Archive

Posts Tagged ‘clairvoyance blog’
Comments off

CLAIRVOYANCE BLOG

It is a truth universally acknowledged that LAG SUCKS. Plenty of players experience ping spikes, lag, and full-blown disconnects all the time. But what if there was a quick fix for all your connection issues? For some players, there might be: switching from WiFi to an Ethernet connection when you play could have drastic results for your in-game experience.


I’m Viscarious, a product manager on Riot’s live services team. About six months ago, I lagged out of a game so hard that I decided I had to do something about it. It was the last game of my placement series, and I was playing as Sona supp. Our jungler was thundering into bot lane for a gank, and my Jinx ADC landed a sick double trap on our opponents (Leona and Ezreal). First blood was guaranteed. I was just about to follow-up with my ult when the lag gods smote me. The next thing I knew, Jinx was dead and the enemy Ezreal was snowballing. All I could do was apologize and resist the urge to kick my router.

I had to wonder: what if my problems were caused by my WiFi connection? What if, to claim my rightful spot as a true PC gaming god at the top of the challenger ladder, all I had to do was switch to Ethernet? To understand the problem as deeply as possible (and because this is a Clairvoyance blog post) I dug into some data to find out.

 

TERMS TO UNDERSTAND


 

PLAYING PING PONG AND LOSING SOME PACKETS

 

WIFI VS. ETHERNET: AVERAGE PING DELTA
IN MILLISECONDS

When you look at League players around the world, the average ping for players on a WiFi connection is between 6.7ms and 11.7ms higher than for those using an Ethernet connection.

However, this usually doesn’t manifest as consistently higher ping. Instead, increased ping is frequently experienced as ping spikes; your ping increases significantly over a short period of time, then goes back down a few moments later(usually after your whole team is already dead and the enemy Yi is dancing around your nexus). The chart above doesn’t show the severity or duration of ping spikes—just the average ping difference over the course of many games.

 

WIFI VS. ETHERNET: AVERAGE PACKET-LOSS DELTA

IN PERCENTAGE POINTS

Similarly, we found that players on WiFi had between 1.9 and 3.7 percentage points higher packet-loss than players on Ethernet. While this may not seem like a lot, there are two things to consider:

1. Every time a packet is lost between your computer and Riot’s servers, a request has to be made for that packet again. You want your packet-loss percentage numbers to be as close to zero as possible because, depending on when the packet-loss happens, you could miss a last-hit, fail to activate your ult, or even whiff the smite on Baron (that’s definitely what it was, right junglers?).

2. Similar to our ping chart, this doesn’t show the magnitude or duration of packet-loss when it happens. Players most commonly feel lag when there are big spikes in packet-loss. The 1.9 to 3.7 percent figure is just the average difference between WiFi and Ethernet players over the course of many games.


 

WHO’S PLAYING ON WIFI?

Out of curiosity, we analyzed who is playing on Ethernet and WiFi and came across some pretty interesting results:

Some regions rely on WiFi way more than others. While over 90 percent of games in KR are played on Ethernet, well over half of games played in NA and OCE are on WiFi.

 

PERCENT OF GAMES PLAYED BY CONNECTION TYPE

BY REGION

Part of this is likely due to the popularity of hard-wired PC bangs in Korea, but it’s also probably because we NA players are complete scrublords.

We saw a slight increase in Ethernet use in Ranked games compared to ARAM and Normal games across all regions.

 

PERCENT OF GAMES PLAYED BY CONNECTION TYPE

BY GAME MODE

But most interesting was the change in Ethernet vs. WiFi use by rank. Across all regions, a higher percentage of high-rank games were played on Ethernet. Although we’re not able to draw a causal relationship between playing on Ethernet and an increase in your rank, it’s clear that players at higher ranks are more likely to play on Ethernet. My best guessplanation for this is that highly ranked players are more likely to do everything possible to play on glorious, photo-worthy, wired battlestations.

 

PERCENT OF GAMES PLAYED BY CONNECTION TYPE

BY TIER


 

HOW DOES WIFI AFFECT GAME PERFORMANCE?

We now know two things: 1) WiFi has an adverse affect on connection quality, and 2) it’s unlikely that Faker has ever played on a WiFi connection. So the next question is whether your WiFi or Ethernet connection affects your in-game performance in a measurable way.

For this analysis we decided to look at a range of gameplay metrics including: Minions Killed (CS), Gold Earned, K/D/A, Mastery Grade, and Win/Loss ratio. We decided to isolate our analysis to ranked games and players who played the same champion on both Ethernet and WiFi within the timeframe analyzed. Basically, we didn’t want to compare the same player’s Mastery Grade with Fizz on Ethernet to their Mastery Grade with Fiddlesticks on WiFi (since ping affects some champs more than others).

We didn’t find conclusive evidence that playing on WiFi negatively impacts certain specific performance metrics such as CS, K/D/A, or Mastery Grade.

Our analysis included only NA region players, but the results should apply globally.After all that gloom and doom we found about ping spikes, the results surprised us. No matter how we cut the data, we didn’t find conclusive evidence that playing on WiFi negatively impacts certain specific performance metrics such as CS, K/D/A, or Mastery Grade.

We have a few hypotheses regarding this:

1. While lag experienced from WiFi can negatively impact gameplay, it usually only has noticeable effects intermittently, so it’s difficult to isolate the effects within the course of a game (i.e. it’s hard to find the signal among the noise)

2. Players on WiFi may be able to adapt to added ping and packet-loss (playing around it, effectively).

3. Since there are four other players on the team, the impact of one WiFi player may not significantly influence the outcome of a game, especially if there are other players on WiFi on the other team.

When we removed the constraint that players had to play the same champion, we found that the win-rate on Ethernet was 1.1% to 1.7% higher than on WiFI.

Surprisingly, when we removed the constraint that players had to play the same champion across WiFi and Ethernet, we found that the win-rate on Ethernet was 1.1 to 1.7 percentage points higher than on WiFi. This is pretty consistent across regions. We’re really not sure why removing the champion constraint had this result. One hypothesis is that by controlling for champion across connection types, we’re also focusing more on players’ main champs. If you main a champ, you probably get used to dealing with the ping spikes that come with WiFi, but those same spikes hit you harder when you’re just learning a champ.

 

ETHERNET VS. WIFI: DELTA IN WIN-RATE

IN PERCENTAGE POINTS


 

MY PERSONAL EXPERIENCE

The same night I lost that disastrous promo game, I popped open a browser tab and bought a 50 ft. Ethernet cable. After untangling my cat from the wire, I ran diagnostics on my network and found some pretty drastic differences in the quality of my connection on Ethernet vs. WiFi.

 

CONNECTION QUALITY OF WIFI GAME SAMPLE


Ping



Packet-loss



Both


 

CONNECTION QUALITY OF ETHERNET GAME SAMPLE


Ping


Packet-loss



Both


 

CONNECTION METRIC WIFI
SAMPLE
ETHERNET
SAMPLE
AVERAGE PING 62.4ms 38.2ms
AVERAGE PACKET-LOSS 3.9% 0.0%
JITTER 33.7ms 0.21ms
PING ‘SPIKES’ (>2X LOWEST PING) 19 0
PACKET-LOSS ‘SPIKES’ (>15% PACKET-LOSS) 9 0

Sure enough, I was experiencing higher ping, more packet-loss, and more connection issues when on WiFi. Especially interesting was theway I was experiencing lag. While there were periods of stability over the course of the game, there were also short periods with large spikes in either ping or packet-loss. These were the times that I felt lag most acutely.

 

 

My experience was pretty extreme, but let’s throw this to the commenters: what’s your own experience using either connection type? What lengths have you gone to improve your connection (longer than 50ft?) and how many of you intend to switch to Ethernet after reading this post?

 

Ping Winrate and Vayne Probs Banner

Riot’s latest dev blog explores how high ping affects winrate depending on a Champion’s skill cap:

 

Related:

Heating Up Banner

Warding and You Blog

 

ARTICLE BY CHRISTOKKIES, DESIGN BY NANCYMON

Nothing quite kills the joy of online gaming like latency. At one time or another we’ve all experienced the pain of lag spikes during a pivotal, game-deciding moment.

One thing that I’ve been mulling over when it comes to latency (or lag) is how champions in League of Legends might be affected differently by slow communication between a player’s client and the game server.

My brother, who lives in Japan, plays on the NA server so we can play together. He intentionally avoids playing AD carries because, his logic goes, this mechanically intensive role is not conducive to being played in conditions of high latency.

But just how true is his supposition? Are certain roles (or champions) more susceptible to a drop in performance from increased latency?

 

METHODOLOGY Banner

To test this, I went all out. I created a series of statistical models that attempt to predict a game’s outcome (win or loss) based on one’s ping. As it turns out, the use of latency as a predictor of a game’s outcome is contingent on the champion being played. In other words, the outcome of a game with bad ping is easier to predict with certain champions more than with others.

The use of latency as a predictor of a game’s outcome is contingent on the champion being played.

Classical linear regression models make predictions about continuous variables, where numbers have a logical order (such as age or or number of wins on a champion). Because the outcome variable in this case is categorical and binary (win or loss), a type of regression known as logistic regression was used to determine the estimated probability of an outcome given latency. I analyzed over 95 million different occurrences of a champion appearing on Summoner’s Rift. To emphasize relative differences in latency between players, I used difference from the average in-game ping (which we’ll just call ‘ping difference’ going forward) as the predictor variable rather than the absolute ping values. For example, if a player’s average ping during a game is 75, while the average ping of the game for everyone else is 70, that predictor value will be 5 (rather than 75).

 

FINDINGS Banner

It appears that more mechanically intensive champions are more affected by latency, while tankier champions or those with point-and-click abilities are less affected by latency.

The graph below shows the estimated probability of Vayne being on the winning team as a function of ping difference. As the graph demonstrates, the lower the average difference in ping, the better chance this champion that relies on extremely precise positioning will tumble into victory rather than the enemy team.

Vayne Stats 1

The estimated probability of Vayne being on the winning team as a function of her difference in ping from the average in game.

coeff. = -.0014, z-value = -41

 

This relationship is similar for Xerath. The higher a Xerath’s latency is relative to other players in the same game, the harder time the Shurima demigod has landing skillshots on his opponents.

Xerath Stats

The estimated probability of Xerath being on a the winning team as a function of his difference in ping from the average in game.

coeff. = -.002, z-value = -20

 

I believe this intuitively makes sense. Landing a Xerath ultimate or a Vayne condemn when your opponents’ movements are more responsive than your own is difficult to put it mildly.

 

PRESS R TO WIN Banner

Certain champions, however, do not exhibit a strong relationship between latency and estimated probability of winning.

Looking at Warwick, for example, differences in the average latency between other players in the game is not a strong predictor.

Warwick Stats

The estimated probability of Warwick being on a the winning team as a function of his difference in ping from the average in game.

coeff. = -.0002, z-value = -4

 

Similarly, Singed’s probability of landing on the winning team does not seem to be affected much by latency.

If a Singed has 30 less ping than the average champion in that game, he has a roughly 50 percent chance of being on the winning team. And if that Singed has 30 more ping than the average champion, the chance of him being on the winning team is…well… still around 50 percent.

My interpretation is that the skills necessary to excel at Singed are what I would consider more strategically focused. When and where do I ward? When do I split? When do I TP or group?

Latency should only be affecting the outcome of a match to the extent that it differs from that player’s normal ping.

Further, I am reluctant to rule out any effect of lag on Warwick’s or a Singed’s ability to win. Matchmaking is designed with the intent that players are matched against other players in such a way that they win 50 percent of their games on average. If high latency is systematic for a player, then the effect of lag on a champion’s ability to win should be muted. For example, if a player who normally has good ping suddenly has bad ping, then that player will probably be far more likely to lose that particular game than if that player consistently plays with bad ping from game to game. In other words, latency should only be affecting the outcome of a match to the extent that it differs from that player’s normal ping.

Finally, I would like to caveat these findings by noting that the interaction between latency, champion, and the estimated probability of winning may not be causal. The models shown here only have a single predictor variable, and it’s possible that if we throw other variables into the mix that the effect of latency disappears. Having said that, I think that the data support the hypothesis that the effect of latency on the outcome of the game differs by champion. A future analysis using more complex models could perhaps provide more evidence of causality.

Singed Stats

The estimated probability of Singed being on a the winning team as a function of his difference in ping from the average.

coeff. = -.0003, z-value = -2

 

CAN WE GENERALIZE THESE FINDINGS TO ROLES

Let’s come back to the question of whether my brother is correct to avoid playing as an AD carry because of his high latency. To answer this question, I ranked the champions by the extent to which their estimated probability of winning was affected by latency (for the more statistically inclined, I ranked the champions by the normalized regression coefficient given by the Z value). I then took the median of the rankings for each champion by role. For example, Yasuo appeared to be the most affected by lag, so he was rank 1. Vayne was the second most affected—making long-distance duo-queuing with a Vayne main somewhat ill-advised.

We do see a difference in the effect of latency on the estimated probability of winning based on role.

As it turns out, we do see a difference in the effect of latency on the estimated probability of winning based on role. Champions categorized as ‘AD carries’ (median rank = 21) appear to suffer the most from latency, followed by the mid role (median rank = 50). Finally, support, top, and jungle all had relatively higher rankings (70, 79, and 93 respectively).

These findings are in line with the notion that roles are differentially affected by latency, and in fact, my brother is probably better off playing Top, Support, or Jungle over AD Carries or mid given his higher-than-average latency.

 


If you have any questions, feel free to ask me at @NoL_Chefo or e-mail me at nolchefo@gmail.com.