Guild Wars 2

You do realize we’re talking about large scale fights where it’s next to impossible to see someone channelling their moves anyway right?

Also it’s not as if endurance regen would be replenishing at the normal rate while everything is else moving slow.

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?).

Time dilation would hurt the pace of the game, not work well with ppt, remove some level of skill from the game, and affect everyone. I would hate it and so would many others I know. We’d quit if it was implemented. Punishing others who didn’t take part in the huge zergfest is always the wrong approach. I wouldn’t even want to be in a time dialated fight. Even if I was part of the zergfest.

One thing I’m curious about, and I apologize if it’s been asked already, is this a strictly server-side issue or can we throw a more powerful cpu at the problem to fix it. Specifically, if I upgraded my 3.4ghz i5 to an i7, would I see marked improvement?

DevonCarver.5370:

We are very, very focused on WvW.

Well i know you keep saying this over and over, but can you tell me why is it that PvE wise we can actually see a lot of development efforts and regarding wvw we see very little ? On the panel Anet was stating that for you the playerbase and what the playerbase wants is important, that you want to “listen” to the playerbase.

Now ask yourself a question. Are you as a company really doing that ? How in the world with all the performance problem´s wvw has in it´s current state can you implement a league system ? Is making favor of lag competition for you ? Has the playerbase asked for such a system ? Why aren´t you focusing on solving the technical problems for providing a solid fundament for upcoming content changes and actively promoting such fixes to the community?

draugris.9872:

DevonCarver.5370:

We are very, very focused on WvW.

Well i know you keep saying this over and over, but can you tell me why is it that PvE wise we can actually see a lot of development efforts and regarding wvw we see very little ? On the panel Anet was stating that for you the playerbase and what the playerbase wants is important, that you want to “listen” to the playerbase.

Kind of tough to make comments like that when you don’t know the internal dynamics of how the ANET dev process works. I’m not going to pretend I know everything that happens there, but from what it sounds like they have separate teams working on PvE content, while unrelated teams work on WvW content. So his statement of “We are very, very focused on WvW” could mean than his team/subordinates are dedicated to WvW, while other dev teams are not.

If ANET operates like any other company looking to stay profitable, I’d imagine there’s a good deal more resources going into PvE content because that’s where the money to be made is. Again, without knowing the true internal dynamics of ANET, this would be the best answer as to why we’re seeing a lot more new content in PvE as opposed to WvW and to a greater extent SPvP.

Grove.2835:

No, DO make more maps. More maps and lower the map cap.

Well that would definitely solve the skill lag problem. Less epic than what it COULD be, but more epic than the lagged solution we have now.

I’d still rather have both. Reduced skill lag so that the current maps would run well even with 200 players clashing, and new maps (preferably the borderlands made unique each).

Deamhan.9538:

timidobserver.7925:

It is a hardware problem. The hardware starts to get pegged out when a bunch of people are using a bunch of skills on each other at the same time in the same place. Assuming that spending hundreds of thousands of dollars upgrading their hardware isn’t a viable solution at the moment, making a new map and creating more capture points is probably next best bet.

Upgrading the hardware isn’t really the answer. Thier servers handle it just fine, it’s the clients computers that can’t.

I’ve never had my fps drop below 10. I’ve talked to people in the exact same fight have their fps drop down to 1.

When you load into a map, you don’t load the entire map. This is why, when you enter into a map that has a massive zerg war going on, you don’t automatically bog down even when facing the direction the zerg battle is at.

Only when you get within a certain range do you begin to lag and this is because only at a certain distance is the zergfest finally loaded on your end.

The tricks they were using before with culling and now particle effects culling is to reduce the amount of information being processed client side.

What I think they should do is to kind of create a local overflow. Think of it like culling on steriods. Instead of both ally and enemy players not rendering, after the zerg gets to a certain size, a local overflow occurs and players from all three sides in the immediate area would enter the overflow.

As seen by the client side, it would appear to be culling, but unlike culling, it’s more than just a lack of rendering.

This is the most ignorant and uniformed post I’ve seen on this topic in a very long time.

Clockradio.3257:

and to a greater extent SPvP.

Yes you are right, the SPvP community is even more screwed than we are. The scene was never really big and people / teams started leaving already. But of course, we do not know the technical problems they have. On the other side , in one year the only technical improvement made was the removement of culling, that´s it. Sorry that´s not enough. I don´t say that everything has to be done in 1 day but as they don´t really speak to the community beside´s the usual “soon, very soon, around the corner” phrases i have the impression that they simply don´t put strong efforts in fixing the issues.

Grove.2835:

No, DO make more maps. More maps and lower the map cap.

I’m getting 60 – 80 FPS on huge zerg fights (Low Settings) however as soon as I start using skills or spamming it, nothing is happening. I do hope that this gets fixed soon. WvW is unplayable at the moment whenever a zerg is nearby.

I just wanted to share with you guys a glimpse of how this works. Maybe it’ll show you how overwhelming an issue it is.

I made this image just now in Paint.NET. It was quick and monochromatic, but it should get the point across.

With only one player in the area, there’s only one way skills can affect anyone. A player can cast a skill on itself (like the guardian’s “Contemplation of Purity”). With two players, Player A can cast a skill on Player B, and Player B can cast on Player A. Player A can also cast skills on himself, as can Player B.

When we add another player, things get a little more complicated. I’m not going to go into all of the details, since they’re easier to illustrate in the picture.

I stopped at 5 players, because beyond that, the interactions will be too many for me to illustrate clearly, and the diagram will begin to hurt my brain. Just know that the numbers get absurdly large when even a few more players come together, and I can’t imagine what a 20v20 fight would look like on paper.

With 5 players in the same vicinity, there are quite a few ways for them to interact. My diagram only shows single-target interactions, but there are many ways in which a skill can target multiple players at once. The server has to compute these interactions for each skill. Think about the guardian’s “Save Yourselves,” for instance. The server has to determine which targets are in range, which ones have conditions to draw, how that adds up for the guardian using the shout, whether any are affected by Pure of Voice, etc.

TL;DR: It’s really far more complicated than you’d expect, and as more people cluster together, the possible skill interactions add up to a ridiculously high number. You want less skill lag? Spread out. QED.

Okay, I made this graph to see what a 20v20 would look like.

I came up with 1560 one-way interactions, but keep in mind that (as stated in the bold text above) this is an oversimplification. There can also be many combinations of these when a skill is used, particularly an AoE skill or one with combo fields. If one were to track all of these (as the server does), then the growth would indeed be exponential (y=2^x – 1)

As an aside, that’s also why we can’t raise the AoE cap (as I understand it).

RoyHarmon.5398:

I just wanted to share with you guys a glimpse of how this works. Maybe it’ll show you how overwhelming an issue it is.

I made this image just now in Paint.NET. It was quick and monochromatic, but it should get the point across.

With only one player in the area, there’s only one way skills can affect anyone. A player can cast a skill on itself (like the guardian’s “Contemplation of Purity”). With two players, Player A can cast a skill on Player B, and Player B can cast on Player A. Player A can also cast skills on himself, as can Player B.

When we add another player, things get a little more complicated. I’m not going to go into all of the details, since they’re easier to illustrate in the picture.

I stopped at 5 players, because beyond that, the interactions will be too many for me to illustrate clearly, and the diagram will begin to hurt my brain. Just know that the numbers get absurdly large when even a few more players come together, and I can’t imagine what a 20v20 fight would look like on paper.

With 5 players in the same vicinity, there are quite a few ways for them to interact. My diagram only shows single-target interactions, but there are many ways in which a skill can target multiple players at once. The server has to compute these interactions for each skill. Think about the guardian’s “Save Yourselves,” for instance. The server has to determine which targets are in range, which ones have conditions to draw, how that adds up for the guardian using the shout, whether any are affected by Pure of Voice, etc.

TL;DR: It’s really far more complicated than you’d expect, and as more people cluster together, the possible skill interactions add up to a ridiculously high number. You want less skill lag? Spread out. QED.

You are of course correct, except that there are way more bits of data that need to be communicated between players than just the skills and the math relationship is quadratic, not exponential. Position, armor/appearance, buffs/conditions, etc. all get passed. The required server-side bandwidth gets huge quickly, but ANet shot themselves in the foot by the way they chose and handled lots of these matters. I really don;t think that the people who were designing content and writing software were well coordinated with the people who engineered the supporting systems.

Oroibahazopi.9074:

Well I’d rather have it being playable and slowed down than lagged out and not having any skills work.

Only thing that changes is you have more time to react, it doesn’t “gum” it up. Right now you can’t react at all in large fights because you can’t activate skills I know which I prefer. Also more reaction time only really nerfs the mindless aoe spam, direct instantaneous attacks will be just as dangerous.

You don’t play much Eve do you?

As an Eve player for the past 6 years, TiDi works for reasons already mentioned by Mr Carter himself. GW2 doesn’t operate under the same combat system or meta that Eve does. If a 10 minute brawl in WvW was stretched out to 2 hours like some of the 10 minute brawls in Eve Online are, I would probably quit WvW and GW2 entirely. Kitten that Kitten, yo.

Cactus.2710:

You are of course correct, except that there are way more bits of data that need to be communicated between players than just the skills and the math relationship is quadratic, not exponential. Position, armor/appearance, buffs/conditions, etc. all get passed. The required server-side bandwidth gets huge quickly, but ANet shot themselves in the foot by the way they chose and handled lots of these matters. I really don;t think that the people who were designing content and writing software were well coordinated with the people who engineered the supporting systems.

So, you’re saying that it increases at a constantly increasing rate, and not an increasingly increasing rate?

I honestly don’t understand how skill lag isn’t a No. 1 concern above everything else to the point that ArenaNet is throwing every resource at the problem. It’s probably the one thing in PvP that would make a casual player quit on the spot. People don’t put up with lag, especially when it causes horrible gameplay. (I understand they’re working on it, but I doubt that it’s a top priority, considering it’s still a huge problem in the game one year later.)

I’m also not sure Devon said time dilation wouldn’t work in Guild Wars 2. I actually agree with that, but time dilation actually does happen in Guild Wars 2 right now. After the server lags for a bit, everyone in the zone appears to gain haste. All attacks fire off faster, even downing people. This happens to everyone. So if time dilation doesn’t work — and I agree it doesn’t — why is that happening?

Lopez.7369:

I honestly don’t understand how skill lag isn’t a No. 1 concern above everything else to the point that ArenaNet is throwing every resource at the problem. It’s probably the one thing in PvP that would make a casual player quit on the spot. People don’t put up with lag, especially when it causes horrible gameplay. (I understand they’re working on it, but I doubt that it’s a top priority, considering it’s still a huge problem in the game one year later.)

I’m also not sure Devon said time dilation wouldn’t work in Guild Wars 2. I actually agree with that, but time dilation actually does happen in Guild Wars 2 right now. After the server lags for a bit, everyone in the zone appears to gain haste. All attacks fire off faster, even downing people. This happens to everyone. So if time dilation doesn’t work — and I agree it doesn’t — why is that happening?

I experience skill lag about 0.001% of my gameplay in WvW so its honestly not a big issue for me. I assume it’s actually the veteran players who are most affected, not new players, because it essentially equalizes the playing field. But I would much rather they put the effort into improving WvW in other ways (like new content/map).

If they wanted to fix skill lag it would probably be something like a reduced AoE cap (but increased damage per AoE target) in large fights beyond a certain threshold. Also, the rare bug you’re referring to is… a bug.

I have an off topic question. Are the new maps going to be introduced today?

Seigfried.5938:

I have an off topic question. Are the new maps going to be introduced today?

No. And it looks like they’re not coming in the next update either.

Tom Gore.4035:

Seigfried.5938:

I have an off topic question. Are the new maps going to be introduced today?

No. And it looks like they’re not coming in the next update either.

Thanks

Seigfried.5938:

Tom Gore.4035:

Seigfried.5938:

I have an off topic question. Are the new maps going to be introduced today?

No. And it looks like they’re not coming in the next update either.

Thanks

Actually now that I looked again at the Anniversary Bash coverage, Devon says that borderlands bloodlust IS actually coming in the 17th of Sep patch… Here’s hoping

DevonCarver.5370:

Oroibahazopi.9074:

Well I’d rather have it being playable and slowed down than lagged out and not having any skills work.

Only thing that changes is you have more time to react, it doesn’t “gum” it up. Right now you can’t react at all in large fights because you can’t activate skills I know which I prefer. Also more reaction time only really nerfs the mindless aoe spam, direct instantaneous attacks will be just as dangerous.

You know which you prefer out of the system as it exists now and some system you’ve made up in your head. Obviously the one you’ve dreamed up is one you like more, but the real implementation of it would have a host of problems you haven’t mentioned. When we have a solution that is better than the current system, we’ll get it in, but all of the things we have tried so far have been no better than the status quo.

All due respect, but I would rather be casting skills at 25-50% rate, and have them actually work.

It would be better than just auto-attacking until we die, which is what happens now. Nothing like pounding on your #6 key and seeing nothing happen, while your HP slow drains away.

Seems like the simplest solution to skill lag is stop rewarding people for balling up into giant lag producing blobs. You could, crazy as this sounds, reward people for being more strategic, tactical and skillful.

ANET created packed server issues by ignoring the obvious impact coverage had on PPT, then they went out of their way to do everything they can to reward blobbing in every aspect of WvW.

Surprise, we have skill lag!

The solution is very easy. Transfer to a low pop server and you wont even know what skill lag is.

Dopamine.7502:

The solution is very easy. Transfer to a low pop server and you wont even know what skill lag is.

True, but only to an extent…18th ranked NA server (of 24 total) and we get it from time to time, much more so on the weekends during heavy playtimes.

Fix skill lag now.

playandchill.3908:

Fix skill lag now.

Why? You left the game until WvW gets lots of fancy stuff. Was that 22 weeks since you’ve logged in, or you played for 22 weeks and then stopped logging in?

Just arrived from WvW, and the game is unplayable in the Desolation server. You see animations freak out, your skills wont work.. This way you cannot defend vs a 60 man zerg in your garrison. Because nothing works.

Adding every 2 week new skins to the game isnt very helpfull too for WvW ( i think ) Doesnt the server has to load even more then?

My advise is to hire someone from NASA. Or maybe someone from the chinese industry i dont know..

But the way it is going now, i have not much hope for WvW.

in 2 weeks League.. League Of Lag.. LOL GW2 style

RoyHarmon.5398:

playandchill.3908:

Fix skill lag now.

Why? You left the game until WvW gets lots of fancy stuff. Was that 22 weeks since you’ve logged in, or you played for 22 weeks and then stopped logging in?

My sig is wrong i left this game 25 weeks ago, i dont even update it.

Anyway Decided to relog tonight to play that new patch and because of this season hype.

Terrible lag, was awfull. I play mesmer and cant use 75% of my skill.

Tbh as long as they dont fix this pure kitten i dont care if WvW has fancy stuff.

Fix this Anet, this is ridiculous.

DevonCarver.5370:

Swamurabi.7890:

Are there different causes for lag server side? Can there be lag derived from map population (WvW or PvE) as well as lag from server population?

Strictly speaking the lag isn’t due to population per se, it is due to concentrated groups of players using skills on each other. The server calculates things every time a skill fires off and the more people it hits the more calculations it does. So if you get enough people fighting against one another, or even fighting with one another, you start to see server CPU problems.

So you basically say you(as a company) released a game with a broken feature which you used to market the game in every occasion. Not even mentioning the graphics engine related lags..

Our keep got attacked on a border by 2x enemies with both around 40 ppl!

Lets get on the siege!!! On the Cannons!!

“Guys we have a problem”

“What is that Lord Magnetron”

“Well the skills from the cannon wont work”

Oww nevermind, they already face bashed thrue the gate..

And we all stand there and look at it, cannot do anything.. it is seriously fighting without a weapon nothing works.

PRO TIP: Do not invest in siege WvW abilities, because they wont work if you need to defend.

This is the daily what T1-3 servers on prime time have to face.

Please do something, hire someone from NASA, Seriously.

EDIT: Left WvW Again, soon i will leave this game. Makes some ppl happy though

Anet’s hardware is inferior.

I think he is saying that the combat is too complex to make use of a feature used in EVE. Clearly he’s never played the game. EVE combat is way more complex than GW2.

It cannot be fixed.

Cut the map cap. Copy and paste another BL in or another EB for the top teir matches. That is all.

Time to beat a dead horse:

I can hit 5 people at a time, 50 people can all hit me at once. Put in a defensive AoE cap to offset the the offensive one, and reduce skill calculations immensely.

King Amadaeus.8619:

Time to beat a dead horse:

I can hit 5 people at a time, 50 people can all hit me at once. Put in a defensive AoE cap to offset the the offensive one, and reduce skill calculations immensely.

I wonder what that would do to world bosses… Or keep lords, for that matter.

RoyHarmon.5398:

King Amadaeus.8619:

Time to beat a dead horse:

I can hit 5 people at a time, 50 people can all hit me at once. Put in a defensive AoE cap to offset the the offensive one, and reduce skill calculations immensely.

I wonder what that would do to world bosses… Or keep lords, for that matter.

There are world bosses in WvW?

Also the cap could easily be applied to players vs players only.

Mora maps less lag!

Just going to run a little example…

Let’s say that we have one server box running all three BLs and EB. That server is over loaded with the amount of information it’s trying to process by all the activity going on in these maps. So it begins to lag.

If you add a new map and then lower the population per map, you will still end up with that one server trying to keep up with the same amount of information.

You would need to add a new server and then divide the maps across the servers or just reduce the population limit of the servers without adding any new maps.

Devon:
“Strictly speaking the lag isn’t due to population per se, it is due to concentrated groups of players using skills on each other. The server calculates things every time a skill fires off and the more people it hits the more calculations it does. So if you get enough people fighting against one another, or even fighting with one another, you start to see server CPU problems.”

I’d like to think and hope that greed wouldn’t overwhelm your higher-ups and that with all the money your company is making (especially off me for example – i think i’ve paid an extra $150 since launch), that your company could upgrade the servers to ones that can handle larger-scaled skirmishes. Not saying it’s a better mmo, but Aion, an older mmo handled more players at once in a full 360-degree space with zergs of enemies casting skills, handling more graphics and effects and with far less lag. Their servers, along with perhaps their programming team has to contribute to their success. I just don’t understand how a newer and very successful mmo hasn’t resolved these issues yet even after a year. I’m sorry, but i feel justified in expecting more. Playing against other players is where it’s at, especially when you’ve conquered the content. I’m grateful for all the content you provide, but the lag and lack of expansion in wvw is just very disappointing.

Make new maps and make them intelligently and you may even alleviate the skill lag problem.

How about a certain percentage of the players just get literally taken out of the equation? Say there are 40vs40 people fighting. Within a certain geographic range, after certain lag thresholds have been exceeded, a number of them are mutually made invulnerable and otherwise unable to interact, ideally even indirectly. Hits of them miss (with a message that it’s due to rationing), hits against them miss, their AoE fields, while still affecting some of your allies, don’t do anything against you and vice versa.
For example, in the situation RoyHarmon visualized, if there were 5 players with each 4 relations for a total of 20 interactions, it would get cut down to 5 players and 3 relations each (one 2), for a total of only 14 interactions.

It would greatly prefer this over the current unplayable state.

anyway, let me finish my map completion first !

From what I have been reading so far the best way this issue can be resolved is to spread people out over the map, so less interactions then occur with each other. Maybe the best solution would be to spawn events where servers are rewarded more for doing so. Events like taking and holding say four locations within a 15 minute time range would reward the server with 500 points to their total score (just throwing numbers about here). Surely then if people want to win they must then spread out, which results in fewer dense battles and so reducing skill lag. putting these event on each map simultaneous could make this solution work even better.

Just disable guardians staff 1 and warriors autoattack from being able to hit more than one target in wvw and immediately 75% of the calculations would not longer be necessary (and they might learn what it’s like to be another class).

Create a GvG map >>> No more guild in WvW >>> no more lag

err… GvG can be held in the JP area right ?

Deamhan.9538:

Just going to run a little example…

Let’s say that we have one server box running all three BLs and EB. That server is over loaded with the amount of information it’s trying to process by all the activity going on in these maps. So it begins to lag.

If you add a new map and then lower the population per map, you will still end up with that one server trying to keep up with the same amount of information.

You would need to add a new server and then divide the maps across the servers or just reduce the population limit of the servers without adding any new maps.

That’s not what will happen. For example 100 vs 100 zerg fight is actually close to an order of magnitude more intensive than 100 individual 1v1s would be (assuming they happened simultaneously). The whole issue is with AoE abilities/boons/healing etc… AoE is usually hitting the target limit in zerg fights while 1v1 is always single target and likewise with small scale fights.

This is compounded by the fact that in zerg v zerg essentially everyone on the map is hitting everyone else all at once while roaming fights are sporadic and spread out over time.

That’s why the solution can’t really be just throwing more server cpu power at it unless they could somehow distribute the load to additional servers just for zerg fights. It would have to be something like reducing AoE limits in big fights or else… De-emphasize huge zerg vs zerg encounters in the first place (what a concept!)

Ducking Frunk.5127:

From what I have been reading so far the best way this issue can be resolved is to spread people out over the map, so less interactions then occur with each other.

People aren’t dumb mathematical constants though. If a commander determines that a massed assault to overwhelm the defending force is what’s needed, he won’t equally distribute his men among the “other objectives”, he will take his entire zerg to the keep/SM because no matter how skilled or bunkered in, once the keep is open 20 people won’t stand a chance against 60. And if he does that, what is to stop the other commanders to call in most of their people on the map as well?
Unless they make the maps much bigger (much bigger than feasible, I’m thinking fifty times bigger) no amount of secondary objectives will resolve megablobbing in centerpiece keeps.