Showing Posts For Habib Loew.6239:

The Basics of WvW guide and more!

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Stickied for greater visibility. Good job getting this together!


ArenaNet Gameplay Programmer

Question to Devs

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Some minor housekeeping, but you should drop the “s” in weeks from the following statement on the WvW section of the website:

“Three huge teams —each representing a server, or world—battle for control over objectives on four massive maps in wars that last for weeks.”

Good catch, thank you! I’ve passed that feedback on to the folks who make words appear on web sites. I don’t know how their update schedule goes but I imagine it will be corrected soonish.


ArenaNet Gameplay Programmer

What is the Loot Bag despawn timer?

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Loot bags despawn after 3 minutes.


ArenaNet Gameplay Programmer

Updated World Ratings (Nov. 23)

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

May we ask why can’t you have this kind of information running on a live website where people could keep checking it?

A live feed is something that we’d like to do eventually, but we’re currently focused on in-game issues.


ArenaNet Gameplay Programmer

Question to Devs

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Currently we feel that one week matches are the right length for WvW and so we don’t have any plans to change the match length at this time. Should the game evolve to a point where two week matches make sense then we will, of course, consider making that change.

I don’t know if or when world transfer cooldowns might change.


ArenaNet Gameplay Programmer

Bugged World Bonuses

in Bugs: Game, Forum, Website

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Just out of curiosity since this bug took so long to find and fix, what was causing the bug to begin with? I’m not looking for any highly technical reason, just being curious in general.

The fixes are related to translating time from one format to another under a very specific set of circumstances and to the parts of the system that are supposed to recover if something goes wrong when translating time from one format to another. It’s a bit like having your breaks and airbags fail at the same time. Either one is bad, but both together… ouch.

I don’t believe we’ve announced our next patch date so I can’t be any more specific than saying that it’s in an upcoming patch. Sorry to be so vague, but rules are rules. It should be coming fairly soon though.


ArenaNet Gameplay Programmer

Perma Stealth thieves are ruining WVW.

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

In an upcoming patch the WvW capture points will be updated to follow the same rules as the PvP capture points with respect to invisibility & invulnerability. As a result invisible & invulnerable players will no longer contest capture points.


ArenaNet Gameplay Programmer

WvW match rollover times

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

When match reset time comes around the game lets the final score tick finish and then spends a bit of time preparing the next match. This process is initiated at 00:00 GMT, but as the last score tick can add up to 15 minutes the reset occurs a bit after that.


ArenaNet Gameplay Programmer

Updated World Ratings (Nov. 23)

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

And as text:

Na
Name Rank Delta Rating Delta Deviation Volatility
Sea of Sorrows 1 2 2127.837 88.359 197.998 0.976
Jade Quarry 2 0 2124.213 -8.267 193.726 0.969
Stormbluff Isle 3 -2 2120.239 -77.749 196.339 0.972
Blackgate 4 0 2021.188 -4.921 196.695 0.97
Sanctum of Rall 5 0 1932.766 7.854 194.096 0.97
Isle of Janthir 6 0 1867.742 -3.012 196.421 0.97
Tarnished Coast 7 0 1794.92 -14.904 194.74 0.969
Fort Aspenwood 8 0 1750.002 -13.201 194.286 0.972
Yak’s Bend 9 0 1715.815 27.865 194.619 0.97
Crystal Desert 10 0 1678.151 4.837 198.673 0.97
Ehmry Bay 11 2 1578.213 129.605 192.14 0.977
Maguuma 12 -1 1533.943 -6.246 194.008 0.971
Northern Shiverpeaks 13 3 1472.12 163.447 193.197 0.983
Darkhaven 14 -2 1468.092 1.897 197.831 0.969
Dragonbrand 15 0 1410.882 -2.896 192.316 0.968
Henge of Denravi 16 -2 1287.479 -128.341 193.021 0.984
Sorrow’s Furnace 17 0 1240.69 -43.554 193.595 0.97
Eredon Terrace 18 0 1160.411 -119.236 192.625 0.985
Borlis Pass 19 0 1160.349 12.341 192.304 0.969
Anvil Rock 20 1 1107.952 21.228 192.485 0.968
Gate of Madness 21 -1 1104.462 -33.69 192.5 0.969
Devona’s Rest 22 0 914.811 13.584 203.686 0.971
Kaineng 23 0 708.928 5.676 197.005 0.969
Ferguson’s Crossing 24 0 673.128 -18.908 198.263 0.97

Eu
Name Rank Delta Rating Delta Deviation Volatility
Blacktide 1 0 2080.882 38.567 195.287 0.976
Vizunah Square 2 0 2007.62 38.356 192.394 0.968
Arborstone 3 2 1909.961 103.867 192.885 0.976
Desolation 4 -1 1831.874 -77.651 194.623 0.976
Seafarer’s Rest 5 2 1778.243 64.469 190.949 0.97
Elona Reach 6 -2 1772.383 -51.736 192.668 0.971
Baruch Bay 7 -1 1736.189 -51.751 192.544 0.97
Miller’s Sound 8 2 1713.516 56.397 190.981 0.97
Riverside 9 -1 1686.744 7.929 190.901 0.967
Kodash 10 1 1630.124 -24.299 190.779 0.967
Far Shiverpeaks 11 -2 1597.3 -72.564 191.088 0.97
Piken Square 12 2 1586.539 51.86 191.295 0.97
Fort Ranik 13 -1 1576.633 -32.285 191.624 0.968
Augury Rock 14 -1 1575.791 -16.731 191.697 0.967
Jade Sea 15 0 1487.887 -35.271 191.48 0.968
Abaddon’s Mouth 16 1 1480.031 19.731 190.982 0.967
Gandara 17 -1 1458.372 -9.666 191.302 0.967
Drakkar Lake 18 2 1418.417 92.736 191.145 0.972
Aurora Glade 19 -1 1387.797 -10.191 191.747 0.968
Gunnar’s Hold 20 -1 1339.367 -22.905 191.041 0.969
Whiteside Ridge 21 2 1280.662 88.771 194.926 0.974
Underworld 22 -1 1254.721 -70.046 191.401 0.972
Ring of Fire 23 -1 1186.075 -76.22 194.662 0.971
Dzagonur 24 0 1147.283 -11.514 193.968 0.968
Ruins of Surmia 25 0 1059.206 63.981 212.428 0.987
Fissure of Woe 26 0 831.381 55.669 206.4 0.975
Vabbi 27 0 650.236 -122.258 210.745 1.004


ArenaNet Gameplay Programmer

Updated World Ratings (Nov. 23)

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Here are the updated world ratings & rankings as of the end of the match on 11/23

Attachments:


ArenaNet Gameplay Programmer

Bugged World Bonuses

in Bugs: Game, Forum, Website

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

This bug (group of bugs really) has been found and fixed. The fix will be available in an upcoming patch.


ArenaNet Gameplay Programmer

Can you fix it (culling) ?

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

I’d like to correct a common misconception that I see popping up regularly in these threads. The culling that exists in Gw2 today is the same as what we shipped with. It was not introduced after the game launched, but was in fact present on day one. I suspect that people didn’t notice it as much in the beginning due to the game being all shiny and new, but it has been there all along.

I would also like to add a personal observation: I understand that the current state of culling is frustrating. It frustrates us too and we’re working hard to fix it. Often in these threads people propose a variety of options for addressing the issue along with the suggestion that their proposed changes are simple. This is a tempting line of thought but in my experience very little of what goes on in game development is actually simple. In particular, the changes required to deal with culling involve fundamental changes to core systems. Both the changes and the systems involved are complex, which is why it’s taking us so long to implement them. As I hope I’ve made clear in my posts fixing the culling issues is a major focus of the WvW team right now and I firmly believe that we will be able to achieve significant improvements when our current development efforts are complete.


ArenaNet Gameplay Programmer

Updated World Ratings (Nov. 16)

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

And as text:

NA
Name Rank Delta Rating Delta Deviation Volatility
Stormbluff Isle 1 0 2197.988 -2.831 195.818 0.969
Jade Quarry 2 0 2132.48 26.788 193.689 0.97
Sea of Sorrows 3 1 2039.478 1.659 198.865 0.976
Blackgate 4 -1 2026.109 -24.361 195.21 0.971
Sanctum of Rall 5 0 1924.912 -0.552 194.853 0.973
Isle of Janthir 6 0 1870.754 -1.099 197.507 0.972
Tarnished Coast 7 0 1809.824 -2.486 195.741 0.971
Fort Aspenwood 8 2 1763.203 108.081 197.884 0.977
Yak’s Bend 9 0 1687.95 19.724 193.638 0.971
Crystal Desert 10 -2 1673.314 -17.295 193.219 0.97
Maguuma 11 2 1540.189 105.508 191.473 0.974
Darkhaven 12 0 1466.195 0.992 195.833 0.969
Ehmry Bay 13 1 1448.608 28.568 192.319 0.969
Henge of Denravi 14 -3 1415.82 -106.424 195.144 0.983
Dragonbrand 15 1 1413.778 4.39 195.34 0.968
Northern Shiverpeaks 16 1 1308.673 -51.394 194.871 0.97
Sorrow’s Furnace 17 1 1284.244 50.002 199.351 0.971
Eredon Terrace 18 -3 1279.647 -135.394 192.521 0.988
Borlis Pass 19 1 1148.008 50.675 193.854 0.971
Gate of Madness 20 -1 1138.152 -45.641 195.036 0.971
Anvil Rock 21 0 1086.724 -5.637 193.527 0.968
Devona’s Rest 22 0 901.227 34.052 199 0.971
Kaineng 23 0 703.252 -44.8 194.759 0.97
Ferguson’s Crossing 24 0 692.036 12.579 197.537 0.971

EU
Name Rank Delta Rating Delta Deviation Volatility
Blacktide 1 2 2042.315 134.668 195.21 0.983
Vizunah Square 2 0 1969.264 -14.406 192.207 0.968
Desolation 3 -2 1909.525 -117.156 193.637 0.978
Elona Reach 4 1 1824.119 22.646 195.21 0.972
Arborstone 5 -1 1806.094 -82.23 195.631 0.973
Baruch Bay 6 0 1787.94 59.157 194.889 0.97
Seafarer’s Rest 7 2 1713.774 36.651 190.693 0.968
Riverside 8 -1 1678.815 -9.648 191.815 0.968
Far Shiverpeaks 9 1 1669.864 6.642 191.614 0.969
Miller’s Sound 10 -2 1657.119 -27.355 191.072 0.969
Kodash 11 0 1654.423 -3.704 190.897 0.967
Fort Ranik 12 1 1608.918 23.779 192.846 0.968
Augury Rock 13 -1 1592.522 -2.941 191.415 0.967
Piken Square 14 2 1534.679 54.313 191.562 0.971
Jade Sea 15 0 1523.158 25.376 192.025 0.968
Gandara 16 -2 1468.038 -48.744 191.621 0.968
Abaddon’s Mouth 17 1 1460.3 23.338 190.915 0.968
Aurora Glade 18 1 1397.988 30.677 191.451 0.968
Gunnar’s Hold 19 -2 1362.272 -77.382 191.086 0.971
Drakkar Lake 20 1 1325.681 10.74 191.63 0.968
Underworld 21 -1 1324.767 -42.204 192.594 0.972
Ring of Fire 22 0 1262.295 -12.222 195.065 0.968
Whiteside Ridge 23 0 1191.891 16.388 200.309 0.972
Dzagonur 24 0 1158.797 -2.002 194.819 0.969
Ruins of Surmia 25 1 995.225 227.122 210.927 0.998
Fissure of Woe 26 1 775.712 51.167 212.206 0.976
Vabbi 27 -2 772.494 -306.59 223.33 1.026


ArenaNet Gameplay Programmer

Updated World Ratings (Nov. 16)

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Here are the world ratings and rankings as of the matchup ending on Nov. 16th.

Attachments:


ArenaNet Gameplay Programmer

Can you fix it (culling) ?

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

tl;dr: We are working on a fix (a collection of fixes, really) for culling and character loading issues. The fixes require significant changes to a number of game systems and thus require time to implement. We’re not yet ready to discuss a release date for the current set of fixes, but we’re working hard to improve the experience as quickly as we can.

Now for the long version:

We currently use server-side report culling to limit the number of characters that any given game client is aware of. By limiting the number of characters that we report to any given client we also limit the bandwidth used (by the server and the client) and avoid situations where the client is overwhelmed by the number of characters that need to be processed and rendered. While this system has some obvious advantages, and it works well in PvE, the large battles that are the signature of WvW tend to highlight the deficiencies of this approach.

There are also some client-side issues which have contributed to the perception of how our culling system works. Once a character is reported to a given client there’s a non-zero amount of time required to load and initially display the assets associated with that character. Extra load time varies depending on how beefy the client machine is (those with more memory, faster CPUs, more CPU cores, and faster drives experience shorter load times). One of our engine programmers recently completed an optimization pass on the character loading process and so we should be seeing improvements to that part of the issue very soon. Even so, the bulk of the issue remains with the server-side culling as it doesn’t matter how fast your client can load and draw a character if it hasn’t even been told that character should exist yet.

As you may have heard we already have a fix for the server-side culling implemented for sPvP. Because sPvP has dramatically less players we were able deploy our fix immediately without worrying about downstream side effects. WvW, however, operates at a much larger scale than sPvP and so we have a number of additional hurdles to clear before we can turn on the server-side fix. In order to address the culling issue we need to ensure that clients, including min-spec clients, are able to handle rendering and processing many more characters. We also need to ensure that the bandwidth needed by any given client remains reasonable and falls within our min-spec for connectivity. The WvW team is working to address both the bandwidth and the client performance issues even now. The changes that we’re making are complex and have a large impact on the way the game engine works. Because of the level of complexity involved, and the core systems that are impacted, these fixes take time to implement correctly. As such, I can’t give you a date when we’ll be done.

At the end of the day our goal is to dramatically improve the experience of large battles in WvW and provide a substantial increase to the number of players that can be seen by any given client.


ArenaNet Gameplay Programmer

PVP Changes in WvW

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Currently WvW uses the same skills as PvE, so PvP balance changes do not have an impact on WvW.


ArenaNet Gameplay Programmer

WvW Rewards

in WvW

Posted by: Habib Loew.6239

Habib Loew.6239

Gameplay Programmer

Next

I can’t say too much just yet, but we’re doing some work related to rewards and character progression in WvW even now. It will be a little while before we’re ready to talk about more details though. Rewards, progression, and incentives for play are hot topics around the office. We all care about this stuff and once we have some more details hammered out we’ll let you all know what we’re up to.


ArenaNet Gameplay Programmer

Updated World Ratings (Nov. 9)

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

And as text:

NA:

Name Rank Delta Rating Delta Deviation Volatility
Stormbluff Isle 1 0 2200.819 45.443 194.14 0.97
Jade Quarry 2 0 2105.692 -2.232 194.784 0.972
Blackgate 3 0 2050.47 -43.307 194.311 0.974
Sea of Sorrows 4 0 2037.819 17.785 198.441 0.984
Sanctum of Rall 5 0 1925.464 14.191 195.09 0.979
Isle of Janthir 6 0 1871.853 -32.987 198.84 0.977
Tarnished Coast 7 0 1812.31 96.844 192.795 0.974
Crystal Desert 8 1 1690.609 48.644 192.942 0.973
Yak’s Bend 9 2 1668.226 94.496 191.748 0.975
Fort Aspenwood 10 0 1655.122 47.306 191.851 0.971
Henge of Denravi 11 -3 1522.244 -149.767 194.729 0.987
Darkhaven 12 1 1465.203 -21.416 192.475 0.969
Maguuma 13 1 1434.681 22.91 191.616 0.969
Ehmry Bay 14 3 1420.04 29.959 196.07 0.97
Eredon Terrace 15 -3 1415.041 -144.386 193.213 0.989
Dragonbrand 16 -1 1409.388 -1.706 191.88 0.969
Northern Shiverpeaks 17 -1 1360.067 -32.521 195.846 0.969
Sorrow’s Furnace 18 0 1234.242 2.83 199.312 0.97
Gate of Madness 19 0 1183.793 8.08 196.785 0.972
Borlis Pass 20 0 1097.333 -10.948 196.393 0.972
Anvil Rock 21 0 1092.361 2.746 195.231 0.969
Devona’s Rest 22 0 867.175 32.415 196.908 0.972
Kaineng 23 0 748.052 3.418 194.796 0.97
Ferguson’s Crossing 24 0 679.457 -35.689 196.218 0.973

EU:

Name Rank Delta Rating Delta Deviation Volatility
Desolation 1 1 2026.681 70.034 192.157 0.974
Vizunah Square 2 -1 1983.67 -18.038 192.468 0.969
Blacktide 3 2 1907.647 143.159 195.17 0.98
Arborstone 4 -1 1888.324 -52.04 192.123 0.971
Elona Reach 5 -1 1801.473 -104.987 199.905 0.976
Baruch Bay 6 3 1728.783 55.784 191.061 0.969
Riverside 7 -1 1688.463 -43.996 194.979 0.97
Miller’s Sound 8 0 1684.474 6.652 191.957 0.971
Seafarer’s Rest 9 3 1677.123 47.384 190.646 0.968
Far Shiverpeaks 10 -3 1663.222 -64.049 192.787 0.971
Kodash 11 -1 1658.127 -0.372 190.659 0.968
Augury Rock 12 -1 1595.463 -46.843 190.391 0.968
Fort Ranik 13 0 1585.139 -17.994 193.141 0.969
Gandara 14 0 1516.782 15.293 191.91 0.968
Jade Sea 15 0 1497.782 2.585 192.212 0.968
Piken Square 16 2 1480.366 78.544 192.379 0.972
Gunnar’s Hold 17 -1 1439.654 -41.317 191.437 0.969
Abaddon’s Mouth 18 -1 1436.962 -36.339 191.31 0.968
Aurora Glade 19 0 1367.311 8.489 191.925 0.969
Underworld 20 2 1366.971 111.884 195.717 0.976
Drakkar Lake 21 0 1314.941 18.226 191.438 0.969
Ring of Fire 22 -2 1274.517 -26.661 191.38 0.968
Whiteside Ridge 23 2 1175.503 54.31 217.102 0.975
Dzagonur 24 -1 1160.799 -42.641 194.286 0.97
Vabbi 25 -1 1079.084 -76.573 204.106 0.976
Ruins of Surmia 26 1 768.103 14.785 211.632 0.975
Fissure of Woe 27 -1 724.545 -66.234 212.406 0.977


ArenaNet Gameplay Programmer

Updated World Ratings (Nov. 9)

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Here are the new ratings as of November 9th.

Attachments:


ArenaNet Gameplay Programmer

The math behind WvW ratings

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Just a quick clarification. The formulas

wAPercent = wAScore / (wAScore + wBScore)
wAGlickoScore = (sin((wAPercent – 0.5) * Pi) + 1) * 0.5

make no assumptions about winners or losers. The first simply calculates a relative score percentage between the two teams involved. This is a number between 0.0 and 1.0. The second transforms the relative percentage into a Glicko 2 score using a sine wave which has the effect of making large score differentials have a less than linear impact on the Glicko 2 score (see the linked graph). The result is still in the range 0.0 to 1.0. Effectively this means that as score differentials get larger and larger the actual Glicko 2 score for a team approaches 0.0 or 1.0 more and more slowly.

We perform these operations pairwise on all the teams, so we are calculating 6 matches (two for each team) which fits the Glicko 2’s one-way nature. This is because the rating changes are asymmetric.

Because the ratings of two teams matter when calculating the results of a match (a lower rated team beating a higher rated team results in more change than a higher rated team beating a lower rated team) if two teams are very close in score then there is generally very little change in their ratings. Again, I encourage you to read the Glicko 2 website for a more thorough explanation of that part of the process.


ArenaNet Gameplay Programmer

The math behind WvW ratings

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

People often ask how WvW rankings are determined so in this post I will outline exactly the system that we use.

The short, short summary is that after each battle we use the score differential between the worlds that fought each other, along with their previous ratings, to calculate new ratings. Once ratings have been calculated we re-sort the list of worlds and form new groupings of 3 for the next battle.

Far more details regarding the ratings calculation for the mathematically inclined:

WvW world ratings are calculated using the Glicko 2 rating system. Full details regarding Glicko 2 are available at http://www.glicko.net/glicko.html. How an algorithm is applied is often nearly as important as the algorithm itself, so here are the details of exactly how we use Glicko 2 in the context of each 3 world matchup. My explanation of these details assumes basic familiarity with the Glicko 2 algorithm.

Assume the worlds are wA, wB, and wC. In order to handle a 3 way battle we treat each match as having two battles for each world. So when calculating the new rating for wA the two battles are wA vs. wB and wA vs. wC. Likewise for wB and wC. Naturally, we do all the calculations before updating any of the ratings. Note that this usage is supported by Glicko 2.

We use a Tau of 0.6 and a k of 1.0

In standard Glicko 2 scores are represented as 0.0 for a loss, 0.5 for a draw, and 1.0 for a win. We use a slightly modified version that takes into account score differential. I’ll explain how using the wA vs. wB battle as an example. To calculate the Glicko 2 score for wA in the wA vs. wB matchup we do the following:

wAPercent = wAScore / (wAScore + wBScore)
wAGlickoScore = (sin((wAPercent – 0.5) * Pi) + 1) * 0.5

where wAScore and wBScore are the raw scores from the end of the match.

That last transform is easiest to visualize as a graph .

We perform the same score calculation for each world and then plug those results into Glicko 2. This means that ratings change over time as a result of battle outcomes and that the rating for a given server reflects the history of that server’s performance.


ArenaNet Gameplay Programmer

Merge the EU/US Servers

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

The North American and European data centers are different in one important respect: they are located on different continents. North American players connecting to the NA data center (and European Union players connecting to the EU data center) will generally experience lower latency and a higher likelihood of playing with larger groups of other players, as those in the same data center tend to operate during similar times of the day. So there are real distinctions between the data centers which their EU/NA affiliations make clear and for that reason we will not be removing their continent designations.

We don’t match up worlds from multiple data centers for similar reasons. Ultimately, the server that runs a WvW map must live somewhere in the world and the players who connect from that same continent will have a distinct advantage over those connecting from another continent due to lower latency. In order to keep things as fair as possible to all involved we keep the matchups within each data center.

Of course it is always possible for an EU player to choose to play on an NA server, or vice-versa, but in doing so that player is choosing to take on the burden of additional latency. That situation is vastly different from our matching system placing an entire team at a latency disadvantage without their knowledge or consent.

tl;dr: data centers are on different continents, latency is an issue with inter-continental connections, data center distinctions are here to stay.


ArenaNet Gameplay Programmer

(edited by Habib Loew.6239)

Thank you for WvW

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Hi all,

First and foremost: Thank you for the positive feedback! We work hard to make WvW fun for you all and it’s very gratifying to hear that you’re enjoying it.

Now I need to make a bit of a clarification about culling being fixed. I think you’re referring to this post in the PvP forum: https://forum-en.gw2archive.eu/forum/pvp/pvp/Please-fix-this-sooner-rather-than-later-Vid-attached/first#post622762

We have a fix for culling in Structured PvP but that fix does not yet extend to WvW or PvE. The issue is that Structured PvP uses small maps with small player counts so we have fewer challenges to overcome with regards to resources like cpu and bandwidth. We are doing the work to extend the culling fix out to WvW but that change will not be in the next patch.


ArenaNet Gameplay Programmer

Please fix this sooner rather than later - Vid attached

in PvP

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Ya, it was a visual culling bug. It will be fixed next patch.

Cheers.

I just want to clarify this post: we have a fix for culling issues in Structured PvP but that fix does not yet extend to other game types.


ArenaNet Gameplay Programmer

WvW match rollover times

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

This is just a reminder that WvW matches roll over at 00:00 UTC and are thus not subject to daylight savings time. So, if you’re in a region that participates in daylight savings time then the matches will now roll over one hour earlier.


ArenaNet Gameplay Programmer

Updated World Ratings (Nov. 2)

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Nymphy, we use the Glicko 2 rating system which is explained fully here: http://www.glicko.net/glicko.html

Roughly you can think of deviation as a measure of the margin of error and volatility as a measure of the system’s confidence in the assigned rating (lower volatility => more confidence). That’s a massive oversimplification though, so please check out the Glicko web site for the real details.


ArenaNet Gameplay Programmer

Updated World Ratings (Nov. 2)

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Here are the Na results as text:

Name Rank Delta Rating Delta Deviation Volatility
Stormbluff Isle 1 1 2155.376 17.28 196.73 0.971
Jade Quarry 2 -1 2107.924 -70.629 200.628 0.978
Blackgate 3 0 2093.777 48.887 197.611 0.98
Sea of Sorrows 4 2 2020.034 228.238 199.846 1
Sanctum of Rall 5 2 1911.273 171.98 194.186 0.991
Isle of Janthir 6 -2 1904.84 -116.866 208.716 0.984
Tarnished Coast 7 1 1715.466 -11.547 192.287 0.971
Henge of Denravi 8 -3 1672.011 -124.326 199.786 0.985
Crystal Desert 9 1 1641.965 37.683 193.617 0.976
Fort Aspenwood 10 3 1607.816 88.837 192.46 0.973
Yak’s Bend 11 0 1573.73 18.066 192.409 0.973
Eredon Terrace 12 -3 1559.427 -162.188 194.93 0.989
Darkhaven 13 -1 1486.619 -55.047 192.187 0.971
Maguuma 14 0 1411.771 -62.997 191.715 0.97
Dragonbrand 15 0 1411.094 -25.427 192.695 0.971
Northern Shiverpeaks 16 0 1392.588 -24.429 195.842 0.969
Ehmry Bay 17 0 1390.081 54.242 196.852 0.971
Sorrow’s Furnace 18 0 1231.412 -28.54 196.949 0.97
Gate of Madness 19 0 1175.713 -75.491 201.596 0.977
Borlis Pass 20 1 1108.281 87.815 203.77 0.976
Anvil Rock 21 -1 1089.615 -7.157 198.365 0.97
Devona’s Rest 22 1 834.76 63.103 198.867 0.976
Kaineng 23 1 744.634 -2.525 197.698 0.972
Ferguson’s Crossing 24 -2 715.146 -61.396 199.675 0.977

Here are the Eu results as text:

Name Rank Delta Rating Delta Deviation Volatility
Vizunah Square 1 1 2001.708 11.377 193.991 0.97
Desolation 2 1 1956.647 41.605 193.529 0.976
Arborstone 3 -2 1940.364 -52.88 193.083 0.972
Elona Reach 4 0 1906.46 -2.197 198.926 0.972
Blacktide 5 1 1764.488 26.804 194.53 0.969
Riverside 6 1 1732.459 55.875 191.462 0.971
Far Shiverpeaks 7 -2 1727.271 -25.283 195.32 0.971
Miller’s Sound 8 5 1677.822 59.381 193.496 0.976
Baruch Bay 9 2 1672.999 30.848 190.273 0.967
Kodash 10 -2 1658.499 -14.269 191.257 0.969
Augury Rock 11 1 1642.306 12.599 190.428 0.968
Seafarer’s Rest 12 -3 1629.739 -41.368 191.112 0.968
Fort Ranik 13 -3 1603.133 -43.499 190.441 0.97
Gandara 14 0 1501.489 -38.636 191.439 0.968
Jade Sea 15 0 1495.197 -19.553 191.928 0.969
Gunnar’s Hold 16 0 1480.971 21.371 191.143 0.97
Abaddon’s Mouth 17 0 1473.301 16.5 191.339 0.969
Piken Square 18 2 1401.822 83.148 191.693 0.971
Aurora Glade 19 -1 1358.822 -38.246 192.122 0.972
Ring of Fire 20 -1 1301.178 -62.348 192.027 0.97
Drakkar Lake 21 0 1296.715 -21.144 191.944 0.971
Underworld 22 0 1255.087 -15.98 194.505 0.97
Dzagonur 23 0 1203.44 42.381 193.257 0.97
Vabbi 24 1 1155.657 23.693 227.617 0.979
Whiteside Ridge 25 -1 1121.193 -26.447 193.413 0.968
Fissure of Woe 26 1 790.779 73.54 217.567 0.977
Ruins of Surmia 27 -1 753.318 -86.825 208.731 0.976


ArenaNet Gameplay Programmer

Updated World Ratings (Nov. 2)

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Here are the current ratings after the match that ended on Nov. 2

Enjoy!

Attachments:


ArenaNet Gameplay Programmer

Orbs will be removed from WvW in an upcoming build.

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

After a careful reevaluation of orbs of power we have decided to remove them from WvW in an upcoming build. As implemented, orbs tend to strengthen teams who are already winning and make it even more difficult for underdog teams to fight back. In addition, the current implementation seems to be irresistible to hackers/cheaters and will require significant modification to prevent cheaters from having an unfortunately large impact on the state of any given WvW game. Under the circumstances we believe that removing orbs completely is a better choice than attempting an in-place redesign/re-implementation as it will immediately put a stop to all orb hacking. It is likely that orbs, or some orb-like mechanic, will return at some point in the future, but only after we are confident that they will not exhibit the sorts of issues that we see with orbs today.


ArenaNet Gameplay Programmer

World Ratings update for October 12

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Zemi, this thread explains how the ratings are calculated in some detail: https://forum-en.gw2archive.eu/forum/pvp/wuvwuv/World-Rankings-as-of-9-28-New-Format/first

The short, short version is that ratings are a relative estimate of performance so if e.g. a higher ranked server beats a lower ranked server then the ratings aren’t as likely to change as if a lower ranked server beats a higher ranked server since the expected result matches closely to what actually happened. The closer the prediction is to the reality the less the rankings will change, the further off it is the more the rankings will change. Please see the referenced post for many more details.


ArenaNet Gameplay Programmer

World Rankings as of 9/28 (New Format)

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

This post was updated on 10/5 but the data is still from the previous matchup. The most recent matchup data is in the sticky thread: https://forum-en.gw2archive.eu/forum/pvp/wuvwuv/World-Rankings-as-of-10-5/


ArenaNet Gameplay Programmer

World Rankings as of 10/5

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Here are the results as raw text for those of you using them on websites, etc.:

Na
Name Rank Delta Rating Delta Deviation Volatility
Henge of Denravi 1 0 2139.487 -3.537 196.388 0.969
Jade Quarry 2 1 2085.029 41.021 197.145 0.971
Eredon Terrace 3 1 2027.236 39.637 205.82 0.973
Stormbluff Isle 4 -2 2011.719 -35.714 194.363 0.969
Isle of Janthir 5 0 1786.07 7.875 197.26 0.97
Blackgate 6 2 1743.089 49.651 192.673 0.971
Crystal Desert 7 -1 1690.733 -45.73 199.915 0.973
Dragonbrand 8 -1 1687.043 -40.129 195.482 0.971
Tarnished Coast 9 2 1666.646 96.312 191.598 0.974
Sea of Sorrows 10 0 1595.67 0.101 191.894 0.968
Maguuma 11 3 1587.102 139.361 192.691 0.979
Fort Aspenwood 12 -3 1586.597 -11.611 194.594 0.969
Yak’s Bend 13 0 1457.452 -71.957 193.38 0.974
Sanctum of Rall 14 3 1449.646 78.437 190.841 0.971
Northern Shiverpeaks 15 4 1440.231 105.522 207.212 0.986
Gate of Madness 16 -4 1436.989 -96.345 191.637 0.973
Ehmry Bay 17 -2 1363.606 -69.045 193.984 0.972
Sorrow’s Furnace 18 -2 1340.206 -48.253 190.801 0.969
Darkhaven 19 -1 1336.879 -30.3 191.127 0.969
Anvil Rock 20 0 1129.947 -39.867 201.759 0.978
Ferguson’s Crossing 21 1 1085.325 63.302 223.228 0.978
Borlis Pass 22 -1 1061.163 -64.62 208.581 0.977
Devona’s Rest 23 0 654.809 -68.877 211.564 0.976
Kaineng 24 0 631.153 15.263 221.084 0.976

Eu
Name Rank Delta Rating Delta Deviation Volatility
Vizunah Square FR 1 0 2072.236 14.06 204.729 0.972
Far Shiverpeaks 2 0 1884.769 -1.941 197.626 0.969
Riverside DE 3 0 1854.658 -11.918 198.315 0.969
Arborstone FR 4 2 1847.79 144.93 192.929 0.981
Elona Reach DE 5 2 1794.193 98.412 190.852 0.973
Augury Rock FR 6 -1 1690.918 -32.422 192.184 0.968
Desolation 7 -3 1688.912 -115.888 194.941 0.976
Seafarer’s Rest 8 2 1677.895 47.171 190.441 0.969
Kodash DE 9 -1 1646.664 -27.389 190.599 0.967
Aurora Glade 10 3 1641.794 78.984 190.601 0.971
Blacktide 11 0 1631.726 5.739 190.221 0.966
Gunnar’s Hold 12 -3 1590.644 -71.523 191.311 0.97
Gandara 13 3 1557.526 12.234 193.869 0.968
Baruch Bay SP 14 -2 1554.792 -53.024 190.558 0.968
Fort Ranik FR 15 -1 1536.396 -14.812 190.589 0.967
Jade Sea FR 16 -1 1484.671 -64.86 191.326 0.975
Ring of Fire 17 3 1455.14 102.203 192.382 0.974
Underworld 18 -1 1429.923 -16.101 192.559 0.969
Abaddon’s Mouth DE 19 -1 1418.271 4.247 193.775 0.97
Piken Square 20 1 1322.07 19.127 192.494 0.968
Miller’s Sound DE 21 2 1299.155 92.668 192.961 0.974
Vabbi 22 -3 1236.665 -121.163 192.382 0.977
Drakkar Lake DE 23 2 1222.095 88.186 207.59 0.976
Whiteside Ridge 24 -2 1179.482 -75.948 193.669 0.972
Dzagonur DE 25 -1 1129.986 -17.423 194.712 0.976
Ruins of Surmia 26 0 885.46 -30.012 200.423 0.975
Fissure of Woe 27 0 766.273 -54.658 205.182 0.974


ArenaNet Gameplay Programmer

World Rankings as of 10/5

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Here are the world rankings as of the end of the 9/29 00:00 UTC to 10/6 00:00 UTC matchup.

Attachments:


ArenaNet Gameplay Programmer

The real problem here is invisible enemies. Give their algorithms time to match servers properly.

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Hi mcl,

Thanks for pointing that out! It was my intention to be clearer on that point but that seems to have gotten lost in my attempt to acknowledge that better hardware is better.

So, let me be very clear: better hardware is better and will generally allow the game to look prettier but the issue with asset loading is not hardware dependent so buying an uber gaming rig won’t really help with the invisible player issue at this time.


ArenaNet Gameplay Programmer

The real problem here is invisible enemies. Give their algorithms time to match servers properly.

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

(Continued from the previous post…)

I’d like to take a moment to briefly address a few of the points of serious contention that have come up in this thread and hopefully clear up any confusion:

This is a multi-faceted issue that involves both reporting from the server to the client and client asset loading. Much like a super villain team-up these two parts of the issue are combining to make an unpleasant situation worse. We’re pursuing both parts of the issue and hope to see incremental improvements as we get fixes in and tested.

The reporting issue is really all about performance tradeoffs. Every decision that the server makes about what to report to which client consumes resources, as does the act of reporting itself. These resources (server CPU, network bandwidth, etc.) are finite and are the same ones that are used by every other aspect of WvW as well. How we make use of those resources determines the number of players we can support in a given map and how smooth the simulation feels. The system that we have in place now constitutes an attempt to strike a balance between a perfect simulation that handles all the details and makes them available to every client immediately and a simulation that supports a reasonably large number of players while maintaining smooth performance under most gameplay situations. This is a true dilemma because we really want to achieve both of those goals completely and simultaneously but that just isn’t possible at the moment.

Could we throw more hardware at the problem? Maybe, but the servers that we’re running on now are Serious Business™ and simply buying faster CPUs likely wouldn’t gain us even a linear increase in performance. With today’s hardware I believe that we’re likely to gain far more improvement from code changes than we would from slightly faster CPUs. As you might imagine the way that we manage client/server communication is pretty well core to the way our game works so making changes to that system is a tricky affair that must be undertaken with great care and much testing. Further, since we can’t really create any more resources (CPU, network) every substantial change involves making a hard decision about performance, scale, and completeness of the various aspects of Gw2. Some of the most robust, correct, and appealing solutions to this issue are also the ones that will take the longest to implement correctly, thus adding response time into the mix of factors we need to consider. In a situation like this, sadly, there are no easy answers. That said, we’re evaluating possible changes to reporting even now and are committed to making WvW into the best experience we can.

The client issue relates to the way that we load assets when preparing to display characters. WvW hits this more than most other parts of the game because players are pretty much the most complicated characters that we have and, especially at higher levels, they tend to be quite varied (so things like texture caching don’t help us as much as they might elsewhere in the game). WvW tends to have much higher player densities than the rest of the game so that’s why we see these issues coming up in WvW more than elsewhere. This asset loading issue will be influenced by client hardware (kind of like saying water is wet, I know) but we see this issue crop up on even high end systems so it’s clear that hardware is not the major determining factor. So, while better hardware may improve the situation a bit it won’t make the client issue go away completely. At this point we have a solid repro of the client issue and we’re aggressively pursuing fixes.

All of us who have worked on WvW (and many of those at ArenaNet who have not) are deeply invested in making WvW the best it can be. I personally have dedicated over a year of my life to developing this game type. Other have spent even longer. I know it can be terribly frustrating to deal with these issues (I’m a gamer too, I’ve been there!). I also know that frustration can make it tempting to believe that our silence on the forums means we’re ignoring issues with the game. Please believe me when I tell you that is simply not the case. We must always balance our time on the forums with our time spent working on the game. If we go silent for a while it’s generally because we’re busy working hard so that the next time we post we can have something substantial to tell you.

TL;DR: The issue is real, we’re aware of it, we’re working on fixes/improvements, the fixes/improvements are complicated and I can’t provide you an ETA.

Keep fighting the good fight and we’ll be back to let you know when we can share more details.


ArenaNet Gameplay Programmer

The real problem here is invisible enemies. Give their algorithms time to match servers properly.

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Hi all,

I just wanted to pop in to mention that we haven’t forgotten about you.

We’re aware that this is a serious issue and we’re actively working to improve the experience. I can’t talk about specific work that we’re doing yet, but I can say that we’re attacking the problem on multiple fronts and working hard to get WvW into a better state. This is one of those issues that involves multiple systems interacting in sometimes unexpected ways and as such it’s tricky to address. I would love to be able to snap my fingers and fix it today (which would make me a wizard, so that would be doubly awesome) but unfortunately the changes we’re working on will take time to implement and test thoroughly. Our goal is to make WvW the best experience it can be so we want to be very sure that any changes we make won’t have unintended, negative side effects.

(Continued in the next post…)


ArenaNet Gameplay Programmer

World Rankings as of 9/28 (New Format)

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

DeeJayTC: Yes, that’s how we do the matchups.

Sir William PD: The cycle ends at 23:59 UTC today (Friday)


ArenaNet Gameplay Programmer

World Rankings as of 9/28 (New Format)

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Just to clarify how the matchups work:

The data in this post is the result of the 9/22 (UTC) – 9/29 (UTC) one week matchup. The results from that matchup determined the groupings for the current (9/29 – 10/6) matchup. The results of the current matchup, which are calculated at the moment the games end, will determine the groupings for the subsequent matchup.

Currently we determine the matchups very simply by taking groups of 3 in order from the list.


ArenaNet Gameplay Programmer

World Rankings as of 9/28 (New Format)

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Well, it’s not a “for dummies” version but I have updated the post with details about how we do the calculations. That (and the details at the Glicko web site I linked) should be everything that you need to do the calculations yourself, if you so desire.

Note that Glicko 2 already incorporates the idea of multiple contests between rating recalculations.


ArenaNet Gameplay Programmer

World Rankings as of 9/28 (New Format)

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Hi Folks,

We’re trying out a new format for the world rankings that includes more data about how the rankings and ratings changed as a result of the last matchup.

Deviation and Volitility are used by the Glicko 2 rating system (http://www.glicko.net/glicko.html). They’re not absolutely necessary for understanding the ratings but I know there are some folks out there who will like to have them.

Enjoy!

Update:

Details have been requested so here’s the rest of what you need to understand how we’re using Glicko 2. These are fairly specific details so feel free to ignore them if you don’t get excited by math

Assume the shards are sA, sB, and sC. In order to handle a 3 way battle we treat each match as having two battles for each shard. So when calculating the new rating for sA the two battles are sA vs. sB and sA vs. sC. Likewise for sB and sC. Naturally, we do all the calculations before updating any of the data.

We use a Tau of 0.6 and a k of 1.0

In standard Glicko 2 scores are represented as 0.0 for a loss, 0.5 for a draw, and 1.0 for a win. We use a slightly modified version that takes into account score differential. I’ll explain how using the sA vs. sB battle as an example. To calculate the glicko score for sA in the sA vs. sB matchup we do the following:

sAPercent = sAScore / (sAScore + sBScore)
sAGlickoScore = (sin((sAPercent – 0.5) * Pi) + 1) * 0.5

where sAScore and sBScore are the raw scores from the end of the match.

That last transform is easiest to visualize as a graph . Other than that it’s all standard Glicko 2.

Have fun doing the math!

Update 10/5

New ratings & rankings will be coming out later today but because I know some of you want to use the rating data in your own site/applications here’s the data from this post as text:

Na
Name Rank Delta Rating Delta Deviation Volatility
Henge of Denravi 1 0 2143.024 5.851 198.446 0.97
Stormbluff Isle 2 0 2047.433 18.399 195.343 0.969
Jade Quarry 3 1 2044.008 58.468 205.496 0.973
Eredon Terrace 4 -1 1987.599 -24.954 198.027 0.972
Isle of Janthir 5 2 1778.195 86.539 193.653 0.972
Crystal Desert 6 -1 1736.463 -51.728 196.391 0.973
Dragonbrand 7 -1 1727.172 -1.89 199.008 0.972
Blackgate 8 2 1693.438 80.143 191.771 0.971
Fort Aspenwood 9 0 1598.208 -25.806 192.261 0.971
Sea of Sorrows 10 -2 1595.569 -60.171 193.3 0.97
Tarnished Coast 11 2 1570.334 64.755 192.341 0.971
Gate of Madness 12 -1 1533.334 -41.097 191.082 0.969
Yak’s Bend 13 -1 1529.409 -38.597 191.596 0.974
Maguuma 14 0 1447.741 -43.134 192.273 0.969
Ehmry Bay 15 0 1432.651 -22.869 197.303 0.972
Sorrow’s Furnace 16 0 1388.459 -5.07 191.216 0.969
Sanctum of Rall 17 0 1371.209 -21.589 191.251 0.968
Darkhaven 18 0 1367.179 27.301 192.631 0.97
Northern Shiverpeaks 19 1 1334.709 188.69 206.117 0.989
Anvil Rock 20 -1 1169.814 -143.691 204.597 0.984
Borlis Pass 21 1 1125.783 70.803 223.876 0.979
Ferguson’s Crossing 22 -1 1022.023 -34.62 206.1 0.973
Devona’s Rest 23 0 723.686 -53.993 211.691 0.975
Kaineng 24 0 615.89 -8.437 225.012 0.977

Eu
Name Rank Delta Rating Delta Deviation Volatility
Vizunah Square FR 1 0 2058.176 -10.866 206.669 0.973
Far Shiverpeaks 2 0 1886.71 -0.751 198.36 0.97
Riverside DE 3 0 1866.576 9.291 198.116 0.97
Desolation 4 0 1804.8 1.706 196.767 0.969
Augury Rock FR 5 0 1723.34 16.141 192.558 0.968
Arborstone FR 6 4 1702.86 52.072 191.129 0.972
Elona Reach DE 7 1 1695.781 27.423 190.429 0.968
Kodash DE 8 -1 1674.053 3.88 190.99 0.967
Gunnar’s Hold 9 -3 1662.167 -18.389 192.88 0.968
Seafarer’s Rest 10 -1 1630.724 -31.319 190.548 0.97
Blacktide 11 1 1625.987 11.587 190.345 0.966
Baruch Bay SP 12 1 1607.816 0.391 191.085 0.966
Aurora Glade 13 -2 1562.81 -63.208 190.425 0.969
Fort Ranik FR 14 1 1551.208 21.225 191.183 0.968
Jade Sea FR 15 2 1549.531 141.963 192.603 0.979
Gandara 16 -2 1545.292 -21.452 190.526 0.967
Underworld 17 2 1446.024 48.356 192.714 0.97
Abaddon’s Mouth DE 18 -2 1414.024 -99.479 193.711 0.973
Vabbi 19 -1 1357.828 -43.918 193.274 0.971
Ring of Fire 20 0 1352.937 0.545 193.992 0.971
Piken Square 21 1 1302.943 43.763 194.132 0.969
Whiteside Ridge 22 -1 1255.43 -49.342 193.454 0.97
Miller’s Sound DE 23 1 1206.487 26.538 194.218 0.971
Dzagonur DE 24 2 1147.409 170.19 195.991 0.986
Drakkar Lake DE 25 -2 1133.909 -68.605 192.424 0.97
Ruins of Surmia 26 -1 915.472 -137.637 198.32 0.98
Fissure of Woe 27 0 820.931 -37.014 199.609 0.973

I’ll be sure to include text data along with our pretty (?) pictures in subsequent posts.

Attachments:


ArenaNet Gameplay Programmer

(edited by Moderator)

A different kind of Kaineng War Report.

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Good luck Fubuki! All of us who work on WvW are pulling for you.

Kaineng consistently has the best queue situation of any NA server so it seems like an obvious win for any guilds looking to play together. I hope that your PR efforts continue to pay off!


ArenaNet Gameplay Programmer

There is a 4th enemy in WvW - Lag

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Hi all,

An encounter with Lagzilla is always rough and it’s a tricksy beast that can manifest for a variety of reasons. Ultimately the experience of lag comes down to one or more of client performance, network issues, and server performance. If the lag you’re experiencing is a result of client performance issue then turning down your graphics settings, running at a lower resolution, etc. can help. If not then there’s unfortunately very little you can do. Let me explain a bit more about the different issues that can cause lag:

Client Performance
When your client is trying to do too much work (usually this comes down to animating and rendering the game world since that’s the bulk of the work done on the client, though it can be other tasks as well) the time it takes to create each frame goes up and, thus, your frame rate goes down. When the frame rate drops a little the game just doesn’t feel as smooth but when it drops a lot (say, into the single digits) then the experience can start to feel like lag in that you have to wait longer to get feedback about your actions. Mostly though, this situation just feels choppy. You can check your frame rate in Gw2 on the options menu and if it’s unacceptably low then changing your client settings may help.

Network Issues
The internet is a big, complex, self-healing network. It truly is a marvel of the modern world. Despite that it’s designed to be fully redundant and self healing the internet still runs into problems sometimes. Sometimes the server you’re connecting to is just too far away and thus it takes your packets a long time to make it to the server and back. Other times a fault with some portion of the network along the path from you to the server can introduce delays or cause packets to simply be lost. Any of those issues can cause a feeling of lag by increasing the time between when you try to do something and when it happens on the server as well as increasing the time between when something happens on the server and when you see it on your client. While some amount of latency is just a fact of life on the internet (nothing is instantaneous, curse you speed of light!) serious issues with the network are relatively infrequent. When they happen it’s no fun at all, but thankfully there are lots of people at lots of network providers who spend their time preventing these issues and fixing them when they occur so they’re usually fairly short lived.

Server Performance
Server performance is a lot like client performance in terms of lag. The work the server is doing is different, of course, but fundamentally it still has a limited amount of time to present any given frame of the game and if it takes too long doing that work then things start to back up and everyone on the server experiences lag. In large WvW fights, such as when the majority of the players on the map are fighting over Stonemist Castle, this is most likely what’s causing you to experience lag. When a lot of players are in relatively close proximity to each other the server has to do more work because all the interactions (or, in some cases, potential interactions) between those players need to be calculated. This is the N-squared problem that you may hear computer scientists going on about all the time, assuming you hang out with computer scientists. Of course we’ve done what we can to optimize these processes but at the end of the day more work is more work, no matter how you slice it, so there will always be some player density that’s just too much for the system to handle in the time allotted. Ultimately we end up in a balancing act between trying to allow the maximum number of players into WvW and minimizing the amount of lag that players encounter.

I hope that this information is helpful when trying to determine if changing your client settings will have a positive impact on lag you may be experiencing. All of us at ArenaNet are also gamers and, believe me, we hate lag just as much as you do! We are constantly looking for ways to improve performance and the work of optimization never stops.


ArenaNet Gameplay Programmer

The real problem here is invisible enemies. Give their algorithms time to match servers properly.

in WvW

Posted by: Habib Loew.6239

Previous

Habib Loew.6239

Gameplay Programmer

Next

Hi all,

For a variety of performance reasons we limit the number of characters that are reported to any given game client. This report limiting (or culling, as it’s also been referred to) is generally distance based and limits both the amount of bandwidth and client side processing (rendering, etc.) required to play the game. Ideally this shouldn’t be something that you notice happening as characters will simply fade in when they’re “far” away from you. In the best case this happens far enough away that even if you’re looking right at them when it happens it isn’t too visually distracting. Unfortunately, there are some situations in the game in which this setup doesn’t work as well as we’d like and it seems that those situations come up in WvW rather more often than in other parts of the game. The higher player densities that we see in large battles are an obvious example of where this system goes awry. If only the nearest N characters are reported to you but there are N+100 characters within effective battle range then many of those characters will be invisible. There’s never a great time to be dealing with invisible characters, but I think that it’s fair to say that during a large battle is one of the worst times.

In WvW one of the things that we see exacerbating the issue is this: From the moment a character is first reported to your client to the first moment that your client is able to render it a non-zero amount of time passes. During this time your client is doing things like loading textures from disk, which can be (at least in computer terms) fairly slow what with all that accessing of spinning, physical storage media. So that means that a character who is moving towards you can potentially appear first at a much closer point even than the one at which they were reported because, of course, they were still moving during that load time.

MajorKong’s screenshot doesn’t immediately look like the situation I just described so it may be that there’s some kind of bug lurking in there as well. We will certainly be looking into that possibility.

That was a lot of detail but really I’m posting to let you know that we’re aware of the issues associated with invisible enemies and we’re working on finding both the root causes and effective solutions. I understand that these experiences can be quite frustrating but please rest assured that we do care and we are working on improving the experience.

Thank you for reporting your issues here on the forums and for your patience and understanding as we work to resolve them.

Also, MajorKong, that really was a great screenshot even if it does show a bug and bugs make me sad.


ArenaNet Gameplay Programmer