Appreciate the feedback.

I mentioned this in some other posts but we have put fixes in EOG late last year and into this year, and it did resolve some key pains in End OF Game (EOG) (improved time to get to post game + fixes in reconnect bug showing up incorrectly), but clearly, it was not enough and there is a larger issue with EOG. EOG is compromised of not just the client, but the game engine plus services.

We are doing a deep dive this quarter to determine where the performance bottlenecks are in EOG, and will work with other teams to resolve them.

As far as the client getting worse, this is not true. Although reddit is important for us to see region-based trends and address issues that multiple people report, it is not a true measure of success for the work our team is doing, on a global scale

We use a survey (you may have got it), that is run in Brazil, South Korea, North America, Vietnam, and China.

We launched this survey in September 2020, and measure the client satisfaction on a scale from Very dissatisfied to very satisfied (5 point scale), plus other things

When we first launched the survey, we had 50% of players Very dissatisfied + dissatisfied with the client in these regions combined. Today, we have seen over a 10 % drop, so this proves the client on a global scale is better. So now is 39% Very Dissatisfied + dissatisfied good enough? heck no, we still got work to do.

We also track a number of other data points (Memory footprint, crashes, JS errors, loading times, plus many more) that have all gotten better since last year, which is driving the positive results we are seeing when players are filling out this survey.

For region breakdowns, South Korea has the highest rate of satisfaction with the client, followed by NA, China, and BR. If we look at why BR has the lowest satisfaction, we know that network and infrastructure in that region is not as good as South Korea, so we know there is a component of network + infrastructure services we do not have control over. But again, we are trying to make fixes that improve the client globally for all our players, not just a specific region.

This is where the CEF upgrade comes in, and we have highlighted the benefits players will see with this in our latest post, followed by our continued investigation on EOG, plus coming up with a way to improve the client between game sessions. We noticed a lot of folks leave their clients on and leaving it on tends to produce errors we are unable to reproduce, so we are trying to put in changes that will 'refresh' the client so players get the best experience starting and getting into and out of a game.

I feel confident when we deliver on the above, the client will continue to get better and better!

Thanks

SuperMrBlob

Hey Am1t8,

Happy to see you here. Hoping you could answer a question I had after reading the latest dev blog.

To me, one of the head-scratchy things about the CEF upgrade is that other brilliant game clients existed in Nov 2019 - it's not like the technology available at that date was holding Riot back. Doesn't this suggest that a CEF upgrade is simply a band-aid fix to the clear technological issues the client has? At worse, its gloss of performance could hinder diagnosing these issues. I'm sure the upgrade will help, but doesn't more work have to also be put into fixing some of the client's fundamental architectural issues (without knowing more, e.g., completely deleting the plugin architecture)?

We are 16 versions behind CEF, there will be good performance and reliability gains with this upgrade. A lot of the crashes that we get in clients today we are hypothesizing (based on the crash data) are related to outdated CEF version, so we expect crashes to drop after we upgrade.

on architecture, yes, we need to continue to work on reducing plugins, right now, we want to deliver more player value, but we need to get to our final architecture state also

420-IQ-AUTIST

When we first launched the survey, we had 50% of players Very dissatisfied + dissatisfied with the client in these regions combined. Today, we have seen over a 10 % drop, so this proves the client on a global scale is better. So now is 39% Very Dissatisfied + dissatisfied good enough? heck no, we still got work to do.

So was there additional context data factored into this?

Is this 50% decreased to 39% rating purely on players/accounts that voted in both surveys? Or just total account participation?

If not then you could have 2 situations arise that invalidate this data.

1. Players that were exceedingly dissatisfied with a billion dollar company not making meaningful updates to their game unless it directly relates to money making, might have gone ahead and left the game. Therefore they are not present to continue voicing their dissatisfaction, and the rating "gets better"
2. New players who only joined after the first survey but before the second giving a positive rating because "yay i like this new game" without any thought or extended experience of the failures of the client.

Beyond these edge cases, it's frankly an obvious and borderline malicious twist of the data to say that "results of an opinion gathering survey" correlate to "the client has gotten better". Because come on now, we know that player opinion does not equal measurable improvements.

Appreciate the comment.

We work with an advisory team internally that publishes this survey and it's an 'always-on' survey, meaning a random sample size throughout the week gets an opportunity to take the survey, this allows us to correlate our fixes after every patch to the results in this survey. We track with a four-week rolling average and have confirmed that the positive changes we make do influence the sentiment while outages (that may not even be related to the client) hurt sentiment.

and again, we have a number of data points that show the client has gotten better since last year, and we will continue improving it for our playersthanks