Guild Wars 2

Just to chime in quickly here with something that I seem to recall but am too lazy to sift through pvp changelogs to look for: The matchmaking algorithm used to create teams with as close of an average rating as possible, i.e. pick ten players then sort them so that the resulting average team ratings were as close as possible. Then at some point when the MMR was made visible, if memory serves, the system was tightened up to force teams of roughly equal skill to face other teams so as to better reflect the nature of the game mode, i.e. the algorithm was changed to first pick five players for a team and then match the teams together as closely as possible while maintaining reasonable queue times. Thus far what I remember. Now the addendum:

This however resulted in (some) people complaining that they were getting dragged down by having a random noob thrown into their team … and for making the good players thus get dragged down to the average while the bad players got dragged up to the average. Which is quite the opposite of what the matchmaker should be doing – its main purpose is to create reasonable matches that get you to your skill rating the quickest. Those matches are not always going to be good quality matches by default because unless you’ve played over a hundred games or so, your rating is still fairly volatile in terms of accuracy, and as far as I’m aware the system does not sort together people with a similar number of games played.

In my personal experience after this change went through, the match quality did improve overall – more matches that were not blowout matches and matches that did have swings, which was not something I had been seeing very often prior to this. HOWEVER, this has generally only begun being the case after me having played somewhere between 50 and 75 games in that particular season.

So to conclude, you really can’t make the system suiting everyone. People at the top, if they queue together, by that very virtue of being at the top, will not have enough people within a reasonable skill bracket to queue with/against and thus will be matched against lower-rated players. Doesn’t matter who, that’s just the nature of the top 1% (let’s say it’s 1%). Those matches will either be what we have now, blowout matches that confirm the rating of the top-rated players (while ideally also not giving a big bump to the lost rating of the losing team) or matches where those top players are split and the match is just random chaos until someone identifies a weak link and then a blowout happens, and the weakest link is (possibly) ridiculed over team chat for being so weak and dragging everyone down. Personally? I prefer the system we have now when you can’t really blame any single player on the team for dragging everyone down and if you do end up on the losing side in a serious blowout match, it’s usually not going to cost you too much (barring smurfing and match manipulation, both things which the matchmaker can never account for). And yes, I’ve been on the receiving end of said blowout matches. Despite that my rating has remained stable over the past two weeks with a deviation of roughly 50 points (or about 3 games’ worth of MMR) either way (lucky winstreaks or loss-streaks moving it either way).

So in my opinion, if anything needs changing, it’s not the baseline algorithm, it’s the details of how closely the two teams should be in terms of rating (directly relating to queue times) as well as things not related to the matchmaker directly, namely class balance, map design and game modes available.

Simple fix to what is a deliberately degrading “you – the losing team – suck!” is to be more subtle about it – have the two teams facing each other, with the winners cheering and the losers adopting a sad stance. This forced “these guys are the heroes, those are the corpses” is just not appropriate for when games end 495-500. :|

This looks like a sorting problem. Looking at that particular team matchup I would have to guess that the system first assembled the teams with the aim of keeping the team average rating low-variance (this is a good thing) – and then matched two teams based on their average ratings. In the rating range in question, there may not be many teams close enough in their average ratings so that the likelihood of skewed matches increases a lot. But this ought to go both ways (not quite in equal measure though because of the gaussian distribution of player ratings) – so if you find yourself getting matched against the same higher-rated people a lot, maybe give this whole queue-dodge thing a try?

Perhaps I should expand on the decay bit. What I would expect decay to be, would not be a flat -100/+100 per day (depending on whether the player in question is below or above average), but rather something like exponential decay. This can potentially be made far less aggressive for rankings below the average but I firmly believe it should still be present after a period of inactivity. One could also – while keeping the basic decay formula the same – modify the decay time (in the exponential decay scenario) multiplicatively with a weight factor that would be dependent on the starting rating in question. So for example, around the average rating this factor would be nearly 1 thus resulting in a normalized decay time, it would go down the higher the rating is (thus making the decay kick in and be faster) and it would go up the lower the rating is (thus making the decay kick in and be slower).

One could make the baseline rate of decay be such that over the course of half the season, a given player (with a decay time factor of nearly 1) would drift back to the average rating. Obviously, this is almost never going to be the case, and people lower on the ranking list would have slower decay, while people higher up would have faster decay. From here on out, with a model of exponential decay, it would be merely a question of playing with some sample numbers until a satisfactory model is found.

To those of you saying all the statistical evaluation being easy to address … You do realize that we’re still talking about a computer evaluation, yes? If you are so sure these issues are easy to address, please, educate us (and Arenanet) by providing details of how to do it. Because all I have seen so far is “it should do this and this” but not the details of how it can be accomplished – nor the reasoning why those particular things would be the ones used.

What I’m talking about is – how do you weigh individual contribution based on match statistics? Which ones should be used? Should they be universal or class-dependent? If they are class-dependent, shouldn’t they instead be role-dependent? If they should, how do you automatically determine the role someone is playing without human interference?

Most importantly, can you also outline how such a proposed change would impact the system on a global scale? I mean, sure, I have seen examples of individual changes, but did anyone think of the larger implications? Until all of this has been adequately addressed, I have no reason to believe the current system is inherently bad.

Sure, I would like to not get dragged down during individual matches by having “bad luck” with the team, but as has been said time and time again… Bad luck averages out over a large sample size of games. So does good luck. The matchmaker has no preference for individual accounts to be put on “good teams”, thus, even though you only control roughly 10% of what goes into a match, assuming the other 90% average out to a 50/50 win/loss over time, it’s your 10% that will decide whether you rise or fall in rating. Of course, it might take more than 50 or even 200 games, but if you’re serious about PvP you’re likely to play more than that. In which case there is no issue here.

The one and only change I would like to see to the system as it stands right now, would be for decay to occur towards the average rating instead of towards 0. I mean, the soft MMR reset puts everyone’s rating closer to the average, not towards zero, so it only makes sense that by being inactive the skill is assumed to drift towards that same point.

Showing both teams’ average pre-match rating after the match would be nice indeed to provide the context for a game’s result. It would certainly help evaluate our own performance (“oh of course it was a stomp, the average rating was 200 points higher” – should never happen but you never know) – as well as give a bit more insight into the matchmaker. We’ve already had an indication of individual skill-level post-match before – the badges. Never have I once seen anyone get harassed because of it in the past. And if anyone does – block+report ought to be sufficient. Individual pre-match ratings would be an even better reality check than showing the (geometric?) average rating – “am I going insane or did I really have to carry hard this match?”.

Of course, this won’t address any of the other issues there may have been with the match, such as compositions – but assuming there were no other issues, this would be something that is good to have, and I thus consider good to have regardless of the other issues. More data usually helps.

On that note – it would also be nice to see your current queue MMR range if the queue is taking longer than x time, would help people dodge the skewed matches in the first place and avoid a lot of grief.

To add a thought here – we had to deal with the extreme bitter cold that not even Sons of Svanir can endure unprotected … Meanwhile, in the same general area, Braham Eirsson acts as if he’s not even noticing it’s cold. That more than anything caught my attention – and I think he’s falling much like Svanir himself did. So now we just need a sudden twin sister or something to come up :P

You mean more like, Rifle is a ranged utility kit complementing weapon kits that do the main job of doing damage, yes? Because that’s how lackluster it is. And it doesn’t even give you more utility than a hammer, as has been outlined by Pakkazull a few posts further up. The only thing a rifle has going for it is range, nominally, and three out of five rifle weapon skills require you to be up close and personal to be useful, thus negating the argument of range. At which point, as many people have already said, the rifle is completely outclassed by the hammer – objectively, by any sane objective metric (which have been provided already – more CC on hammer, more utility, more damage, same role essentially).

Obtena.7952:

Having choices fixes bitter too … you don’t have to play the options you don’t like, you play the ones you do. That should make you feel good. Every class has choices people don’t like. They don’t play them, they are happy people.

Anyways, it’s clear Rifle and FT aren’t choices for you, they are for other people. No reason to be angry about that.

“Having a choice” by definition means that any of the presented options are viable. Evidently, right now engineer rifles are subpar in most if not all circumstances, thus making them a non-viable option for players seeking the maximum efficiency, and thus making it an illusion of choice instead of being an actual choice – from this perspective. Now, granted, not everyone may play with the maximum efficiency in mind, but the only other metric, “fun” is highly subjective and ought to be a secondary consideration during game balance (which isn’t to say it shouldn’t be considered, but it should not be the primary factor).

On a grander scale, with regards to your argument that giving people choice is a cheap way to enact procedural balance … I don’t quite agree. This approach works well if all weapons/classes in question have a specific, (more or less) narrow role that they fill. Whether their role is currently desired or not is then up to the players to decide. But making a class/weapon be mediocre at every imaginable role relegates it to the “only play this if you don’t mind being ineffective” role. The best approach I’ve seen to date is a kind of rock/paper/scissors system of counters. So tell me, please, within this context, where does an engineer rifle (and while we’re at it, the flamethrower) fit in? In which situations do they excel over other options that are available?

(On a more on-topic note:) … No, I do not think we’ll be seeing any (significant) balance adjustments with today’s patch, unfortunately. I think the trailer showed those weapons because they fit the narrative more than anything else.

Well, when we first meet him, he keeps chanting the number “97”, or so our character seems to think. Also, Auris’ title is something like “poorly anchored”. Now I had a hunch and checked what the level 97 fractal was – surprise surprise it’s Thaumanova. So I’m thinking maybe Auris was a part of an investigation into fractals and got into an accident in the Thaumanova one?

And otherwise, it seems that each time you do something and come back to him, he has a somewhat different recollection of events to you. At first when you hand him the ??? item he seems to examine it for a while and then presses a button to see what happens. But when you finally break the tier 1 stabilizer and come back to talk to him, he says he doesn’t know what a rift stabilizer is, but is familiar with the tech. This seems to be at odds with the initial interaction – why would he have to fiddle with the item if he were familiar with the tech?

Same problem here. To help narrow it down – I could advertise a party in LFG while I was not a member of one, but could not advertise while in a party.

I welcome this addition with one minor caveat: I’m still holding out hope for updrafts to exist in locations where one might expect them – not forming paths like in Verdant Brink, but just in logical places. No need for many, just a way to get back up into the air in certain places that does not require an on-foot climb. Would be a nice touch.

Josh Davis.6015:

AegisFLCL.7623:

Snip.

I’ve thought through the implications of removing cele from PvP and what would happen to elementalist quite a bit, actually. It’s pretty risky. I think elementalist would need a new 4 stat combo to be viable at roughly the same level (maybe something like 1050pow/1050condi/560vit/560healing).

This just seems like a recipe to make other unintentional (?) impacts, like making an Engineer/Scrapper stronger (?). It feels like the problem needs to be addressed at the class level, not the amulet level, if the issue is indeed with just one class underperforming due to the removal of an amulet.

Another point that came to mind and is very specific: if you make diamond skin a condition cleanse thing, the ele will have ridiculous amounts of condition cleansing readily available in the bunker builds (which will invariably use water/earth). It might perhaps be better to make it a short-duration resistance kind of trait? Like, gain resistance for 1s after being affected by a condition, ICD 5-10s? The Earth trait line flavor is, after all, all about mitigation, not cleansing (same holds true for most Earth-type skills).

Some of the changes sound good on paper, however, what I find missing is (among other things) a reduction in scrapper sustain. There is a problem when a class running an offensive set of stats (marauder) can outbunker a bunker. Overall, I feel this will not address the power creep introduced with HoT – there is more damage being flung around, but there are also many more ways of bunkering down. Both sides need to be looked at in equal measure – and so far it reads to me like it’s mainly the defensive side being looked at (but only on certain classes? Why not across the board?). Most notably, the high amount of hard CC introduced with HoT is something that was not a consideration before – burning 2 stunbreaks and still being stunned/dazed … by one character, is perhaps a bit too much? (Main offenders: Dragonhunter, Druid)

About runes/sigils/amulets: I found PvP to be far more enjoyable and varied before the Celestial Amulet was a thing. It’s a no-brainer for certain builds and/or classes because the drawbacks of not specializing into one thing aren’t big enough. (Example: playing a Tempest/Ele? Run celestial amulet if you want your sustain and eat it too!). So I would propose giving it a very hard look at and perhaps further reducing the amount of stats it provides (personally, I’d like to see it fully removed, but that may not be the best thing). As someone who plays his elementalist a fair bit more than every other class, the celestial amulet has been a source of power like no other since it gives so many benefits at once – and the problem isn’t with the class, the problem is with the amulet here. If I had to choose between going for more damage or more sustain, that’d be an interesting choice – with the cele amulet, I can get a more than reasonable amount of both. I suspect there are other instances where this is also true.

… And here we thought it was just the AR. Good catch on the other stats, can confirm this happened to my group as well. That is one OP jellyfish :P

When previewing armor pieces with the Winter’s Presence shoulder piece already equipped, if I click on “clear preview” the amount of falling flakes generated in the preview window increases with each click. This is not gamebreaking or anything, since simply resetting the preview window by closing and reopening it fixes the issue, but it’s there nonetheless.

Not sure if the last patch was supposed to have fixed this, but it’s still happening.

Same problem in our Arena

Here’s an update: the issue seems to worsen a lot as peak (european) times approach, it worked without a hitch this morning. (Playing on Gunnar’s Hold). The issue is accompanied with the instances having trouble closing and resetting too, since at the times when I had issues getting in, I also had issues with the map not reloading/kicking me when the “match timer” was over.

Having this issue intermittently as well. Either I get a message that the world I’m trying to join is full (this is an instance, right, surely more can be spawned?) or I get the loading screen to Winter Wonderland, which promptly changes back to the DR loading screen – thus effectively blocking entry to the activity. I haven’t tried the other two as extensively, but they seem to also be affected.

I can also confirm the “map is full” message after instance reset (last I remember I didn’t get booted out properly when that happened)

Win 7 64 bit
GW2 64 bit client

Started happening yesterday

Since the first 17th November patch the Guild Portal destinations are blank. Our guild was unable to utilize this upgrade to ease the travel to guild mission destinations.

Furthermore, the Scratch Sentry Defense guild challenge mission had no way to be started. We launched the mission, we got into the instance, and the NPC that would normally trigger it had no option to begin. We tried restarting, we tried re-entering the instance etc. With no effect. We met another guild having the same issue.

Gilded Hollow guild hall issue. Our guild just reached guild level 15 and we would have liked to upgrade the mine, but there’s now a solid rock wall in the way, no way to get to the mine proprietor. It was accessible and fine before the guild level change took place.

Edit: since then, the rock wall has been reappearing and disappearing on guild hall re-entry, sometimes within the same instance. One time it was there for my guildie and not for me, while we were in the same map instance. :-/

This has happened to pretty much every revenant upkeep skill I tried

The invisible mushroom registered for the torch collection, but no other. I’ve since sold it since it was taking up too much space, however, this is probably a bug. Unless of course you need to find a different, yet same, mushroom for all the profession weapon collections :/

This may help in tracking down the issue: I have exactly three fractal weapon skins unlocked – the Axe, the Longbow and the Spear. The achievement however tells me I have obtained the fractal Scepter, Axe and Spear.

Gaining the elite spec rune recipe immediately on the character that finished the spec is slightly counterproductive. As an example, I have a fresh revenant whom I finished getting to max herald today – got the Rune of Herald recipe… but, here’s my question, where did it go? My revenant does not even have a crafting discipline trained :/

This would be much better as an account-bound recipe item. (And maybe refunds to those who experienced this issue?)

The Northwind Collection: I never gained the Machined Warhorn item which you supposedly get for finishing your specialization. I maxed Tempest out before the iteration build went live, never received it then. Logged onto my Daredevil thief and got the equivalent item, which made me recheck the ele. No such luck

In Sign Cutting

Issue: I’m still having the requirement of “train itzel poison lore” as a requirement show up. I had reached this story step prior to the patch that is currently listing that in release notes as “removed”. I tried resetting it via restarting the story step, to no avail.

Edit: restarting story chapter 13, going to auric basin, then starting chapter 14 again and going back into tangled depths seems to solve this issue.

The Alt-F4 thing upon being downed might be easily solved by introducing a simple penalty: anyone crashing/logging out/disconnecting while in a downed state in wvw subtracts a point from the server score :P

I’ll just chime in here and say that completely removing white swords in all cases is a terrible idea – any server that does not have 24/7 coverage for a(ll) map(s) will be forced to spread their already scarce resources even thinner, in many cases meaning people won’t bother to defend at all. However, removing white swords, or at least delaying them by anywhere up to 2 minutes, when a server currently enjoying the “outnumbered” buff would have triggered them, is not that bad of an idea, I think. Conversely, outnumbered defenders should always have the luxury of white swords, perhaps even near-instant ones – although this last bit is debatable given how it’s feasible to generate arbitrarily outnumbered situations for defenders in the mid tiers by leaving just a few people behind to defend and go wreck havoc on another map in the meantime.

Points per kill … I’d rather this stayed as Points-per-Stomp, like Bloodlust. If that’s not an option, then it should at least be tied into the “outnumbered” buff – when outnumbered, you don’t give points for being killed, and/or get double points for kills. But really, stomps are better here, for the same reason why the Bloodlust buff doesn’t give points for “mere” kills but requires actually using a finisher on an opponent.

… That said, the amount of communication is much improved!

As far as I could tell, all you faced in that video were power builds, mostly glassy. How would this measure up against someone specced into conditions and tanking? Just outrange and kill or run?

I used to run a very similar build myself (a while ago) before this buff, so I’m curious how it’s working out nowadays, given the prevalence of condition spam.

Same problem. Wiki outright refuses to load at most times, forum takes a good while to load. (Also some ingame skill lag for no good reason)

This has been happening some time since the feature patch for us, unfortunately I didn’t note the exact point in time it started happening. It’s definitely been consistently the case since the latest patch, however.

I could swear I saw the sun actually rising in Lion’s Arch the last day before it was attacked. It was a very subtle change in night/day, but the darkness clearly was slowly receding – and then when I looked up into the sky I saw the sun disk actually move to its “day” position. It’s a shame it’s not more of a gradual move and proper dusk and dawn with all the lighting that go with that, but saying there’s no day/night transition isn’t entirely true. I do agree that there could be more of that, especially random weather would be pretty epic.

This is awesome, made my morning

1. Ranger
2. Elementalist
3. Necromancer

Third slot was a tough call, between a Thief needing some love on some skills and weapon sets, a Necro needing some balancing on condition output and a Warrior needing some balancing on potential raw damage output (mostly in PvE)

I respectfully disagree – this can be sorted out by having a pet turtle. Turtles sort out everything.

1) Commander and squad functionality
2) Making PPT matter to all who participate (Rewards)
3) Discouraging zerging/Encouraging small(er) coordinating groups

As said above – the potential for people to just troll someone else is too high with such an implementation. Being able to block particular commander tags (for yourself only, mind) would be an infinitely better solution.

L2identify the problem source. The issue isn’t with the skills here, they’re perfectly fine (you know, that blind field, it’s only melee range, just, y’know, use ranged attacks for a bit?); the issue is with the trait that gives 2 init back on each application of stealth. If you want to point to anything being broken, that’s your best bet. Just saying.

… And before anyone says anything about me hating on thieves – I play one, I do my best to not abuse stealth. Just pointing people who want to rage about thieves to the most likely culprit of their raging.

Shoe.5821:

With all due respect, Devon, I believe time dilation “can” be made to work with the GW2 combat system – I seriously doubt there’s no server-side time ticks at all (even if it’s just 10ms or 1ms), that’s just not how programming (in a simulation) works. The question is, how small are those ticks, and how would you present the slowdown of time to the client. As well as what happens with the points acquired. … And the points thing is really a big issue, because I can see no way this won’t get abused. :-/

it’s not a question of tech, it’s a question of sense and gameplay

time dilation doesn’t matter much to gameplay in EVE’s 1Hz deterministic system. but in GW2, noone would land any big hits because you would have all the time in the world to dodge.

it would also add a lot of down time and frustration. imagine someone finishing you off at 1/3 speed, it’s like the game itself is teabagging you

Eh, I’m not saying it would the perfect solution, I’m merely saying it can be made to work – it is questionable, however, how much of a gameplay experience improvement would be felt at a reasonable slowdown (obviously EVE’s max of 10x is not workable in GW2). However, there as here, the idea would be of graceful performance degradation, not to eliminate lag as such. As it stands right now, you don’t even get to use your non-“autoattack” skills much (they do trigger, sometimes, seemingly at random) – I’d much rather my skills would respond at least somewhat reliably and my opponents would get to make decisions there. There’s only so many dodges and anti-CC skills – I would expect a world with something like time dilation to shift to a very CC-AoE-heavy meta in zerg environments (oh wait, it’s already there, no?).

With all due respect, Devon, I believe time dilation “can” be made to work with the GW2 combat system – I seriously doubt there’s no server-side time ticks at all (even if it’s just 10ms or 1ms), that’s just not how programming (in a simulation) works. The question is, how small are those ticks, and how would you present the slowdown of time to the client. As well as what happens with the points acquired. … And the points thing is really a big issue, because I can see no way this won’t get abused. :-/

Seeing how this would
a) Give (more?) guilds (more) incentive to join into WvW
b) Provide actual benefits to defending (you know, other than PPT)
with the only downside I can think of being competition between guilds who gets to claim what … Yyyyyeeaaaaah, I think it’s reasonable to give this a thumbs-up.

Am trying this right now as well and … it does feel kinda strange. However, seeing how the shortbow is more of an “oh snap I need mobility and defense, nao!” weapon (usually, anyway), and I end up in WvW roaming situations, I figured having a different melee weapon set to, y’know, surprise people I meet might be interesting. It is. Very interesting.

New approach to hinting at the content, I think I like it. It’s certainly gotten me curious about what oh-so-predictable interruption will happen this time … sorry, sarcasm went into overdrive again. Will fix it asap.

Is it a solo thief giving you problems?

Class: Elementalist, about 70 tries.
Build: Gear is mix of Celestial and Zerker (mostly zerker). Traits used were, oddly 20/30/0/10/10. Scepter/Dagger. Ether Renewal, Lightning Flash, Mistform, Signet of Air, Useless Elite of the Day (went with Glyph of Elementals). Some cheap +power and +cond dmg food was used.

0/10 on the camera, 3/10 on the fight. Challenges shouldn’t be about avoiding oneshots all the time (and fighting with the camera for a good view of the battle field). Having some oneshots is good, but Liadri’s fight goes way overboard on them for my taste.

I do seem to recall having our golems eaten during a ninja attempt by a single cannon (half a year ago), so yes, this has always been like that. Also, I would think that cannons and mortars are the defensive anti-siege siege weapons (that can be built as upgrades), so yes, this is working as intended. Most of them are in positions that can easily be cleared, so do so before an assault – or risk being annihilated :p