamQ. Everyone knew before the change that the queue times were very long; and depending on when you play they still are. By reducing the number of game modes from 2 to 1, ANet increased the player base of the remaining game mode (5v5) while losing players who only enjoyed 8v8. That condensing of the player base increases the number of players in the queues, thus reducing queue time. Combined with the rewards rework, the new PvP structure encourages players to participate in SoloQ over hotjoin for better rewards in a nearly identical game mode; which further increases the queue population.
As for when they did it:
- They removed 8v8 in the same update that they removed hotjoin because it was the only time they’d ever get a chance to do it. Removing 8v8 was unpopular and would assuredly reduce the player base, so this had to be done during an update that was large enough to hide the player exodus. The reason it was removed in the same update as the full rewards rework was because the rewards rework was guaranteed to bring in players who would want to check out the changes; and some of them stayed, although that number has been dwindling. The full rewards rework update was the only content update that could hide the removal of sPvP’s most played game mode. Had it been removed in a separate or standalone update, the exodus would have been obvious.
What I think they should’ve at least done if they were trying to increase the PvP player base:
- There should’ve been more changes and tools to teach new players the game mode. Overall, the game does a poor job of teaching the basics across all game modes. ABSOLUTELY NOWHERE in the game are you taught about boons, conditions, crowd control, stun breakers, the value of weapon swapping, your profession mechanic, runes, sigils, combo fields, combo finishers, and the list goes on. Players may unlock some things at later levels to reduce the learning curve, but even though players are given new tools to use, they are never taught how to use them. You could even argue that players don’t even need to learn about most of those things until they reach sPvP, WvW, or dungeons. I would think that if ANet really wanted to help players prepare for tournaments and PvP in general, they would’ve upgraded the Heart of the Mists with more tools to help new players (and veterans) learn game mechanics and test builds, similar to GW1’s Isle of the Nameless; or at least implemented an optional tutorial that taught game mechanics.
- Since they are encouraging/forcing players into SoloQ to get a decent playing experience, they should have updated the matchmaking algorithm at the same time so that new players can’t get stuck playing veterans after a lucky win. New players should start near the bottom with other new players so that they can learn the game while facing off against equally skilled peers. From there they can learn and gain experience with the game mode, moving up the ladder as they get better.
Finally:
- PvP is a lot more difficult than PvE, and it will take time for new players to adjust. However, there no longer is a training ground in sPvP for new players. 5v5 does not allow for small victories that encourage new players to keep trying. What it does do is it punishes the player (and their team) more for their failure, raising the stakes to levels new PvPers shouldn’t be forced to encounter.
- They would’ve addressed the spectator mode abuse by now, which has created a toxic environment for both new and veteran players. While other features such as build templates and PvP leagues would be nice things to have, those features should be much less of a priority than fixing game-breaking issues in hotjoin. Hotjoin is the new player experience for PvPers, akin to Queensdale in PvE (and its past champ train plight). When a player’s first few experiences in sPvP are toxic and often defined by uneven teams and one-sided beat-downs coupled with reduced rewards, no small victories, and no sense of personal improvement, you are inevitably going to drive away new and veteran players; and that is truly game breaking.
Thus, regardless of whether players like 8v8 or not, I find it disingenuous for ANet to say that they made this change to help new players when, in my opinion, it clearly was not.
Anyways, still like to know what players favorite thing in GW1 was.
As a PvPer, I personally enjoyed Alliance Battles (AB, 12vs12) and Fort Aspenwood (FA, 8vs8) the most. Both were fairly PvP balanced game modes that allowed for a large number of players to compete. There weren’t so many players that your contribution didn’t matter (very much unlike WvW, specifically zergs), but there weren’t so few that victory depended on the competency of a single player (unlike 5vs5 sPvP). You could find fairly balanced engagements (3vs3, etc) very quickly, unlike WvW where you could go 30+ minutes solo or in a group before finding something remotely similar. Most importantly, those game modes allowed for an unparalleled variety of builds to be played; if you wanted to play a Grenth’s Balance spike necromancer or some other non-meta build, you could and could still be very effective. In sPvP, if you’re not running something close to the meta, then you probably don’t have enough teammates to cover your build’s weaknesses or you might be running something that their compositions can’t cover for; like using full berserker stats on a team with no bunkers/tanks, even though there’s nothing wrong with your individual build. If sPvP were 8v8 or 10vs10, you could hide behind teammates for a little while if your health dropped too low, then reengage once you recovered; you’re disappearance for 20 seconds isn’t going to be critical to victory. In 5vs5 though, it will be.
Anyways, as for GW2’s persistent PvE world, overall I like it but the biggest thing holding back the game relate to core design decisions made very early in development that would be very time-consuming to change for existing maps. GW2 is designed so that the maps and most events can handle anywhere from 10-300 players; they might not handle it well or be as fun, but at least they don’t bug out, can be completed, and everyone can still get credit for participating. Most of the small and large events in the game are actually fun and can handle groups of around 20 players well, ranging from low-level events in Queensdale to world bosses such as the MegaDestroyer (MD). The issue is that there is nothing preventing that from that limit from being surpassed. And since more players makes events easier, you could argue that the game’s design encourages that limit to be passed. The world boss schedule also encourages this: although now more people can experience the MD event since its easier to get the necessary 20 players, it also waters down the experience because instead of 20 you get to 200. The Marionnette event, Dry Top, the map wide Scarlet Invasions: all of those and more were Anet experimenting with remedying this problem by spreading players out; and had various levels of success. Personally, I don’t think it’s something that can be solved well without reducing the map limit to around 100 for (most) future designed maps which would be smaller, with no modifications of existing maps since that probably isn’t the best use of their time. Basically, the existing maps were made for 300ish players to be spread out, not all grouped up in one area.
The second biggest issue, AI competency, is heavily linked to the first; and is an area where GW1 greatly surpasses GW2. The reason though is that if you have events such as the Orrian Temples and Tequatl where 100’s of creatures are fighting in a small area, you not only can’t afford to spend the server’s processing power on more complex AI, but no one would even notice if you did: there are too many particle effects and the creatures would die too quickly. So it’s not worth it in the open world. Some system of scaling AI based on player population could possibly be developed but that would be a massive undertaking: not only to develop and test, but to then implement to all of the game’s events. While open-world events wouldn’t benefit from more complex AI, creatures that would benefit from more complex AI include dungeons and player-owned summons. And in fact, some player owned summons such as ranger pets and mesmer clones have seen improvements. Unfortunately, whether for technical or for developer-time related reasons, NPC enemies haven’t seen similar improvements.
The predominant reason for removing 8v8 was so that more players would feel inclined to join the 5v5 solo and team queues. If hotjoin is just an unranked soloQ, most people would just soloQ to try out new builds and to get better rewards while they’re at it; especially if they don’t care about their soloQ rank. More players in the queues also means that the matchmaking system works better in finding appropriately skilled opponents.
Removing 8v8 in the same update that they released the reward tracks was a way of masking the effect of 8v8 removal. If 8v8 had been removed in a separate update, the drop in players would have been obvious and extremely unpopular. The influx of players from the new reward tracks masked the dismissal of many players who no longer found the game mode fun.
As for helping new players learn the game mode, I would argue that removing 8v8 actually lowers the number of players in sPvP, especially in the long term. A new player could still have fun in a 7 vs 8 and learn the game at a slower pace, where their inexperience has less of a negative impact on their team, which can still succeed despite that inexperienced player . When they eventually want better rewards they could move to 5v5. There, they would have started off near the bottom of the ladder, because they’re new and need experience, and work there way up while learning the game mode in the arenas; or possibly practice some more in the 5v5 hotjoin arenas.
Instead, the current system throws them immediately into the 4v5 hotjoins where they will get hopelessly defeated by more experienced players repeatedly until they get demoralized and perhaps never enter sPvP again. It is very hard to convince new players to stick around or even show up after that kind of experience, and knowing more is to come for the foreseeable future. Just look at the 3rd place team in nearly every WvW matchup since launch if you need evidence of that fact.
Finally, if they actually wanted to help players learn the game, they would have put in a better tutorial and upgraded the HotM to have more build testing tools; similar to how GW1 had Isle of the Nameless with its condition dummies for learning the different effects, CC or Condition focused NPCs that roughly mirrored actual player builds, etc.
The level 80 upgrades such as orbs, crests, and jewels don’t appear to be working because the backpieces are level 78. I’m not sure if the level on the back items were meant to be 80, or if this is working as intended. Lower level upgrades such as crystals are working fine for me.
I don’t think it would be an issue if it was implemented as weapons. For example, you equip a weapon called “Invisible Daggers” that is a mostly invisible dagger. It would also likely be made as a very rare/expensive weapon, like Volcanus, etc. so that everyone is not running around with invisible daggers, shields, etc.
The bigger issue would be when those weapons are brought to sPvP or WvW. For PvP balance the weapons would need to become highly visible at some point so that players can recognize and react to attacks (‘tells’) such as Kill Shot. So making the weapons have a visible form during skill activation would be a necessity; which would defeat the purpose of invisible weapons. Although there might be ways to use particle effects and transparencies to achieve an acceptable result; especially since the attack ‘tells’ are tied to the skills themselves and not weapons. So skills like Kill Shot still show their animation, just with a semi-transparent rifle.
So we’ll get NOTHING for GW2’s 2nd birthday??! o_o
Not necessarily. There’s a relatively new birthday vendor NPC in the center of Lion’s Arch, so they’re at least aware its coming up.
http://wiki.guildwars2.com/wiki/Birthday_Vendor
As for the Balancing Act reveal, it’s probably related to profession balances; if it is, you’ll probably get more value from watching the more detailed videos that are in their “Ready Up” shows and/or reading the text summaries. They are revealing the planned changes for 2 professions every 2 weeks on the shows. I think the next one is this Friday.
In case you haven’t seen them:
Thief and Warrior
Video: http://www.twitch.tv/guildwars2/c/4859579
Text Summary: http://dulfy.net/2014/08/08/gw2-ready-up-upcoming-warrior-and-thief-class-changes/
Engineer and Ranger
Video: http://www.twitch.tv/guildwars2/c/4743783
Text Summary: http://dulfy.net/2014/07/25/gw2-upcoming-changes-to-engineer-and-ranger/
These have been in the GW2 .dat file since at least June 19, 2013. They were never released, and the codes don’t work anymore; likely due to the wardrobe changes. You can still see screenshots in the reddit thread though.
http://www.reddit.com/r/Guildwars2/comments/1gpb8n/school_clothing_codes/
Are you doing hotjoin only? There is a 2 tiers per day limit for hotjoin and custom arena matches. To advance it further in a single day, you have to play in either Solo or Team Arenas.
http://wiki.guildwars2.com/wiki/Reward_track#Reward_Tracks
I was going to do a post a long time ago in the WvW forums, but I never did as it started to get too long. Summary of the overall WvW Portion, not necessarily relevant to this post.
To help make matches more competitive (and other stuff) by reducing the extreme differences in population:
- Reduce the number of maps from 4 to 3, thus reducing max player count (and max population difference) from ~500 to ~375. Reduces the number of maps a small force needs to cover during off hours and the score by which they can fall behind by.
- Remove the borderlands maps, which removes the necessity for all 3 sides to have a map in which they have an advantage.
- Make a main, matchup based WvW version of the EotM map. Provides a familiar map for newish players who have only experienced the overflow version of EotM. Address outstanding issues.
- Make a new (maybe smaller) map that incorporates a bloodlust-like mechanic.
- Move the borderlands jumping puzzle to Obsidium Sanctum (1 completion counts as credit for the 3 borderlands versions)
- Allows for potentially swapping in new WvW maps every once in a while.
- Change bank and crafting station access.
Anyways, the new bank access portion was basically this:
With the borderlands removed, the new banking location would be the Aetherblade Airship (currently located above Gendarran fields). As added flavor/style, the airship could be moved around to different cities/zones every once in a while, like Divinity’s Reach, Rata Sum, etc so that players aren’t completely disconnected from the rest of the world (i.e. map chat, nearby events, etc) and remain in a PvE area. Of course, the airship can be placed in many zones as well, and a random zone is selected each time. Bank upgrades on WvW keeps remain untouched. Talking to an NPC or using the item again would return you to either: the location you were previously (excluding WvW and instances) or the zone the airship is in.
Access to the airship is granted in two different ways:
- Using the Captain’s Airship Pass(or any version of it).
- Free access by talking to an NPC in the WvW maps, HotM, or from whatever zone it’s currently docked in.
From a design standpoint, it seems that the new LA map is focused around not having too many players in the same spot, thus there are 3 banking locations. I assume this design would extend to the Heart of the Mists. Thus, placing a bank NPC in the HotM would make it the new AFK spot and detract from its purpose as a sPvP staging area; and since it conflicts with spreading out AFK players is less likely to happen. It may also come as a shock to new players who have never entered the mists before and weren’t expecting to get sent to an area where their gear/trait selection is changed, are set to level 80, can’t figure out how to leave, etc. On the positive side, it could have the potential to attract new players into sPvP while they’re there, or encourage regular players to play a game or 2 while they’re there.
Just some things to consider.
As for the original change that highlighted this behavior, the removal of 8v8, I don’t think it has worked out and the default “Play Now” needs to change back to 8v8. It looks like the purpose of the change was to (1) increase the number of people entering into ranked matches and possibly (2) help new players learn the game mode. The number of participants in ranked matches has probably increased due to the rewards changes alone, especially since hotjoin is basically the same thing as ranked arenas but with better rewards in this all 5v5 format. But with regards to (2) learning the game mode, the removal of 8v8 has removed a way for players to test out builds and learn the conquest game mode in a slower-paced and more forgiving manner. 8v8 allows newer players to learn over time while enjoying a low level of success; 5v5 beats them down constantly until they do or quit. Neither 8v8 nor 5v5 teaches them what/how to learn the game; but at least 8v8 lets them fumble around in a more enjoyable manner, thus making it more likely that they’ll return. IMO, having low to moderate success in an 8v8 has a much greater chance of encouraging future participation and giving them time to learn than constantly getting beat in a 5v5. And then if/when players want to earn better rewards, they can move to the ranked ladders. As for the transition from 8v8 to 5v5 goes, that’s the whole point of the lower end of the ladder. Newer and less skilled players start near the bottom, and as they get better and learn right from wrong they move up; always playing with players of equal rank and skill.
In conclusion, for spectator mode my preference would be to have a timeout for players who enter spectator mode so that the likelyhood of that player being able to switch to the team they want to stack on is greatly diminished. At this point simply fixing the auto-balance abuse wouldn’t be enough to get me back into sPvP personally since it now lacks a game mode that I enjoy, but it would be a big step in the right direction.
Really this issue is motivated by the same things that caused you guys to drop 8v8.
What many players want is a casual enviroment to test builds/classes that can effectively simulate a game of conquest, albeit without even matchmaking.
What spectator mode (and more to that effect, uneven teams) does is prevent a ‘reasonable’ simulation of conquest for things to be tested. That limits the hotjoin experience to either very new players, farmers, or players like me who have played so many thousands of tpvp games that its more of a moving golem experience.
But that said seeing how it operates, i can see many players new and old alike being turned off due to team unbalance.
I agree with this sentiment. There were a few issues with hotjoin before the removal of 8v8, but they were usually masked by having large team sizes. Personally, I have barely done any sPvP since the 8v8 arenas were removed, excluding the month after the change was made since I was checking out the new Courtyard map and reward structure. Nowadays, its very difficult to find an 8v8 arena with people in it. Anyways, the two biggest reasons by far for me leaving sPvP were:
- The removal of 8v8 standard arenas. I enjoyed them over 5v5 for multiple reasons, especially for their semi-chaotic nature.
- The SpectatorModeAbuse (SMA for short) to switch teams.
I can understand why players would at least believe that there was an increase of SMA after 8v8 was removed, although it was present both before and after. Reducing the team sizes means that each player’s individual contribution is more important than before, and that a player’s lack of skill puts a greater burden on the team; and a player’s high level of skill is an even greater boon. In an 8v8 the burden of a poorer player could be better hidden because other players could pick up the slack, and over the course of filling out the teams the teams’ skill levels would tend to balance out. In a 5v5, that’s not going to happen, and the performance of a single player who is significantly better or worse can greatly affect the odds of victory. So attempting to stack on the team with the better player(s) is a lot more tempting and the probability of being rewarded for doing so are greater, because your chances of victory are much higher and there are less players on the other team to try and make up the skill difference. Also, with less players in the 5v5 match, your more likely to notice it happening.
I think the timeout after entering spectator mode is one change that is definitely needs to added. Regardless of why people do it, they are able to do it because they are pretty much guaranteed the spot on the team they want to stack onto. Introducing a timer for re-entry allows another player to take that spot before them, greatly reducing their success rate; though not doing much to tackle the original incentive.
As for disabling spectator mode, on the 8vs8-10vs10 custom arenas, there were generally 2 approaches to balancing the teams.
1. Standard AutoBalance.
2. Disabled SpectatorMode to prevent team switching.
Standard AutoBalance After 2-3 games, players would recognize who were the better players and would attempt to get onto that team at the start of the game. If a player failed to get onto the stacked team, then the SMA may begin. Depending on a variety of factors, this could go on for a long period of time or kill the arena, at which point that may have been the last active 8v8 arena until the next day.
Disabled SpectatorMode This method was a lot more effective at balancing teams than AutoBalance. What would happen is that as players entered the room, they would immediately be assigned to a team and would be unable to swap teams since spectator mode was disabled. The random assignment made the vast majority of games balanced. The downside of this was that, in order to make changes to their build players would have to either leave the room or go AFK (as viewed from other player’s perspective). The AFK timer usually did a good job of keeping the number of AFK players down in this regard by kicking inactive players. The biggest downside of this method was that due to the high velocity of exiting players due to having to make build changes or the AFK timer kicking them, there was a large flow of players exiting the room every few matches. This meant that to keep the semi-chaotic (8v8) nature of the arena up and running, an equal number of players would have to find and enter the custom arena. This requires a near constant inflow of new players that, without the “Play Now” button’s support, meant that the presence of these rooms typically didn’t last for over an hour before they dropped to 4v4 and below level and thus emptied. And of course, randomizing teams doesn’t address the issue of those players who want to play with friends for fun; even if they are the best players in the room, which would be an issue.
I’ve been meaning to ask this for a while, but never got around to it before, so for the folks reading this:
- What is your average roundtrip latency in sPvP, excluding periods of severe lag/disconnects?
- At what latency level do you stop attempting to play sPvP due to latency?
- And if you care to share, what is your average rank in sPvP or tPvP when you play with average/normal latency.
By roundtrip latency, I mean the value that shows up in the Resource Monitor for GW2 (assuming you’re using a Windows OS) that shows how long it takes for a packet of data to go to and come back from the servers.
To answer my own questions:
- My latency is around 250 roughly 80% of the time. Although on rare occassions, such as late at night, it can drop down to around 190.
- I tend to stop attempting to play sPvP once my average latency reaches 270-280 and go to WvW. I stop playing WvW when my latency reaches about 330.
- When I was on the leaderboards (I haven’t done ranked PvP in a while) I hovered around rank 150 in solo queue. I solo queued in team queue and hovered in the 450-550 range.
Note: To see your latency on a Windows machine, press the Start/Windows button in the bottom left and type “Resource Monitor” into the windows search bar, then press enter. Select the network tab at the top, and under the TCP Connections section, look for Gw2.exe. Just in case, Reddit instructions explaining how to check your ping.
Edits: Formatting.
(edited by Roy.7405)
Whether they can make water above a certain level, I don’t know, although I’d like to say anything is possible. But since you seem interested in the mapmaking, I’d recommend you look at:
http://www.twitch.tv/guildwars2/c/3283865
; at least starting at the 4:00 mark. It’s a livestream where Anet talks about how they created the Skyhammer map, and you can get a glimpse of some of the mapmaking tools they used.
I wouldn’t be surprised if Anet has a list of things they want to implement but haven’t gotten around to doing, as some of your suggestions I’ve seen before, but it’s good to see topics like this come up again.
New Buildables:
– Walls/Barricades
– Slow Traps
– Buildable Watchtowers
– Buildable Supply Depot (Allow players to deposit supply, not only withdraw)
– Ramps for high terrain/walls (fragile/low hp)
-Walls/Barricades: I can see issues arising with players placing these in ways that can be exploitive, even with a limit to the number of them in an area and them having low health; such as placing a bunch in front of the portal entrance into a tower. It’s just too risky to implement, as people would constantly find ways to exploit them.
-Slow traps: I like this idea. It could either slow down foes by inflicting cripple, chill, immobilize, or by acting like a guardian’s line of warding. And it can be countered with condition clear, immunity, or stability.
-Buildable Watchtowers: I think Edge of the Mists Sentry Turret consumable would suffice. Admittedly, I haven’t used this item so I don’t know how well it functions.
-Ramps: Anything that can be used to bypass walls shouldn’t be implemented. While it makes sense from a realistic perspective, it’s not very fun to play against; Mesmer portals and Superior Siege Rams are effective enough in this regard.
New Camp Upgrades:
– Dolyak Armor: Equip your Dolyaks with the latest and greatest in combat platemail, allowing them to engage enemies when struck and soak up more damage. (requires guard upgrade first)
– Dolyak Carrot: Increase the speed of your Dolyak, allowing it to run to its location.
I see these as quite reasonable and doable. Alternatively, instead of Dolyak Armor have the dolyak take 20% less damage for each caravan guard alive; that way the guards aren’t completely irrelevant.
New Garrison Upgrades:
– Golem Factory: Upgrade your Garrison with a Golem Factory add-on, allowing for automated Golem creation. Builds 1 Golem every 1 hour, max: 5 Golems. (requires Waypoint Built)
– Runic Portal: Allows portal travel to the Bloodlust Ruins (located between all 5 points) (requires Waypoint built)
-The golem factory, while possible, seems overpowered and it may be hard to track the number made. Although I can see an Asura hammering away at an incomplete golem and dropping it off by the waypoint/supply-depot when finished.
-The Runic Portal would make it too easy for large groups to traverse the map quickly. Also, Anet also probably wants the upgrade panel to be consistent amongst the keeps on all the maps (garrison, hills, bay, EB keeps).
New Map-wide Upgrades: (located in Citadel)
– Upgrade Sentries your world controls into small outposts, consisting of 2 Guards, 2 Archers and 1 Scout (marks nearby targets). Requires Home Borderlands Keeps and Garrison under control.
I could see this as a structure related upgrade (not citadel), as an automatic upgrade if you own a nearby, upgraded structure. Although I think people in general just want the sentries to be stronger in the first place (like EotM).
Zietlogik.6208:New Structures:
– Skritt Hovel (change functionality similar to claimable forces in Eternal Battlegrounds.) Fortifies the location in the NW of the map to be a new claimable location with bonuses.
– Modniir Stronghold (change functionality similar to claimable forces in Eternal Battlegrounds.) Fortifies the location in the NE of the map to be a new claimable location with bonuses.
I don’t know what types of fortifications and bonuses they’d be, but it’s generally agreed upon that these areas are currently a waste of space and can at least be improved. Perhaps they could act as mercenaries, traveling to the ruins and holding the ruins for a server instead of having a player stand there alone; which is quite boring until you are killed by a roaming party. Of course, they probably wouldn’t make it the ruins after fighting at the camps, sentries, yak guards, players, etc., so there spawning location would have to be different.
Something I’d like to see is have the mercenaries on EB be controllable so that they’re more reliable. For example, after you cap the hylek camp, delivering X (like 30) supply allows a player to talk to an NPC that sends them to a nearby location, such as Danelon Camp, Quentin Lake, Klovan, etc to either guard or attack/contest that location. That way the mercenary deployment is less random and more reliable, and won’t just die every few minutes at a camp. There would be a limit of 2 groups that could be sent out, with a 5 minute cooldown after a group has been wiped or deployed.
My questions for the balance team are about their philosophy on build variety in the form of hybrid builds and utility skills. Currently, many professions have builds that are defined by the types of utilities and traits they have selected, with those builds usually opting to go all in for a particular utility type, minus a stun breaker. For example, we have trap rangers, spirit rangers, banner warriors, venom share thieves, shout guardians, etc. who spend their three utility slots on the same type of utility. Some of the utility skills in those builds are effective without trait investment, while others practically require full trait investment in order to make their skills viable/optimal.
The professions that I feel that offer a good variety of hybrid builds via their utilities are mesmers, guardians, and warriors, since many of their utilities don’t require trait investment and can be simply swapped in and out to achieve the desired effect / basic purpose. Contrast that with for example the ranger builds I mentioned earlier. They are not as flexible in utilities and require heavy trait investment before their usage; (almost) no one uses spirits or traps without all of the related traits. My preference is that this is by accident and not by design, and that all utility skills are meant to be viable immediately as soon as they are slotted without having to adjust traits, thus opening the way for more hybrid builds.
Thus, my questions are:
Are (no/some/most/all) utility skills expected to require some level of trait investment before they are used? Or does it depend on the profession and utility type, since traits may be used as a balancing factor and thus some utilities are meant to require more investment before they reach the max possible efficiency?
Are some utilities meant to be staple utilities that nearly every build brings while others are more niche? To provide an extreme example, are you fine with support guardian’s almost always bringing “Stand Your Ground” over “Hallowed Ground” (in PvP at least), or would you prefer Hallowed Ground see more use?
Are you designing with the expectation that, on average, hybrid builds are more, less, or equally as effective/efficient/last-longer in a variety of situations as builds that focus on a particular skill type (at least in sPvP), or are there meant to be diminishing returns on single utility type builds? For example, obviously spirit rangers are likely to be better in large team fights than a hybrid build, but what about 1v1, 2v2, 3v3, 1v2, etc.? Are they meant to perform as well as hybrids as long as they don’t meet their counter builds?
(edited by Roy.7405)
A long, long time ago there was a post in the sPvP forums about scaling up the size of asuras in sPvP that I think got a dev response that they’d look into seeing if it was possible. I don’t know if they determined whether it was possible, but it should be looked into again.
It’s theoretically the best and easiest to implement solution, and could be added game-wide via a “Scale Characters to Minimum Size” toggle in options so that it would affect WvW (since that is PvP too) and PvE (for players who need to see their allies) too. Theoretically, it’s as simple as (pseudocode):
if(scaleCharactersToggle == true && character.race == Asura){
character.scale = Math.Max(character.scale, MINIMUM_SCALE);
}
and is all client-side, not affecting server-load or affecting the position, rotation, skill data, etc of the game engine; just rendering scale. The other plus is that Asura characters maintain their appearance (they’re just slightly larger) so players who want others to notice their character’s appearance still can have others notice it. And finally, it’s completely optional, so if players would rather prefer to see small Asuran characters, they’d just leave the toggle off.
I agree with some of what you said, but this specific point should be their top priority:
Utilities for every class need to be looked at individually … utilities will allow us to have significantly more viable options, balance the utilities first, then
One of the biggest issues holding back most classes are utility skills that only become viable when traited.
Looking at Guardian and Mesmer utility skills, nearly all of them are useful on their own; and they should be the model example for all other profession’s utility skills. Sure, some utility skills may be better than others depending on your build, but those two profession’s skills are all moderately useful regardless of build. The mesmer’s Arcane Thievery may not be that useful to your build if you’re focusing on using primarily glamour skills and traits, but it’s still effective even when untraited and is useful in any Mesmer build. Same goes for the Guardian’s concentration skills; they might not be optimal depending on your build, but they are all still useful when untraited.
Contrast that with Thief and especially Ranger utility skills: most of them need to be traited to be viable. Most thieves are not going to take venom skills without at least 1 venom related trait, and no rangers are going to take “Guard” without the Grandmaster “Nature’s Voice” trait, which removes those skills as a utility option in nearly all builds.
Requiring utility skills to be traited to be viable hurts build options immensely. Utility skills need to be more universally viable without traits.
But no queue on Obsidian Sanctum? I’m not impressed.
I would like to chime in and say MegaServers should be turned off for Malchor’s Leap and kept off in Cursed Shore until the temple events are adjusted; so that capturing traits is actually fun and not a painfully slow wait. I just spent 15 minutes trying to jump between worlds trying to find an uncontested temple / trait-boss via guesting (I was unsuccessful, found out Malchor’s had been megaservered); someone in map chat said they had spent 1.5 hours and still hadn’t found one. Keeping it turned on for Malchor’s Leap could also lead to trolling / hostility if folks attempt to complete the temple defense events.
In general though, those traits needs to be moved to events that occur on a more frequent basis. Getting potentially one shot at Karka per day if you happen to be there for that 10 minute window, waiting around for hours for Lyssa and Grenth temples to become uncontested, Gates of Arah escort scaling issues, hoping (likely to become hostile) people don’t do the temple defense events, none of those things are fun game experiences.
Otherwise, I like the system.
You could also buy the trait in the Heart of the Mists.
Yes, but that doesn’t address the flaws of the trait-capture system when combined with the MegaServers. The capture system was meant to be a fun alternative to simply buying the traits, but their frequency of capture, soon to be affected by MegaServers as well, takes the fun out of the experience and turns it into a painful waiting experience. I liked the skill capture system in GW1 and think it could work in GW2, but there needs to be a guarantee for players to get at least semi-frequent attempts for it to be useful and enjoyable.
For me personally though, I just spent nearly 30 gold unlocking the new sPvP amulets, runes, etc and would rather not spend any more money for traits on my eight professions if there is a potentially more enjoyable way of unlocking the content.
I would like to chime in and say MegaServers should be turned off for Malchor’s Leap and kept off in Cursed Shore until the temple events are adjusted; so that capturing traits is actually fun and not a painfully slow wait. I just spent 15 minutes trying to jump between worlds trying to find an uncontested temple / trait-boss via guesting (I was unsuccessful, found out Malchor’s had been megaservered); someone in map chat said they had spent 1.5 hours and still hadn’t found one. Keeping it turned on for Malchor’s Leap could also lead to trolling / hostility if folks attempt to complete the temple defense events.
In general though, those traits needs to be moved to events that occur on a more frequent basis. Getting potentially one shot at Karka per day if you happen to be there for that 10 minute window, waiting around for hours for Lyssa and Grenth temples to become uncontested, Gates of Arah escort scaling issues, hoping (likely to become hostile) people don’t do the temple defense events, none of those things are fun game experiences.
Otherwise, I like the system.
I assume this is just for PvE/WvW?
(Major) Power
(Minor) Precision
(Minor) Toughness
Do you main Thief? No, but I do play it often when I am in WvW.
When do you feel that pistol whip is unbalanced? In sPvP, when a player is trying to defend a point. Specifically, I feel that it is most imbalanced in 1v1, although this imbalance can pop up in other situations.
Pistol whip requires the stun portion of the skill in order to land most if its hits and high damage, or at least the first few hits. The complete-evasion/invulnerability portion is required to provide the thief some defense while it completes the full channel, specifically in PvE. The thief can also shadowstep/stealth to make it hard to predict when this skill is coming, making it very hard to counter by dodging. All of these together combine to make it extremely difficult to counter with the thief experiencing no risk.
The combination of the stun and damage make this an important skill to move away from (or counter, which you can’t due to evasion). Thus, players would need to use a stun-breaker to break away from it or use stability. Considering that stun-breakers and stability are not equally accessible to all professions/builds and are on long recharges, this results in situations where players have no choice and must simply eat most of the damage frequently because their stunbreakers/stability have already been used.
My suggestion: Would be to rework the evasion portion of the skill and change it from evasion to take X% (50-75ish?) less damage. That way, the attacked player (or an ally) can still counter with a CC/damage or attempt punish the thief for attempting to use the full duration, and the thief still gains a lot of defense.
I believe part of the problem is that there are too many names for these patches/updates, and if you don’t visit the forums often you won’t know what is on the way.
LivingStoryPatch – March 4th.
MaintenancePatch – March 18th, Glory Removal.
WvWSeason2Trigger – (Friday) WvW Tournament Begins March 28th. Not really a patch but still notable.
Feature Patch1 – Not yet determined. After the living story patch.
BalancePatch – Same as FeaturePatch1, just under a different name.
sPvPFeaturePatch – Some time in the distant future.
On a side note, I prefer the old patch preview that had text instead of a video. That way, I knew exactly what was included.
Game Mode:
PvP, but really PvX. I will attempt to use the proposed format.Proposal Overview:
Give stealth to Quick Shot (Ranger Shortbow #3) so that it benefits from Remorseless (Marksmanship-GrandmasterTrait) and to make Remorseless more appealing.Goal:
To make “Remorseless” more appealing to non-longbow rangers.
To add some escape-ability and tactical maneuvering options to the shortbow.
Slightly reduce evasion spam.Remorseless isn’t even good with longbow.
Actually, remorseless isn’t grandmaster worthy trait
I agree that in its current form it’s not grandmaster worthy, and eventually may be moved down to MasterTier. But if it is to remain in the game, giving it some synergies with other skills is necessary, regardless of what tier it’s in.
I also forgot to mention a slight buff to Remorseless.
Proposed Functionality:
In addition to its current effect, your pet gains Opening Strike when it is stealthed.
Purpose: If it is decided that Guard shall keep its same functionality (it still needs improvements to be worthwhile when not traited Nature’s Voice), it will have some synergy.
Game Mode:
PvP, but really PvX. I will attempt to use the proposed format.
Proposal Overview:
Give stealth to Quick Shot (Ranger Shortbow #3) so that it benefits from Remorseless (Marksmanship-GrandmasterTrait) and to make Remorseless more appealing.
Goal:
To make “Remorseless” more appealing to non-longbow rangers.
To add some escape-ability and tactical maneuvering options to the shortbow.
Slightly reduce evasion spam.
Currently, Hunter’s Shot is the only controllable source of stealth that rangers have. This means that outside of kills, which can come infrequently in sPvP and small groups in WvW, this trait that affects only a single target for a short duration rarely activates. This is especially true if the user doesn’t use the longbow. Investing 30 points in a trait that may rarely trigger discourages its use, especially to non-longbow users. Giving stealth to the shortbow would allow non-longbow users to take advantage of this trait as well. Since it is in the Marksmanship line, shortbow users will also benefit from the other Marksmanship traits that have likely been selected.
The stealth would also give the user an opportunity to better position himself for the shortbow’s auto-attack (bleed when hit from the side or behind), instead of being countered easily by the opponent rotating with them as the ranger tries to reposition himself.
Proposed Functionality:
In addition to the current functionality, gain stealth for 3 seconds if the shot hits. Increase recharge to 12 seconds.
Increasing the recharge to 12 seconds reduces the profession’s evasion spammability while balancing out its stealth component. It also allows this stealthing skill to have the same location and recharge on the skillbar as the Longbow’s Hunter’s Shot (which also grants stealth) for consistency.
Associated Risks: Changes primary functionality from spammable evade to tactical stealth.
(edited by Roy.7405)
[WvW / PvP] Proposal: Ranger Spirits, Mandatory Traits, and removing screen clutter
Proposal Overview
By now everyone is familiar with the word ‘Zoo’ and how it relates to the Ranger class. While the screen clutter is still there, the strategic value of all the minions has lessened because of the changes to tab targeting. Body blocking is still a large issue though and I am proposing we remove the spirits from the game.Goal of Proposal
The main advantage to removing the spirits using my idea is it will make the utility skills useful without any trait investment so the class can consider using them regardless of the spec they choose while still allowing players the ability to specialize in spirits if they want. It will also remove the screen clutter and body blocking problems spirits currently have.Proposal Functionality
My suggestion is to remove spirits and instead simply give an ‘aura’ to our pet that does the same thing spirits currently do. Think of it as the spirits ‘possessing’ the animal in some fashion. Spirits will no longer have a duration and will be active 100% of the time until the pet they ‘possess’ dies or is swapped out. Once they ‘die’ they are put on a cooldown for XX seconds before they can be recast. A pet could have one or all of the spirits active at one time on it.When the user activates a spirit’s ability after it has been summoned, a mini spirit will spawn at your pets current location and .5 seconds later it will detonate causing whatever effect the spirits currently do.
If you have the trait to activate spirit abilities on death, this will now activate when the pet dies or is swapped. Like above, a mini spirit will spawn for each active spirit on the pet at the time of its death/swap, and .5 seconds later they will detonate using whatever ability the spirit has. This had the added advantage of the Ranger getting some controllable burst (Storm spirit activated to spawn a mini spirit to explode for AE damage and then you swap pet to summon a second storm spirit at the cost of the XX second cooldown.
The current trait that is no longer needed because spirits will now always move can be changed to reduce the cooldown on death or simply replaced with some other trait.
Associated Risks
The risks of the current spirits remain because you’re sacrificing a utility slot for the spirit. Some of the advantages of the current spirits like their ability to soak AE or body block projectiles is also removed which increases the risk to the Ranger as well. It also puts enormous responsibility on the Ranger to manage their pet properly or risk it dying or needing to be swapped, which would remove all the spirits from play.
Assuming that the summoned spirit actives are destructible (the possibility of a counter should exist), I agree with the majority of this proposal, as I was thinking something similar. One major problem with spirits is that they only become viable when fully traited. Almost no one uses them without all three traits, since each trait adds some functionality that is necessary for it to be viable in whatever game mode you’re in. Obviously with these change values would have to be rebalanced.
The only functional issues with this proposal are Spirit of Nature’s passive and active effect when combined with the Nature’s Vengeance trait. The passive effect (which would be heavily reduced) can be countered by focusing and killing the pet, however triggering spirit actives on pet swap or pet death would make it overpowered (unstoppable AoE ressurection and condition clearing). Nature’s Vengeance would need to be replaced with something else to make this change viable and balanced.
Edit: Also, I forgot to mention that it’s a lot easier to keep track of a single pet than a bunch of spirits among a bunch of AI.
(edited by Roy.7405)