Guild Wars 2

covenn.7165:

GottFaust.5297:

Plato.6128:

As a mathematician, I think the key mistake you are making Mr. GottFaust is not understanding the difference between a conjecture and a theorem. The title of this topic is “The real reasons for the queue bug;” however, you present speculation not proof. If the title were “A theory on the queue bug” I am certain that people here would be more receptive to your ideas.

This is a very valid point. I have made the suggested change. As an engineer, I am very literal-minded. Given the fact that no one can prove, beyond a shadow of doubt, that their theory is more than based on observation until a definitive answer is given from A-Net, I assumed the fact that this is a theory would be obvious, and that the old title would be fine.

Sadly, this is the exact reason why many software engineers are horrible at what they do. I have experienced it first hand in both the public and private sectors. The truth is that designing software well requires a different mindset than the classic engineer stereotype. Of course, some of this is only relevant to the type of software you typically design because the classic mindset does work well enough in some cases.

Anyways, I am a subject matter expert as well and I disagree both with your theories and your reasoning behind them.

I’ll ignore what could be construed as a thinly veiled personal attack in this post, and assume it was meant as a commentary on the industry as a whole.

Care to elaborate as to why you disagree? Dissenting opinions with nothing to back them may well not be posted.

Unfortunately you are ignoring many of the symptoms because you are stuck on your theory. I would suggest taking a step back and taking a look at the symptoms again. Simply put, you are ignoring a large part of the information and not adjusting your theory to compensate for it.

And no, it wasn’t a veiled insult at you directly, but one towards the industry as a whole. This is why I feel that so much bad software gets designed. Even the cutting edge things I am doing with some of the best developers in the country, there are examples of that. The consequences of poorly designed and written code in the gaming industry are relatively minor compared to other areas…. Billions of tax pay dollars and lives aren’t lost when ANET screws up.

covenn.7165:

Unfortunately you are ignoring many of the symptoms because you are stuck on your theory. I would suggest taking a step back and taking a look at the symptoms again. Simply put, you are ignoring a large part of the information and not adjusting your theory to compensate for it.

And no, it wasn’t a veiled insult at you directly, but one towards the industry as a whole. This is why I feel that so much bad software gets designed. Even the cutting edge things I am doing with some of the best developers in the country, there are examples of that. The consequences of poorly designed and written code in the gaming industry are relatively minor compared to other areas…. Billions of tax pay dollars and lives aren’t lost when ANET screws up.

Which legitimate symptoms, pray tell, am I ignoring that have not already been debunked by the A-Net dev post that has been referenced multiple times in this thread so far?

Let’s look at what we can deduce without assuming anything:

• Player for a given server for a given map is in Queue
• Players from the same server actually in that map show out-manned buff

Unless ANet uses the results of ‘culling’ in its determination of when a server has an out-manned buff, then by having any queue AND an out-manned buff would show that either the out-manned calculation AND/OR the queue is bugged.

With the above said I am in the camp that states the queue system is bugged. This OBSERVATION is based upon much the same data that the rest of the population is getting. Seemingly random persons getting into a BL shortly after entering the queue, while others wait hours. Very low observable populations while the out-manned buff and queues are active.

@Gott

Well first of all, you are making a rather large assumption that what he posted is accurate despite the evidence mounting against it. Looks like you need to start completely over in your examination of this issue and remove that assumption.

Start here as it absolutely has not been ruled out and cannot be unless it is your software and you intimately know how it works. Consider the possibility that there may be an issue with how players totals are being reported on maps. Being completely hypothetical, lets say a minor change or rollback to a prior build in a class introduced/reintroduced a bug where disconnected clients weren’t getting cleared from the active player totals correctly. Is it plausible that this same property is being referenced by whatever monitoring code that our responding dev used? Could something like this also explain why sometimes things appear normal and other times they do not, using normal behavior player in how they leave WvWvW zones (exiting game vs. Zoning out) as well as normal functionality like the object being cleared if it was cached (or the property being reset directly from code)?

Yes, I was being completely hypothetical here, but it should be enough to get you started on thinking about this situation a little more clearly.

I think that the OP made a legitimate conjecture. I do not think anyone will dispute that several servers got very stacked, and we are all very aware of the changes that have been made to culling behavior! The question then is whether these two events have sufficiently colored people’s perception into believing that something is amiss where there is no actual problem. Personally, I have no clue as to the truth behind things, but I am leaning toward the camp that something is wrong. After thinking about this a little my conclusion can be summed up as:

1. If there were no problem, the OP’s points while valid would not influence people to the extent that they would think very significant bugs existed (they would just complain).

2. If there existed a “small” bug, the OP’s points in conjuction with this bug could definitely push many people over the threshhold from complaining to thinking that something was wrong.

covenn.7165:

@Gott

Well first of all, you are making a rather large assumption that what he posted is accurate despite the evidence mounting against it. Looks like you need to start completely over in your examination of this issue and remove that assumption.

Start here as it absolutely has not been ruled out and cannot be unless it is your software and you intimately know how it works. Consider the possibility that there may be an issue with how players totals are being reported on maps. Being completely hypothetical, lets say a minor change or rollback to a prior build in a class introduced/reintroduced a bug where disconnected clients weren’t getting cleared from the active player totals correctly. Is it plausible that this same property is being referenced by whatever monitoring code that our responding dev used? Could something like this also explain why sometimes things appear normal and other times they do not, using normal behavior player in how they leave WvWvW zones (exiting game vs. Zoning out) as well as normal functionality like the object being cleared if it was cached (or the property being reset directly from code)?

Yes, I was being completely hypothetical here, but it should be enough to get you started on thinking about this situation a little more clearly.

So, what you are saying is to throw out everything that one of the developers who wrote and maintains the queue code is saying in favor of what is most likely emotionally-exaggerated, due to frustration with the long queues, claims? I’m sorry, but I will not do that.

If you want to be completely hypothetical, there is a more simple solution. Given the operational parameters of the queue supplied by the aforementioned developer, the system is possible to abuse to your advantage. Lets say you are on Server A and you want to dominate Servers B and C hands down. All you have to do is grab a large number of accounts (say 100 for each borderland and battleground on each server you’re facing), transfer them to the target server(s), and have them jump in the queues and then re-queue when they get the popup to enter. This would be easy with a simple bot, and a bit of money. If the bots do get in for whatever reason you can simply have them run around in some far corner forever. At this point you have hundreds of bots taking up legitimate player slots and bogging down the queue thus keeping your competition to a minimum and creating all of the symptoms reported. Again, this is entirely hypothetical and very unlikely.

In general, I find people are terrible at estimating numbers in WvW. I’ve had people say we only have about 5 defending a keep when in reality we have 15. The reason is because of the new culling system where when there are a lot of enemies nearby, you won’t see your friendly dots on the map as it starts to cull them especially if they are spread out.

Another point that nobody seems to mention when talking about the “queue bug” is that the outmanned buff is unreliable. I’ve had it happen often where one person in our party has outmanned buff but everyone else doesn’t. The whole TC incident is becoming some sort of story being passed on without any observable proof and people are just accepting it because of frustration with queue times (is there a video of this incident?).

I maintain that what the developer is saying is true and that the OP has made some reasonable speculations without any observable proof to counter it. If someone has a video of being outmanned while also demonstrating that their area has a queue, by all means post it.

Thanks for changing the title, it’s nice to know that despite having zero clue what you are talking about you have retained the ability to be reasonable. Please make sure to post if you ever actually encounter the queue bug so you can refine your theory with an errant fact or two.

GottFaust.5297:

So, what you are saying is to throw out everything that one of the developers who wrote and maintains the queue code is saying in favor of what is most likely emotionally-exaggerated, due to frustration with the long queues, claims? I’m sorry, but I will not do that.

Your drawing the wrong conclusions. I brought up the same issue he said earlier, from the lack of what the devs posted on the subject.

There is a pretty good chance the DB is not properly clearing offline players, thus the Que thread making DB requests is getting an inaccurate number of open slots. Any dev worth their salt will tell you there’s no such thing as perfect error free code.

Without reading through all of the responses…

Long queues even when getting the Outmanned Buff:

This may be off the mark, but one thing I did notice was that when my wife logged off while in WvW, she was still showing at being there. When she logged in to GW1, she was showing online, but grayed out, in WvW.

Perhaps the server’s “thinking” that everyone with the WvW location is still “physically” in WvW when, in fact, they are either off-line or, in the case of my wife, in GW1.

Hackuuna.4085:

In general, I find people are terrible at estimating numbers in WvW. I’ve had people say we only have about 5 defending a keep when in reality we have 15. The reason is because of the new culling system where when there are a lot of enemies nearby, you won’t see your friendly dots on the map as it starts to cull them especially if they are spread out.

Another point that nobody seems to mention when talking about the “queue bug” is that the outmanned buff is unreliable. I’ve had it happen often where one person in our party has outmanned buff but everyone else doesn’t. The whole TC incident is becoming some sort of story being passed on without any observable proof and people are just accepting it because of frustration with queue times (is there a video of this incident?).

I maintain that what the developer is saying is true and that the OP has made some reasonable speculations without any observable proof to counter it. If someone has a video of being outmanned while also demonstrating that their area has a queue, by all means post it.

Actually the outmanned buff will go on and off, if you are very close to be outmanned. That being said, TC is not the only one experiencing the strange queues and only finding very little people on that map. FA has seen this too and yes we had transfers 3 weeks ago, but no major ones since then and the queuing has only been a biiig issue this week. I hear people blaming in on the culling, but if u look at your mini map u will see the green dots and i can pretty much estimate how many i see there in a second. Believe me our guild is great at estimating numbers and if we say it is 30 then the numbers are between 25 to 35. Yes, and we sometimes laugh at people in map chat when they say OMG there is like 50 here, our guild will check and find normally around 25 to 30 there. But this is different!

We had a Queue and we only owned 1 tower and 1 supply camp believe me there is not many places where players could hide! There were 7 at the camp, around 5 or 6 at spawn and around 12 to 15 at the tower. so thats around 30 players for an entire map with 2 servers attacking with both numbers higher than us.

I don’t think we need to proof that with a video. Don’t u see that EVERYONE is complaining? EU Servers and US servers both have the same issue.

The only way I could see culling affecting this map/queue/lack of bodies thing is if in the code, it’s pre-rendering of enemies has a bug and actually creates two of every player on map … The pre-rendered base model and the actual player.

That said, we do know this started after the Monday patch ……there is a definitive start time for this …… So something is wonky there.

Jayne.9251:

The only way I could see culling affecting this map/queue/lack of bodies thing is if in the code, it’s pre-rendering of enemies has a bug and actually creates two of every player on map … The pre-rendered base model and the actual player.

That said, we do know this started after the Monday patch ……there is a definitive start time for this …… So something is wonky there.

Seems viable enough. Not to mention that, the assertion that a couple dozen servers can’t count and that is the problem, is pretty bold.

GottFaust.5297:

So, what you are saying is to throw out everything that one of the developers who wrote and maintains the queue code is saying

No, what i am saying is to consider the possibility that there may be something he isn’t immediately aware of in play here. Even more so considering like most dev shops, there is probably multiple people modifying source. Also considering the evidence of lack of proper testing that has occurred in the past….

Yeah, you need to open your mind a little bit in order to improve your problem solving skills on issues like this.

Hackuuna.4085:

If someone has a video of being outmanned while also demonstrating that their area has a queue, by all means post it.

Difficult considering it would have to come from two perspective at the same time. You would then have to ‘prove’ that this was the case, which video evidence cannot really provide without itself being susceptible to accusations of forgery.

Jayne.9251:

The only way I could see culling affecting this map/queue/lack of bodies thing is if in the code, it’s pre-rendering of enemies has a bug and actually creates two of every player on map … The pre-rendered base model and the actual player.

That said, we do know this started after the Monday patch ……there is a definitive start time for this …… So something is wonky there.

Without knowing how their code works, that is a possibility. It would be shoddy programming, but I think we have all seen shoddy hacks in the past to get something ‘working’ (using the term loosely heh)

well first it isn’t just TC experiencing this. other servers with various populations have said the same thing. TC did not get enough transfers to account for such a large increase in Q times.

the theory that seems more sound to me, is that the server is still seeing people that are offline as still being online and still in WvW.notice pic in the post below.

https://forum-en.gw2archive.eu/forum/pvp/wuvwuv/Question-regarding-full-wvw-maps/first#post1344455

In the past when someone in your group logged out from the WvW map you would see their group dot on the map. however they would show offline in friends/guild list. Now when they log out the dot is still there, but they are showing online in the friends/guild list.

last patch they introduced a change that allowed players to get back into their dungeon instance if they DCed. The change in the code may be inadvertently effecting WvW.

players that are offline yet the server still showing them online would more than likely also effect a Devs spread sheet or whatever he used when looking at the numbers in the zone.

also you cant just completely ignore peoples observations. Yes counts may be off a little and some people do exaggerate a bit, However believe me when i say many off these people WvW 6 plus hours a day and notice things like large shifts in personnel.

Aggressive culling doesn’t explain anything.

Characters on the other side of the map wouldn’t be visible anyways. Culling does not affect them.

If you’re surrounded by 40 people and half are being culled the only way you would duped into thinking there is no else around would be if you stood completely still and never took a step on your toon. As soon as you move 5 feet, a number of the toons around you will be culled, and a number of others will become visible.

I think the biggest problem with this whole fracas is the fact that A-Net isn’t saying anything about it. It’s really not that hard to see an issue (real or perceived) brought up on the forums and then reply with applicable data. Even something as simple as “we’ve heard your reports of messed up queues and will be looking into it, we’ll update you on what we find next week” is better than this insulting silence we’ve had so far. The very fact that they haven’t addressed most of these concerns is what has convinced many that there is a real problem.

I honestly don’t really care if it’s a bug, an intentional pop-cap nerf, or a few random problems stirring up a whole lot of drama. I just want A-Net to stop treating us like we’re children and TALK TO US. We’re adults (for the most part). We can handle it.

Gondram.8365:

I think the biggest problem with this whole fracas is the fact that A-Net isn’t saying anything about it. It’s really not that hard to see an issue (real or perceived) brought up on the forums and then reply with applicable data. Even something as simple as “we’ve heard your reports of messed up queues and will be looking into it, we’ll update you on what we find next week” is better than this insulting silence we’ve had so far. The very fact that they haven’t addressed most of these concerns is what has convinced many that there is a real problem.

I honestly don’t really care if it’s a bug, an intentional pop-cap nerf, or a few random problems stirring up a whole lot of drama. I just want A-Net to stop treating us like we’re children and TALK TO US. We’re adults (for the most part). We can handle it.

I think this is the post of the day. The silence on the part of ANet really is deafening. Look at all of the posts raising WvW concerns on the front page, not one of them has a response from a Red when they are supposed to be these “community liasons” responsible for posting on the forums and talking to us. Would it be so bad if they took 5 minutes to tell us what’s up? I don’t mean layering facts in cryptic statements either, just telling us what the REAL issue is and that they are trying to fix it. Wild speculation just generates more panic amongst your player base.

GottFaust.5297:

Edit – I forgot to mention that we don’t have queues, and have not had any in months.

Well, apparently, you didn’t play during the time when GoM had queue even during the weekdays.

See someone from your server said that GoM got a queue as well yesterday (and yesterday was weekday already).

Oops, wrong thread! Enjoy the pic anyways.

GottFaust.5297:

Next lets look at the “new” culling system. This one causes the seemingly deserted maps issue. If you remember the way it was described during the initial test, search it out if you don’t, this new system aggressively culls, hides, allied players in favor of showing enemy players. The thought process behind this is that seeing the enemy is more important than seeing your allies. This allows you to get a better count of the enemy strength, works to lessen the impact of tactics that abuse culling like portal bombs, and keeps you from running head-first into enemy zeros that you don’t even see

Thats all well and good except it doesn’t actually achieve this.

Porky.5021:

Aggressive culling doesn’t explain anything.

Characters on the other side of the map wouldn’t be visible anyways. Culling does not affect them.

If you’re surrounded by 40 people and half are being culled the only way you would duped into thinking there is no else around would be if you stood completely still and never took a step on your toon. As soon as you move 5 feet, a number of the toons around you will be culled, and a number of others will become visible.

I’m pretty sure that Gottfaust doesn’t understand that. He also doesn’t understand that culled players would still show on the mini-map if they were there. Basically, he just doesn’t understand how any of this works.

Genev.2450:

“Aggressively culled allied players” does not mean you can’t see people on the minimap anymore, and it only really should happen when you’re in large encounters or very large zergs – not when you’re running around on your own or in very small groups trying to find out where other people are, as people have reported doing.

It should mean exactly that. Culling determines which players the server itself reports to you, the client. If your client is unaware of certain allied players, it doesn’t know they exist. Therefore, it can’t draw a dot on the minimap for these players.

The affinity culling system that they are using now does not aggressively render allies. According to ANet, enemies and allies are on entirely different lists which are independent. Habib has posted on these forums explaining this, and further explicitly stated that the maximum number of allies you can see is half of what it used to be.

As to the OP’s theory, he continually ignores reports of there being queues while a server has the outmanned buff, or of reports of people entering a queue when others ahead of and behind him enter the map with no queue. This should be enough to debunk that culling is responsible, though it could partly be attributed to people counting allies.

When you’re running in zerg, it’s quite easy to spot whether you’ve reached the culling limit as players pop in and out. Based on the ability of your team to defend or attack, you can also estimate whether your numbers are similar to the enemy. I’ve been in lengthy queues, entered a map where it seemed like we had 30 people max, and our performance reflected that. Unless we had 130 people hiding in the jumping puzzle…

covenn.7165:

Hackuuna.4085:

If someone has a video of being outmanned while also demonstrating that their area has a queue, by all means post it.

Difficult considering it would have to come from two perspective at the same time. You would then have to ‘prove’ that this was the case, which video evidence cannot really provide without itself being susceptible to accusations of forgery.

Not really. If you have a queue and you are outmanned, people in your map will be confirming it (“I just came in, there was a queue omfg why are we are outmanned?”). You then click on others to verify (preferably an entire party with outmanned) with video recording and look at the map to confirm location.

You then go through the LA portal and immediately try to transit back to that map. If the queue dialog pops up you have now shown evidence that it’s possible to be at queue/near queue with outmanned buff and the devs will have some explaining to do.

selan.8354:

Hackuuna.4085:

In general, I find people are terrible at estimating numbers in WvW. I’ve had people say we only have about 5 defending a keep when in reality we have 15. The reason is because of the new culling system where when there are a lot of enemies nearby, you won’t see your friendly dots on the map as it starts to cull them especially if they are spread out.

Another point that nobody seems to mention when talking about the “queue bug” is that the outmanned buff is unreliable. I’ve had it happen often where one person in our party has outmanned buff but everyone else doesn’t. The whole TC incident is becoming some sort of story being passed on without any observable proof and people are just accepting it because of frustration with queue times (is there a video of this incident?).

I maintain that what the developer is saying is true and that the OP has made some reasonable speculations without any observable proof to counter it. If someone has a video of being outmanned while also demonstrating that their area has a queue, by all means post it.

Actually the outmanned buff will go on and off, if you are very close to be outmanned. That being said, TC is not the only one experiencing the strange queues and only finding very little people on that map. FA has seen this too and yes we had transfers 3 weeks ago, but no major ones since then and the queuing has only been a biiig issue this week. I hear people blaming in on the culling, but if u look at your mini map u will see the green dots and i can pretty much estimate how many i see there in a second. Believe me our guild is great at estimating numbers and if we say it is 30 then the numbers are between 25 to 35. Yes, and we sometimes laugh at people in map chat when they say OMG there is like 50 here, our guild will check and find normally around 25 to 30 there. But this is different!

We had a Queue and we only owned 1 tower and 1 supply camp believe me there is not many places where players could hide! There were 7 at the camp, around 5 or 6 at spawn and around 12 to 15 at the tower. so thats around 30 players for an entire map with 2 servers attacking with both numbers higher than us.

I don’t think we need to proof that with a video. Don’t u see that EVERYONE is complaining? EU Servers and US servers both have the same issue.

I have had the green dots on the minimap cull and they are no longer a reliable means to me of estimating strength of defense if the defenders are spread out. One example in particular was West Keep defense — I could see a very large mass of FA players on the inner hill catapulting the wall and I felt like I only had about 5 allies with me (using my minimap for estimation). As I was walking from the supply depot to the southern gate to check on a few things I suddenly had old dots disappear while new ones appeared (but I could still roughly see the FA players running around).

I believe this is a new behaviour part of their culling system — it prioritizes enemy locations (not just rendering, but actual locations reported by the server) over allies within a certain radius.

Cactus.2710:

Porky.5021:

Aggressive culling doesn’t explain anything.

Characters on the other side of the map wouldn’t be visible anyways. Culling does not affect them.

If you’re surrounded by 40 people and half are being culled the only way you would duped into thinking there is no else around would be if you stood completely still and never took a step on your toon. As soon as you move 5 feet, a number of the toons around you will be culled, and a number of others will become visible.

I’m pretty sure that Gottfaust doesn’t understand that. He also doesn’t understand that culled players would still show on the mini-map if they were there. Basically, he just doesn’t understand how any of this works.

Culling has two parts: location reporting and rendering. Allied location within a certain radius can be culled if the culling mechanism says that the number of enemies nearby are closer and more worthy of location reporting than the allies.

Why doesn’t the server just report the location of all allies to all players at all times on the map? I guarantee one reason is because it would be a lot harder on the server and on the client (though it obviously wouldn’t be the only reason).

Kimbald.2697:

How can you even remotely present this as ‘how it really is’?

You’re speculating to the extreme, nothing more.

Culling being a reason why people think their server has less capacity? Seriously?

What official message or data are you basing you assumptions on?
I agree many of the speculations about reduced capacity are assumptions without proof too, but you present theories that are just as unlikely and unsupported to be honest…

He quoted the official information. The only changes made to WvW in the patch preceding all the reports of a queue bug mention nothing at all which touches on queues. All the queue behaviour people have been posting in the forums looks like stuff I have been experiencing in WvW for months, everyone heard about swine flu and now they think they have it.

It’s possible that there is some kind of new bug that slipped through, but I personally haven’t read anything that sounds new to me that can’t be explained by queues acting as they always have (people getting in before others who queued longer, having to refresh your queue to get in if you are in it too long, outmanned persisting despite no longer being outmanned, people feeling like they have less numbers than their opponents despite having a queue on the map). I’ve had the game report to me that a map had a queue when trying to get in from an overflow, but once I tried to transfer from a different map in WvW I was easily able to cross over. These are not new problems to WvW.

All of these things were happening in the game before this week. Did you read the threads about loot drops and how everyone was certain the loot tables had been changed and Colin came out and said they investigated it and found that no changes had been made since the November (I think) patch? It’s more likely people are paying attention to the idea of a queue bug, they are experiencing longer queues because of the bandwagon effect (something very real, especially after free transfers) and they “think” something new is happening when in fact, the game is behaving the same way it always has.

Hackuuna.4085:

Culling has two parts: location reporting and rendering. Allied location within a certain radius can be culled if the culling mechanism says that the number of enemies nearby are closer and more worthy of location reporting than the allies.

Enemy and ally culling are handled independently now.

Habib Loew.6239:

This update allows the culling system to handle allies and enemies separately so that being surrounded by a group of allies will not impact the culling of enemies (and vice-versa).

Visiroth.5914:

Hackuuna.4085:

Culling has two parts: location reporting and rendering. Allied location within a certain radius can be culled if the culling mechanism says that the number of enemies nearby are closer and more worthy of location reporting than the allies.

Enemy and ally culling are handled independently now.

Habib Loew.6239:

This update allows the culling system to handle allies and enemies separately so that being surrounded by a group of allies will not impact the culling of enemies (and vice-versa).

I stand corrected. I can’t really explain the allied culling then unless it applies to all the NPCs outside in the West keep defense. It was a fully upgraded keep and there are definitely a lot of NPCs around that supply depot area. Thanks for posting that!

As a vulcan, logic dictates that Anet would comment and close a thread on such a rumor if it was untrue. This is logical since this was what they have done in the past, especially on post that suggest impolite changes without informing their fanbase.

Therefore, until Anet mords comment on such posts, and without basing this on my own personal experience and that of many others that feel the numbers lowered, the rumors are true.

If I was Anet, i would close this thread as soon as possible unless it is true.

GottFaust.5297:

It’s so aggressive, in fact, that it can sometimes cull up to fully half of the allied players you would normally see. This leads, understandably, to the illusion that there are far fewer allied players than there used to be.

Seriously, you think all these people reporting queue problems are standing amidst a crowd of culled, unseen allies and not realising it?

If this were the case we’d be seeing lots of reports of people seeing dozens of green dots on their mini-maps whilst not seeing the actual players themselves. This isn’t happening.

New word from A-Net on the future of Wv3: https://forum-en.gw2archive.eu/forum/pvp/wuvwuv/Upcoming-WvW-Updates/first

Issues 1 and 4 are worded in such a way as to support my theory:

• Issue 1 implies that culling is indeed a problem, and that they are steadily working towards a fix, or fixes.
• Issue 4 implies that the long queues are inherit to large player population and demand through interest in Wv3. Furthermore they are expecting them to get worse with the coming Wv3 enhancements. This also implies that they are looking at ways to make them more efficient and maybe provide some sort of time or position feedback to the player.

GottFaust.5297:

New word from A-Net on the future of Wv3: https://forum-en.gw2archive.eu/forum/pvp/wuvwuv/Upcoming-WvW-Updates/first

Issues 1 and 4 are worded in such a way as to support my theory:

• Issue 1 implies that culling is indeed a problem, and that they are steadily working towards a fix, or fixes.
• Issue 4 implies that the long queues are inherit to large player population and demand through interest in Wv3. Furthermore they are expecting them to get worse with the coming Wv3 enhancements. This also implies that they are looking at ways to make them more efficient and maybe provide some sort of time or position feedback to the player.

No they don’t….they just sidestepped the issue at hand by passing it off/ignore it. Culling isn’t even associated with the queue bug.

I’ve been a software developer for 30 years in many different capacities and I don’t agree with the OP’s theory at all.

nalenb.1425:

I’ve been a software developer for 30 years in many different capacities and I don’t agree with the OP’s theory at all.

I’m sold. This is 100pct absolute proof without a doubt that the OP is completely wrong. Thankyou sir.

nalenb.1425:

I’ve been a software developer for 30 years in many different capacities and I don’t agree with the OP’s theory at all.

I’ll quote myself here: Care to elaborate as to why you disagree? Dissenting opinions with nothing to back them may well not be posted.

If you truely have 30 years of experience developing software you surely know the value and importance of documentation and clear statement of reasons and evidence when disagreeing with a problem assessment.

GottFaust.5297:

nalenb.1425:

I’ve been a software developer for 30 years in many different capacities and I don’t agree with the OP’s theory at all.

I’ll quote myself here: Care to elaborate as to why you disagree? Dissenting opinions with nothing to back them may well not be posted.

If you truely have 30 years of experience developing software you surely know the value and importance of documentation and clear statement of reasons and evidence when disagreeing with a problem assessment.

I was responding to this: “I used this information, some educated observation (I’m a Software Engineer), and some plain old common sense to formulate my OP”

My educated observation is there was little complaints about odd queues and strange populations, then a patch, and now lots of complaints. Years of experience has taught me that however unlikely it seems, there was probably a bug introduced.

nalenb.1425:

GottFaust.5297:

nalenb.1425:

I’ve been a software developer for 30 years in many different capacities and I don’t agree with the OP’s theory at all.

I’ll quote myself here: Care to elaborate as to why you disagree? Dissenting opinions with nothing to back them may well not be posted.

If you truely have 30 years of experience developing software you surely know the value and importance of documentation and clear statement of reasons and evidence when disagreeing with a problem assessment.

I was responding to this: “I used this information, some educated observation (I’m a Software Engineer), and some plain old common sense to formulate my OP”

My educated observation is there was little complaints about odd queues and strange populations, then a patch, and now lots of complaints. Years of experience has taught me that however unlikely it seems, there was probably a bug introduced.

It’s likely related to the new “feature” introduced that allows players that crash to rejoin in the dungeon they were apart of. The system is probably giving them “X” amount of time to get back in before it counts them as logged off completely and unable to get back in the dungeon. Well that system clearly isn’t compatible with WvW now is it? If you get off/crash/DC/AFK, you are off, let the next guy in.

This wasn’t happening before the patch. Now it is here after the patch. Common sense says it was the patch that caused it. HINT HINT ANET!!!!!!!! You being an ostrich on this whole issue isn’t helping you any. A simple acknowledgement that you screwed up and will fix it or you are looking into the issue will shut most of us up. The fact you say nothing only makes it worse. At least pretend you read the forums…

nalenb.1425:

GottFaust.5297:

nalenb.1425:

I’ve been a software developer for 30 years in many different capacities and I don’t agree with the OP’s theory at all.

I’ll quote myself here: Care to elaborate as to why you disagree? Dissenting opinions with nothing to back them may well not be posted.

If you truely have 30 years of experience developing software you surely know the value and importance of documentation and clear statement of reasons and evidence when disagreeing with a problem assessment.

I was responding to this: “I used this information, some educated observation (I’m a Software Engineer), and some plain old common sense to formulate my OP”

My educated observation is there was little complaints about odd queues and strange populations, then a patch, and now lots of complaints. Years of experience has taught me that however unlikely it seems, there was probably a bug introduced.

Yes, there was a patch. The patch introduced the new culling system. On the same day they also cut off the free transfers which caused all the procrastinators and those on the fence to transfer en masse. The theory presented in this thread is not that the issue started on January 28th, 2013, which is obvious, but that the issue is not a bug with the queues or an unannounced decrease in Wv3 population cap. This theory is that the issues reported are a symptom of a couple of, seemingly, unrelated changes causing not only an increase in population, resulting in longer queues, but a lower number of allied players being visible in favor of displaying enemy players, thus causing the illusion of a decrease in Wv3 population.

Assuming that all changes that are viewed by the affected population as negative are bugs is a very ignorant stance to take. Especially when the individuals who have access to the source code and are responsible for its maintenance have stated that everything appears to be working as intended.

Burl.9567:

It’s likely related to the new “feature” introduced that allows players that crash to rejoin in the dungeon they were apart of. The system is probably giving them “X” amount of time to get back in before it counts them as logged off completely and unable to get back in the dungeon. Well that system clearly isn’t compatible with WvW now is it? If you get off/crash/DC/AFK, you are off, let the next guy in.

This wasn’t happening before the patch. Now it is here after the patch. Common sense says it was the patch that caused it. HINT HINT ANET!!!!!!!! You being an ostrich on this whole issue isn’t helping you any. A simple acknowledgement that you screwed up and will fix it or you are looking into the issue will shut most of us up. The fact you say nothing only makes it worse. At least pretend you read the forums…

That capability has been in place for all dungeons except for FotM since launch day. The patch on the 28th simply implemented it for FotM. Wv3 and FotM do not use the same queueing system. Therefore, changes to FotM did not and cannot affect Wv3.

GottFaust.5297:

nalenb.1425:

GottFaust.5297:

nalenb.1425:

I’ve been a software developer for 30 years in many different capacities and I don’t agree with the OP’s theory at all.

I’ll quote myself here: Care to elaborate as to why you disagree? Dissenting opinions with nothing to back them may well not be posted.

If you truely have 30 years of experience developing software you surely know the value and importance of documentation and clear statement of reasons and evidence when disagreeing with a problem assessment.

I was responding to this: “I used this information, some educated observation (I’m a Software Engineer), and some plain old common sense to formulate my OP”

My educated observation is there was little complaints about odd queues and strange populations, then a patch, and now lots of complaints. Years of experience has taught me that however unlikely it seems, there was probably a bug introduced.

Yes, there was a patch. The patch introduced the new culling system. On the same day they also cut off the free transfers which caused all the procrastinators and those on the fence to transfer en masse. The theory presented in this thread is not that the issue started on January 28th, 2013, which is obvious, but that the issue is not a bug with the queues or an unannounced decrease in Wv3 population cap. This theory is that the issues reported are a symptom of a couple of, seemingly, unrelated changes causing not only an increase in population, resulting in longer queues, but a lower number of allied players being visible in favor of displaying enemy players, thus causing the illusion of a decrease in Wv3 population.

Assuming that all changes that are viewed by the affected population as negative are bugs is a very ignorant stance to take. Especially when the individuals who have access to the source code and are responsible for its maintenance have stated that everything appears to be working as intended.

Buddy….we aren’t missing 60-70 invisible players on a map, alright? Stop with the nonsense. They don’t appear on the screen, they don’t appear on the map as green dots. Why?? Because they aren’t there. So when I run around solo and the place is a ghost town, I’m just not seeing them, is that what you are telling me?

When our group of 15 gets run over by 40, I have to wonder why those 60-70 invisible players aren’t pulling their weight.

GottFaust.5297:

nalenb.1425:

GottFaust.5297:

nalenb.1425:

I’ve been a software developer for 30 years in many different capacities and I don’t agree with the OP’s theory at all.

I’ll quote myself here: Care to elaborate as to why you disagree? Dissenting opinions with nothing to back them may well not be posted.

If you truely have 30 years of experience developing software you surely know the value and importance of documentation and clear statement of reasons and evidence when disagreeing with a problem assessment.

I was responding to this: “I used this information, some educated observation (I’m a Software Engineer), and some plain old common sense to formulate my OP”

My educated observation is there was little complaints about odd queues and strange populations, then a patch, and now lots of complaints. Years of experience has taught me that however unlikely it seems, there was probably a bug introduced.

Yes, there was a patch. The patch introduced the new culling system. On the same day they also cut off the free transfers which caused all the procrastinators and those on the fence to transfer en masse. The theory presented in this thread is not that the issue started on January 28th, 2013, which is obvious, but that the issue is not a bug with the queues or an unannounced decrease in Wv3 population cap. This theory is that the issues reported are a symptom of a couple of, seemingly, unrelated changes causing not only an increase in population, resulting in longer queues, but a lower number of allied players being visible in favor of displaying enemy players, thus causing the illusion of a decrease in Wv3 population.

Assuming that all changes that are viewed by the affected population as negative are bugs is a very ignorant stance to take. Especially when the individuals who have access to the source code and are responsible for its maintenance have stated that everything appears to be working as intended.

I know the three servers in our match up are experiencing this problem (tier 3) and none of us have reported any unusual numbers of transfers in. You don’t have any evidence that such last minute transfers happened.

And in any case, ArenaNet could easily query their databases and see. All they’ve said is they will look into it.

The simplest answer is that they introduced a bug.

nalenb.1425:

The simplest answer is that they introduced a bug.

http://en.wikipedia.org/wiki/Occam's_razor

Burl.9567:

nalenb.1425:

The simplest answer is that they introduced a bug.

http://en.wikipedia.org/wiki/Occam's_razor

I live by that every day.

nalenb.1425:

GottFaust.5297:

nalenb.1425:

GottFaust.5297:

nalenb.1425:

I’ve been a software developer for 30 years in many different capacities and I don’t agree with the OP’s theory at all.

I’ll quote myself here: Care to elaborate as to why you disagree? Dissenting opinions with nothing to back them may well not be posted.

If you truely have 30 years of experience developing software you surely know the value and importance of documentation and clear statement of reasons and evidence when disagreeing with a problem assessment.

I was responding to this: “I used this information, some educated observation (I’m a Software Engineer), and some plain old common sense to formulate my OP”

My educated observation is there was little complaints about odd queues and strange populations, then a patch, and now lots of complaints. Years of experience has taught me that however unlikely it seems, there was probably a bug introduced.

Yes, there was a patch. The patch introduced the new culling system. On the same day they also cut off the free transfers which caused all the procrastinators and those on the fence to transfer en masse. The theory presented in this thread is not that the issue started on January 28th, 2013, which is obvious, but that the issue is not a bug with the queues or an unannounced decrease in Wv3 population cap. This theory is that the issues reported are a symptom of a couple of, seemingly, unrelated changes causing not only an increase in population, resulting in longer queues, but a lower number of allied players being visible in favor of displaying enemy players, thus causing the illusion of a decrease in Wv3 population.

Assuming that all changes that are viewed by the affected population as negative are bugs is a very ignorant stance to take. Especially when the individuals who have access to the source code and are responsible for its maintenance have stated that everything appears to be working as intended.

I know the three servers in our match up are experiencing this problem (tier 3) and none of us have reported any unusual numbers of transfers in. You don’t have any evidence that such last minute transfers happened.

And in any case, ArenaNet could easily query their databases and see. All they’ve said is they will look into it.

The simplest answer is that they introduced a bug.

As a software developer you should know by now that the simple answer is rarely the correct one when it comes to systems like this.

As to the server transfer issue: You are aware that they raised the population caps on the servers in general a couple of days before the transfers were cut off specifically because of all the complaints that people couldn’t transfer to servers like Tarnished Coast and Kaineng ( in your tier currently) due to their full status, and that Tarnished Coast received a subsequent volume of transfers so high as to put them back into full status during NA prime time right? You are aware that the only reason Kaineng is even in tier 3 is due to mass transfers to begin with right? You are aware that one of the main reasons they even reset the Wv3 raitings was due to mass transfers greatly changing the Wv3 populations of many servers right?

GottFaust.5297:

nalenb.1425:

GottFaust.5297:

nalenb.1425:

GottFaust.5297:

nalenb.1425:

I’ve been a software developer for 30 years in many different capacities and I don’t agree with the OP’s theory at all.

I’ll quote myself here: Care to elaborate as to why you disagree? Dissenting opinions with nothing to back them may well not be posted.

If you truely have 30 years of experience developing software you surely know the value and importance of documentation and clear statement of reasons and evidence when disagreeing with a problem assessment.

I was responding to this: “I used this information, some educated observation (I’m a Software Engineer), and some plain old common sense to formulate my OP”

My educated observation is there was little complaints about odd queues and strange populations, then a patch, and now lots of complaints. Years of experience has taught me that however unlikely it seems, there was probably a bug introduced.

Yes, there was a patch. The patch introduced the new culling system. On the same day they also cut off the free transfers which caused all the procrastinators and those on the fence to transfer en masse. The theory presented in this thread is not that the issue started on January 28th, 2013, which is obvious, but that the issue is not a bug with the queues or an unannounced decrease in Wv3 population cap. This theory is that the issues reported are a symptom of a couple of, seemingly, unrelated changes causing not only an increase in population, resulting in longer queues, but a lower number of allied players being visible in favor of displaying enemy players, thus causing the illusion of a decrease in Wv3 population.

Assuming that all changes that are viewed by the affected population as negative are bugs is a very ignorant stance to take. Especially when the individuals who have access to the source code and are responsible for its maintenance have stated that everything appears to be working as intended.

I know the three servers in our match up are experiencing this problem (tier 3) and none of us have reported any unusual numbers of transfers in. You don’t have any evidence that such last minute transfers happened.

And in any case, ArenaNet could easily query their databases and see. All they’ve said is they will look into it.

The simplest answer is that they introduced a bug.

As a software developer you should know by now that the simple answer is rarely the correct one when it comes to systems like this.

As to the server transfer issue: You are aware that they raised the population caps on the servers in general a couple of days before the transfers were cut off specifically because of all the complaints that people couldn’t transfer to servers like Tarnished Coast and Kaineng ( in your tier currently) due to their full status, and that Tarnished Coast received a subsequent volume of transfers so high as to put them back into full status during NA prime time right? You are aware that the only reason Kaineng is even in tier 3 is due to mass transfers to begin with right? You are aware that one of the main reasons they even reset the Wv3 raitings was due to mass transfers greatly changing the Wv3 populations of many servers right?

You are aware that most of those transfers were done a week or more in advance? You are aware that most of those individuals were playing already during the match prior? You are aware that they experienced NO issues at all until the patch went live? You are aware that your answer as to why we are all experiencing issues is that Anet addressed some culling issues (actually more they addressed rendering and not culling, they are really separate items), servers got mass transfers, too many people in the JP, too many people farming Skritt, Centaurs, Krait, etc., and that the player base is far to inept as to not recognize that there are 60-70 invisible players on the map causing them to believe there is a population imbalance?

Or the alternative answer is…..they introduced a bug causing some issues with population caps and queues during the last patch.

Sorry…I’m gonna lean towards the bug on this one…

Burl.9567:

GottFaust.5297:

nalenb.1425:

GottFaust.5297:

nalenb.1425:

GottFaust.5297:

nalenb.1425:

I’ve been a software developer for 30 years in many different capacities and I don’t agree with the OP’s theory at all.

I’ll quote myself here: Care to elaborate as to why you disagree? Dissenting opinions with nothing to back them may well not be posted.

If you truely have 30 years of experience developing software you surely know the value and importance of documentation and clear statement of reasons and evidence when disagreeing with a problem assessment.

I was responding to this: “I used this information, some educated observation (I’m a Software Engineer), and some plain old common sense to formulate my OP”

My educated observation is there was little complaints about odd queues and strange populations, then a patch, and now lots of complaints. Years of experience has taught me that however unlikely it seems, there was probably a bug introduced.

Yes, there was a patch. The patch introduced the new culling system. On the same day they also cut off the free transfers which caused all the procrastinators and those on the fence to transfer en masse. The theory presented in this thread is not that the issue started on January 28th, 2013, which is obvious, but that the issue is not a bug with the queues or an unannounced decrease in Wv3 population cap. This theory is that the issues reported are a symptom of a couple of, seemingly, unrelated changes causing not only an increase in population, resulting in longer queues, but a lower number of allied players being visible in favor of displaying enemy players, thus causing the illusion of a decrease in Wv3 population.

Assuming that all changes that are viewed by the affected population as negative are bugs is a very ignorant stance to take. Especially when the individuals who have access to the source code and are responsible for its maintenance have stated that everything appears to be working as intended.

I know the three servers in our match up are experiencing this problem (tier 3) and none of us have reported any unusual numbers of transfers in. You don’t have any evidence that such last minute transfers happened.

And in any case, ArenaNet could easily query their databases and see. All they’ve said is they will look into it.

The simplest answer is that they introduced a bug.

As a software developer you should know by now that the simple answer is rarely the correct one when it comes to systems like this.

As to the server transfer issue: You are aware that they raised the population caps on the servers in general a couple of days before the transfers were cut off specifically because of all the complaints that people couldn’t transfer to servers like Tarnished Coast and Kaineng ( in your tier currently) due to their full status, and that Tarnished Coast received a subsequent volume of transfers so high as to put them back into full status during NA prime time right? You are aware that the only reason Kaineng is even in tier 3 is due to mass transfers to begin with right? You are aware that one of the main reasons they even reset the Wv3 raitings was due to mass transfers greatly changing the Wv3 populations of many servers right?

You are aware that most of those transfers were done a week or more in advance? You are aware that most of those individuals were playing already during the match prior? You are aware that they experienced NO issues at all until the patch went live? You are aware that your answer as to why we are all experiencing issues is that Anet addressed some culling issues (actually more they addressed rendering and not culling, they are really separate items), servers got mass transfers, too many people in the JP, too many people farming Skritt, Centaurs, Krait, etc., and that the player base is far to inept as to not recognize that there are 60-70 invisible players on the map causing them to believe there is a population imbalance?

Or the alternative answer is…..they introduced a bug causing some issues with population caps and queues during the last patch.

Sorry…I’m gonna lean towards the bug on this one…

A few of the statements that you made in this post fly in the face of logic, A-Net’s responses (both text and actions taken), and player reaction and documentation. I won’t bother pointing them out as they should be obvious.

As to the claim that there were no reported queue issues, Kaineng has been reporting them for weeks, yes even before the patch.

GottFaust.5297:

As a software developer you should know by now that the simple answer is rarely the correct one when it comes to systems like this.

Yes. It is less likely they introduced a bug into a large complex system than it is that people are logging in, standing in place with 50 people around them that have been culled and are assuming the map is empty.