Thread / Thread Starter Developer Post Forum
DDOS prevention. Something to look at maybe? RiotGradius
02-18-2014 11:38 PM
General Discussion

RiotGradius

Associate Information Security Engineer

02-18-2014
14 of 17 Riot Posts

Quote:
Originally Posted by Cjkory View Post
Hi RiotGradius,

I was wondering (if you dont mind sharing), what sort of background you (or someone looking for a similar job) have in terms of education and certifications?

I am 25, finishing my bachelor's in CS, and looking at various certifications. One thing that has always been of interest to me is IT Security. I worked with a software security company (both hardware and software encryption) for about 8 years, and it really got me more and more interested.

I know CISSP is and always will be good to have, but I was wondering if you wouldnt mind sharing any other info that could help me along my path.

Thanks in advanced! And thanks for all the hard work you guys do to keep the servers up and running. I know its stressful, but it must also be fun at a level, knowing not only are you trying to prevent current attacks, but also secure for the future!

~Cjkory
tl;dr of my education:

-BS in Computer Science
-Lots of club activities, manly related to secure communications/telecom (since computer science really didnt have those classes)
-Participated in multiple Collegiate Cyber Defense Competitions. (http://www.nationalccdc.org/)
-Went to conferences like DEFCON and SCALE
-Met a few awesome Rioters from OPS during SCALE during my last year of school (woo 5 year plan!)
-Scored an internship, worked my ass off for a full time job, proved that I was passionate about my craft.


DDOS prevention. Something to look at maybe? RiotGradius
02-18-2014 11:31 PM
General Discussion

RiotGradius

Associate Information Security Engineer

02-18-2014
13 of 17 Riot Posts

Quote:
Originally Posted by WuTangFinancial View Post
I know it's been brought up that Riot uses Cloudflare for DDoS protection. But isn't Cloudflare only designed to resist web DDoS attacks on non-UDP ports?

https://support.cloudflare.com/hc/en...are-work-with-

I bring this up because I'm assuming that you're using game servers that are discrete from production web servers hosting the LoL website, and that game servers are communicating with clients over a series of UDP protocols that similarly not defended by Cloudflare.

Does Riot use Prolexic or some of the other non-web DDoS software suites to protect the actual game servers from DDoS? Or are the game servers somehow connected to the web servers such that taking down the web servers with a DDoS attack will actually impact gameplay for players? If so - why are they configured like that?

Also if this is a NTP type attack: any way you could close the port for NTP (123) for a little bit? I get that you still need some way to federate time, but can't you just do some kind of port forwarding via an out of band port for federating NTP while the campaign is underway and your ISP figures out what's up and how to stop this (as you said by just setting up a bunch of ACLs and stopping requests coming in)?
The biggest problem with blocking NTP at the provider level is that we're still working up the chain. The amount of traffic coming through is so huge, that it's taking down multiple provider links. This is why part of our attack strategy is building up relationships with multiple providers and others in the industry. This isn't something we're going to solve overnight, and definitely not something we'll solve by ourselves. Working with many parties can cause inherent lag in the process, but we're kicking ass and taking names in terms of making friends all over the industry.


DDOS prevention. Something to look at maybe? Riot Triggs
02-18-2014 11:30 PM
General Discussion

Riot Triggs

NOC Technician

02-18-2014
12 of 17 Riot Posts

Quote:
Originally Posted by Linna Excel View Post
So how are you guys notified something like this is happening? It's got to be getting late over there so I imagine it to be a phone call at home in the middle of the night or something like that.

I've had a family member who's had to do on-call tech support in the past and even the light stuff tends to be a PITA.
We have people monitoring 24/7.


DDOS prevention. Something to look at maybe? RiotGradius
02-18-2014 11:15 PM
General Discussion

RiotGradius

Associate Information Security Engineer

02-18-2014
11 of 17 Riot Posts

Quote:
Originally Posted by BDubz650 View Post
I'll admit I'm not an expert by any means, but why can't the company that runs the most popular video game in the world just throw enough money at the issue to resolve it? Can't they just purchase more ISPs, more security infrastructure? Hire people to find and stop the attackers? They surely have more resources than the attackers do, they should use them.

Sure these things take time, but its not like this is a new issue, they should've seen this coming, no?

Its gotten to the point recently that more times than not the game is 100% unplayable. Starting to get frustrating.
I guess the beauty of competition is that alternatives exist.
It's not that this a new issue, it's an old issue with new scale. Technically, we could throw money at it and buy more bandwidth, but that's only a temporary solution, and doesn't buy anyone much in terms of stability. As your bandwidth grows, so does the attacker's. This is why we have to find more solutions than just throwing money at it. This is why we have to work with other companies and groups in the Information Security industry, because of the magnitude of these recent attacks. We want this to be more complex than a bandwidth arms race.


DDOS prevention. Something to look at maybe? Riot Triggs
02-18-2014 11:09 PM
General Discussion

Riot Triggs

NOC Technician

02-18-2014
10 of 17 Riot Posts

Quote:
Originally Posted by Vortex IV View Post
Common sense solution, matchmake people together on the same server so that doesn't happen. It'll still happen in party situations but it can't be perfect.
All players in a game play on the same game server.


DDOS prevention. Something to look at maybe? Riot Triggs
02-18-2014 11:07 PM
General Discussion

Riot Triggs

NOC Technician

02-18-2014
9 of 17 Riot Posts

Quote:
Originally Posted by Zathamos View Post
Do you know if uninstalling the game entirely, then reinstalling and allowing it to reupdate would have any affect on whats going on with the 'stuck in a game that already ended and that you cannot reconnect to,' problem? Or would it just be a waste of time because it doesn't have to do with anything the program downloaded? I don't understand and wont try to understand what these attacks are if they are DDOS or whatever. But I didn't know if possibly whatever is wrong could be stored in the downloaded game data of the game storage in the LOL file or if it strickly online having to do with the servers and therefore completely refreshing it might do anything
No, it's not a local problem.


DDOS prevention. Something to look at maybe? Riot Triggs
02-18-2014 11:06 PM
General Discussion

Riot Triggs

NOC Technician

02-18-2014
8 of 17 Riot Posts

Quote:
Originally Posted by The Drezak View Post
Anyways quick question. How are you guys going about Loss Prevention with these attacks? Do you turn it on as soon as you see it happening or wait a bit? I see a lot of people complaining about losing because of the attacks but I feel like you guys would turn it on as soon as you see it happening...

Other question would be any timeline on fixing the issue? I'd assume you wouldn't know this answer quite yet.
Loss prevented is manually turned on by the department I work in (the NOC). It's usually done after we verify it's a large scale issue because it has a significant impact on that server. It shuts down ranked for everyone online. It tags any games currently in progress, and even if we shut it off before that game ends, it still applies. There is always a small % of players who's game ends in the few minutes it takes us to verify the severity of the incident.


DDOS prevention. Something to look at maybe? Riot Triggs
02-18-2014 10:56 PM
General Discussion

Riot Triggs

NOC Technician

02-18-2014
7 of 17 Riot Posts

Quote:
Originally Posted by eertelppa View Post
Probably a dumb question, but are these attacks, to which I assume are causing many of the ping/frame-rate/lag issues, any threat on the security of our personal information? I know you may never be allowed to say yes, bluntly, but should we take precaution at changing passwords and other measures?

Not sure why someone would want to perform this DDoS attack on internet providers other than having fun or something. Seems like the are not collecting information or data from them?

Thanks for working on this and best of luck. Thanks in advance for an answer.
You should always use a secure, long, hard to guess password. Using this site: https://howsecureismypassword.net/ (probably want to get past 1 month.) Also, use two factor auth anywhere you can. However, this has nothing to do with them access your personal information. DDoS attacks basically jam too many packets into a pipe. Like trying to put too many cars on a freeway.


DDOS prevention. Something to look at maybe? RiotGradius
02-18-2014 10:39 PM
General Discussion

RiotGradius

Associate Information Security Engineer

02-18-2014
6 of 17 Riot Posts

Quote:
Originally Posted by The Drezak View Post
First off thank you SO MUCH for posting. That's all I've wanted from riot is them letting us know you're working on it (even though we knew.. reassurance is always nice)

Anyways quick question. How are you guys going about Loss Prevention with these attacks? Do you turn it on as soon as you see it happening or wait a bit? I see a lot of people complaining about losing because of the attacks but I feel like you guys would turn it on as soon as you see it happening...

Other question would be any timeline on fixing the issue? I'd assume you wouldn't know this answer quite yet.
Loss prevented is intended to be an automatic thing when our service level goes below that of what we'd expect. You should see those changes apply fairly quickly (though, I'm not an expert in that area, I mostly just see the emails fly by with "Loss prevented is being enabled")

As for a timeline, that's hard to say since we're looking at a problem that's a huge scale. Just know that we're working tirelessly to get this under control. We want the network to be stable as bad as everyone else does


DDOS prevention. Something to look at maybe? RiotGradius
02-18-2014 10:35 PM
General Discussion

RiotGradius

Associate Information Security Engineer

02-18-2014
5 of 17 Riot Posts

Quote:
Originally Posted by Linna Excel View Post
Is it possible when you are starting to see an unnatural spike in pings that servers just ignores anything it doesn't already have a connection with?
The problem with that is a lot of the time it's not even our network that's being destroyed by the malicious traffic, it's usually our providers or even their providers. This is why as a whole, network operations is moving towards a solution of being closer to providing their own Internet connections to our players. I'll find the post on the EUW forums that references things like the AMS datacenter and Riot Direct and edit it in here.


DDOS prevention. Something to look at maybe? RiotGradius
02-18-2014 10:33 PM
General Discussion

RiotGradius

Associate Information Security Engineer

02-18-2014
4 of 17 Riot Posts

Quote:
Originally Posted by FxE Baylife View Post
Who is your favorite champion?
Been playing a crapton of support recently, lots of Nami/Thresh. Then when I branch into Jungle, been playing Shyv and trying to learn others. And then there's Riven <3


DDOS prevention. Something to look at maybe? RiotGradius
02-18-2014 10:31 PM
General Discussion

RiotGradius

Associate Information Security Engineer

02-18-2014
3 of 17 Riot Posts

Quote:
Originally Posted by UnoDosMoltres View Post
You can't start your own thread?
It's been an idea, but I've decided rather than posting a blanket announcement, I would rather reply to multiple threads and try to explain as many aspects of the situation as possible, and address the concerns at multiple levels, rather than potentially driving the conversation myself.


DDOS prevention. Something to look at maybe? RiotGradius
02-18-2014 10:27 PM
General Discussion

RiotGradius

Associate Information Security Engineer

02-18-2014
2 of 17 Riot Posts

Yup, I've wanted to find the right post to start a conversation in... but so many of them are "RIOT FIX YOUR SERVERS" .. and I really wanted to have a dialog, rather than a storm of that sort of message.

I don't mind chatting with you all (I'd actually prefer it over staying quiet), and I'd really like to address any questions that I can. I'll be honest, it's a ****ty situation to be in, and myself along with a lot of other Rioters, as well as other industry professionals are trying to solve this problem as a whole rather than slowly crawling our ACLS up our provider lists. It really does make me sad that any of your games get ruined because of a DDOS, which is even more motivation to kick our butts into high gear and get this problem solved.


DDOS prevention. Something to look at maybe? RiotGradius
02-18-2014 10:21 PM
General Discussion

RiotGradius

Associate Information Security Engineer

02-18-2014
1 of 17 Riot Posts

Hello, just wanted to see if I could drum up some conversations about the nature of DDOS attacks and what kind of scale we're seeing. The attacks that have recently taken place on many different Internet services are quite large because of a very specific flaw in the NTP protocol. These attacks are called reflection attacks. (http://en.wikipedia.org/wiki/Denial-...Spoofed_attack).

What's difficult to deal with when it comes to DDOS attacks is that the larger the attacks become, the easier it is for the attacker to completely consume all bandwidth available for a specific provider. What this means is that any Internet service that this provider has under them is also unable to communicate back to the Internet. As the attacks become larger, we block them farther and farther up the provider levels, until we're working with Tier 1 ISPs to implement access control lists that prevent these attacks. Here's some more information from Cloudflare about the recent NTP Reflection DDOSes (http://blog.cloudflare.com/understan...d-ddos-attacks)

Us networking guys at Riot are not only working internally to find a solution to this problem, but we're also working with the Information Security industry as a whole to improve the situation. We're working to find a tech solution to block the attacks as they're ongoing, and fixing the underlying problem of open/unpatched NTP servers on the Internet.

Please feel free to toss any questions my way that you might have, I would love to have a chat with everyone about this and answer as much as I can.


Positivity Thread Riot Zephyreal
02-18-2014 02:29 PM
General Discussion

Riot Zephyreal

Player Support Specialist

02-18-2014
3 of 3 Riot Posts

Shoutouts to the Gragas who left laning phase 0-5 and his opponent Zed 7-0. He was super sad about it in chat and I simply told him, "just keep farming so you can two shot the ADC, don't worry about Zed." He stuck with the game and we won after he got a few items, in part due to some amazing ults in team fights!


Skarner Rework Follow Through RiotScruffy
02-17-2014 07:57 PM
General Discussion

RiotScruffy

Game Designer

02-17-2014
21 of 29 Riot Posts

I see a bunch of posts about a mechanic where a Q or auto could extend the E slow. This is actually pretty reasonable and is something that we could do. Realistically if we added this mechanic, we would not be able to buff the slow amount as much. In this case, as a Skarner player, which would you rather have?

-A stronger E slow like proposed with same 2.5s duration

or

-A (less than first option) stronger E slow that can be extended for additional time with Q or auto attack damage

I'm glad that we are having some good suggestions on ways to improve the kit and both of these mechanics could work out well.


"I always get bad teams" Riot Morgageddon
02-17-2014 01:22 PM
General Discussion

Riot Morgageddon

Player Support Tech Specialist

02-17-2014
5 of 5 Riot Posts

Morgageddon chips in.

Previous post has been updated.


"I always get bad teams" Riot Morgageddon
02-17-2014 12:29 PM
General Discussion

Riot Morgageddon

Player Support Tech Specialist

02-17-2014
4 of 5 Riot Posts

Morgageddon pops in.

So, I think a lot of this has to do with perception.

For instance, I remember those last three games. Only one of them did I feel like I was doing acceptable, when in reality, all was probably reasonable for playing as the least skilled member of my team. Why? Because as a human, you notices mistakes more than correctness.

Mistakes stick out. Take a look at a uniform pattern picture. Your eyes naturally are drawn to the one section where it's wrong, or you're already searching for it. Humans love to find things that don't fit, and we love to try and find order when sometimes there isn't.

Last season, I ended D1 at around 55 LP. Statistically speaking, that's the .1% or so of players, very close to Challenger but didn't quite make it. Would you be shocked to hear that after S3 ended, my overall win rate was 54.5%?

I win on average, 4 games more out of a hundred than an ideal 50-50 distribution. That's it. I only played 763 ranked games last season on Riot Morgageddon. I only won 31 games more than the ideal player. I can't assume that I was D1 because of that 31 game differential. I have to believe that I hit that peak of personal skill because of how I played on my team.

Some games I had to jungle, my worst role and just downright terrible for me. I never know when to gank, where to go, what to do. It just doesn't work that well with my mind. But my teammates knew. They called for ganks, they helped me with where and when I needed to be at objectives, and I felt like I actually learned how to better be a jungler because of it. Then there were the games where everything seemed to go wrong, and I didn't hear the end of it from my teammates. Do I remember the games where I improved? Not nearly as much as the ones where I remember how I botched Baron or got caught out 40 minutes in, leading to a Nexus rush and game over.

Doctor's visits. Most people don't remember them, unless something bad happens. They say, "we need to run some tests, you've got an abnormal looking mole," and your mind goes into full panic. If they say, "looks like you're okay, Good blood pressure this time," you probably won't remember that as...vividly as a negative response because you're expected to be fit and healthy. You're not expected to need a check for cancer.

So I'm going to wrap this long response up. I think a vast majority of players remember their negatives games more vividly than their positive ones. That 4-0-6 Orianna game feels amazing, and then the 1-5-2 Morgana feels crushing because you just couldn't get anything going. Humans love to feel good, and the negative feedback response you experience with loss and defeat makes you want to not see it again, hence you remembering/feeling it stronger than you'd expect.

Be nice to your teammates, play together, and most importantly HAVE FUN! If you focus on the good things that happen, your perception of the game will be aligned with things to work on, or things you saw that were good, not how painful the loss can be.


Skarner Rework Follow Through RiotScruffy
02-17-2014 11:32 AM
General Discussion

RiotScruffy

Game Designer

02-17-2014
20 of 29 Riot Posts

Ok cool, lets try out some CC buffs on PBE this week and see how he feels. I'm with you guys, the payoff for hitting the E is not there yet.


Skarner Rework Follow Through RiotScruffy
02-17-2014 11:18 AM
General Discussion

RiotScruffy

Game Designer

02-17-2014
19 of 29 Riot Posts

Quote:
Originally Posted by Hexalyte View Post
Tbh, it shouldn't be a super minor buff to each of them. When I said they both needed something, I meant they really needed it. Your proposed E changes are good and will definitely help for the early ganks, however for him to be able to keep up to make use of this DPS that you added to his kit, he needs a W change as well. Both abilities need a buff, not just a small value change to each.
Definitely, I'm mostly asking where we should put the power considering he will be getting some buffs.


Skarner Rework Follow Through RiotScruffy
02-17-2014 11:04 AM
General Discussion

RiotScruffy

Game Designer

02-17-2014
18 of 29 Riot Posts

Quote:
Originally Posted by Hexalyte View Post
It should be an even match up, just like you could with Kayle Q, then W, to slow, speed, slow speed. Currently, neither of Skarner's abilities are efficient enough for catching up and staying with the target.

His perma-slow was not well designed, but his current design doesn't allow him to do much at all in comparison to so many other super-mobile cc junglers.
Ok good point, maybe we could spit the changes up between E and W.


Skarner Rework Follow Through RiotScruffy
02-17-2014 10:59 AM
General Discussion

RiotScruffy

Game Designer

02-17-2014
17 of 29 Riot Posts

Quote:
Originally Posted by Genow View Post
now just make his W ramp up over 2 S, not 3
We could definitely put more power/reliability into W but it would mean less power in his CC. It's a totally valid proposal, would you rather see more power in his speed or CC?


Skarner Rework Follow Through RiotScruffy
02-17-2014 10:51 AM
General Discussion

RiotScruffy

Game Designer

02-17-2014
16 of 29 Riot Posts

Hey quick update about PBE changes. They wont be up today but should be on pbe later this week. This is the tentative changelist

-E slow increased from 30/35/40/45/50 to 52/54/56/58/60%
-E missile width increased slightly
-E missile speed increased slightly
-E mana cost reduced by 10

These changes are WIP so they may be adjusted after testing and feedback. The main problems we're trying to solve with these are:
-His early game is weak from both not having a strong 1 point ability and weak pre 6 ganks. 1 point in E should hopefully make him a better ganker pre 6.
-Soften up the difficulty of landing the E missile slightly to match the power of its effect
-Generally push his power back into a more CC focused space

These numbers may seem crazy, but as a first pass I think this will really help Skarner out in terms of viability.


"I always get bad teams" Zileas
02-16-2014 09:41 PM
General Discussion

Zileas

VP of Game Design

02-16-2014
3 of 5 Riot Posts

Quote:
Originally Posted by LiftsLikeGaston View Post
And I'm willing to bet they're rare.
Na. visible stats + invisible ability have to add up to total player ability... so I would assume that if you see someone with awesome stats in lots of games who doesnt seem to be climbing overall, they have bad invisible stats. In the reverse, someone who has hundreds of games played and seems to high tier vs the stats they are generating probably has above average teamwork skills. Half of the playerbase has above average teamwork skills at their level (and thus, below average mechanical skills -- otherwise theyd be at a higher level)


"I always get bad teams" Zileas
02-16-2014 09:38 PM
General Discussion

Zileas

VP of Game Design

02-16-2014
2 of 5 Riot Posts

Quote:
Originally Posted by Without Fury View Post
I always feel like these replies should also include " This is the best case scenario " because it is very rare that people just suddenly start listening mid/late game if in the early game they go 0-5. You can't even effectively roam that early in most cases.
There are players that have habitually bad KDA but manage to get to a tier -- those are people with bad mechanical skills and great teamwork, and they will roam correctly and listen ;p


"I always get bad teams" Zileas
02-16-2014 09:33 PM
General Discussion

Zileas

VP of Game Design

02-16-2014
1 of 5 Riot Posts

I agree with OP.

Furthermore, I think this is a common mistake because people look at their stats, and see that they have high stats -- or see that they win their lane, and then conclude the matchmaking sucks... When in fact they are comparing their stats, rather than comparing their total contribution to the team.

The LEAST visible parts of your performance are your teamwork, social skills with teammates, and midgame/late game decision making.... And these count for A LOT.

If you consistently win your lane and have better K/D/A than people playing the same position, but still lose over time and are stable at a ranked tier or an MMR rating... you must still be equivalent strength to those players you are with. So the thing holding you back from progress is actually YOU, and going from there, must be something other than your laning strength and farming ability -- so your ability to work with a team and make good mid/late game calls.

Put simply -- if you are platinum 3 at laning and mechanics, but bronze 3 at teamwork, you are still a Gold V player.


Skarner Rework Follow Through RiotScruffy
02-15-2014 08:55 AM
General Discussion

RiotScruffy

Game Designer

02-15-2014
15 of 29 Riot Posts

I'll put some changes up on PBE next week. Focus will be on increasing the focus and power of CC in his kit/role. Likely they will be tweaks, not new mechanics.


Skarner Rework Follow Through RiotScruffy
02-14-2014 10:01 PM
General Discussion

RiotScruffy

Game Designer

02-14-2014
14 of 29 Riot Posts

Quote:
Originally Posted by Magic Library View Post
I really appreciate you looking for feedback, but I feel like this quote isn't true. In particular I mean the "More dueling/fighting power" aspect. When I read this, I think "Oh, so top lane skarner type stuff". (Somewhat how Shyvana is a popular top laner because she is such a good duelist/fighter)...

In all reality, he was pretty good at that before. He definitely needed tweaks to make it more competitively viable, but it was something he could get away with... And it was quite fun.

The big thing that made him good at this was his sustain and sticking power when you were going to all-in him.

People have said "Oh you can get old sticking power, just get gauntlet, or FM or something" ...

...Well, I can do that on anyone... How is that anywhere near the same thing?

His sustain is also completely gone... Which sucks.... Put his slow on a longer CD...

It feels like there IS NO fighter/duelist concept behind it. When I play the new Skarner I feel like I am required to build full utility tank... Which sucks... Cause I miss "mess mofo's up" Skarner.

Thank you for posting this to get feedback. I sincerely hope you read this, and possibly even reply!

-Magic
Ya I'm hearing this loud and clear. We will be continually updating him until he as useful and powerful as other characters in his role.


Skarner Rework Follow Through RiotScruffy
02-14-2014 05:34 PM
General Discussion

RiotScruffy

Game Designer

02-14-2014
13 of 29 Riot Posts

Quote:
Originally Posted by NUTCRAKKO View Post
Yes, there should be SOME chance. A very SLIM chance that an unfed Annie should be able to beat out a fed Lee Sin.

That said, what does this have to do with you cannabalizing our favorite Scorpion?
It relates to the mechanics of Skarners old q that gave him fewer ways to fight from behind and his opponents fewer options to fight him with when he was ahead.

That said I do think we have shifted his power away from cc into more reliable dps which is not strengthening the fantasy for his core players.


Skarner Rework Follow Through RiotScruffy
02-14-2014 05:21 PM
General Discussion

RiotScruffy

Game Designer

02-14-2014
12 of 29 Riot Posts

Quote:
Originally Posted by Vacus View Post
Frankly? No.

People should be rewarded for being fed. If I'm super fed, and it gives me no significant advantage against my opponents, then why does it matter that I played well in the past?

Having an advantage over your opponent is your reward for outplaying them in the past. If you don't get any advantage, you end up with the bull**** anti-snowball mechanics that made League zero fun at the start of S4. (Those mechanics have since been nerfed and/or removed because it is NOT FUN to have your good play go unrewarded.)


All other things being equal, the guy with more items should win every time that he doesn't make catastrophic errors.
If there is no way for an opponent to possibly win against a character in a fight, at that point the game is effectively over. Allowing for both players to determine if a fight is won or lost is a core element to skill based competitive games. Items can give advantage to the winning player but there should always be some chance for the behind player to win.

We haven't succeeded in creating this balance with all characters but this is what we are trying to move towards.


Skarner Rework Follow Through RiotScruffy
02-14-2014 01:28 PM
General Discussion

RiotScruffy

Game Designer

02-14-2014
11 of 29 Riot Posts

Quote:
Originally Posted by Maulkrieg View Post
On the "whoever had more items wins" part of this discussion, Scruffy...with ADCs, whoever had more items wins in a 1v1.

Midlane/support is arguably the only case where skill starts to take a bit more precedence than items.

That's just how this game works.

If Renekton has THE SAME or more items as me top lane, he's very, very, very likely to win.

Same with Lee Sin. Are they all of a sudden not issues?
These are pretty good points and we acknowledge that many champions (including some listed by you) are not perfect. Over time we try to make improvements to the game that allow for more skill expression and player agency on both sides of a fight. Even if a fed Lee Sin engages in a fight with you, wouldn't it be cool if you still had a chance to outplay him for the win?


Skarner Rework Follow Through RiotScruffy
02-14-2014 10:28 AM
General Discussion

RiotScruffy

Game Designer

02-14-2014
10 of 29 Riot Posts

Quote:
Originally Posted by Maulkrieg View Post
Against assassins, whoever has more items wins, not whoever made the better plays. How are slows more toxic? At least your team has time to respond with slows.

Explain pls

Btw, THANK YOU SCRUFFY. YOU FINALLY PERSONALLY ACKNOWLEDGED AN IDEA. I believe that the secondary CC should be a brief hard CC though. You'll never have time to reach 3 Q stacks even with that secondary effect as hard CC.
In the case of assassins, we are actively working to not make it the case that when an assassin jumps on you, the determining factor for the win is items not skill (looking at you kassadin).

I don't think anyone thinks slows are toxic, the design philosophy is that when both players have the opportunity to make plays and change a fight it results in more engaging fights and a more skill based game overall.


Skarner Rework Follow Through RiotScruffy
02-14-2014 10:21 AM
General Discussion

RiotScruffy

Game Designer

02-14-2014
9 of 29 Riot Posts

Quote:
Originally Posted by Fisherman Fizz View Post
What I'd imagine after those changes (if you include the mana cost reductions) would be:

-He'll have a better skill to leave at one point (E obviously) so he doesn't feel as weak until he's leveled all 3 of them
-His W will still feel underwhelming. Even if it's numerically balanced, the slow ramp up will make it feel weaker than it really is to the person playing him. I think the ramp up time should either be made quicker, made to go in the opposite direction (getting slower over time) or just completely removed in favor of a flat movespeed boost so that the ability feels more satisfying for the person playing as him.
-Without access to more long term CC than only his E which has a long cooldown, he will still be missing a major part of his identity. I understand you think his perma slow on Q was toxic (which I don't agree with), but it was still one of the biggest things that made Skarner feel like Skarner, without something similar to that he won't feel like the same champion anymore. This absolutely needs to be fixed if you want old Skarner players to feel happy about the rework.
-In lane, he still has not been compensated for losing his heal. Please do not forget about his ability to lane when making your changes.


I have some questions, if you could answer any of them I'd really appreciate it:

-What about the slow on his Q made it toxic? Why was it considered bad for him to have this in his kit when other forms of long lasting slows like Red Buff, Frozen Mallet, Frost Shot, etc. also exist?
-How does his win rate with Smite compare to his win rate without it (or more specifically, as a jungler compared to as a laner)? Do the two win rates change at different rates as rating goes up?
-Have you thought about keeping the mark his E used to apply, but possibly adding some other sort of effect (example: Hitting someone with the mark on them could extend the slow for a bit longer)?
To answer your questions:
-Yes getting permanently slowed by his Q was frustrating to play against, but there are actually other gains that we were trying to achieve with the change.

Permanent slow in melee range often only created two situations for Skarner, the case where he catches someone and they can never escape and the case where he can't ever engage on his target. The winner of fights often revolved around who had more items, not who made the better plays. We hope to create more situations where Skarner catches a target in the first place, but more opportunity for escape one the target has been caught.

-I don't have the lane vs jungle data yet, but I will definitely see if there is a significant discrepancy for his lane case

-The mark with E and hit for a longer slow is a reasonable idea, but in that case we would probably just make the E slow last longer baseline to reduce the mechanics clutter.


Skarner Rework Follow Through RiotScruffy
02-14-2014 09:45 AM
General Discussion

RiotScruffy

Game Designer

02-14-2014
8 of 29 Riot Posts

This has been very useful feedback and looking at early data suggests that he's struggling much more in the early game than the late game. I'm tentatively looking to put together some changes for next patch:

-E missile width increase
-E slow values increased (especially at low ranks)
-Maybe mana cost improvements

A few things about these changes:
-We don't want to over buff him and have to nerf him again in the future, so keep that in mind.
-We want to make E a much better 1 point ability to help his early ganks
-We want to generally increase the CC payoff for hitting the E because of the difficulty/risk of landing it

These changes aren't final so please share your feedback.

-Scruffy


Skarner Rework Follow Through RiotScruffy
02-13-2014 01:37 PM
General Discussion

RiotScruffy

Game Designer

02-13-2014
7 of 29 Riot Posts

Quote:
Originally Posted by corals View Post
This is one of the better critiques, also skarner does feel different and that is a fair critique, player attachment is what makes the game go round.

I think they should lower the slow on his E slightly and every time some one attacks a target or slashes them another slow is added for 1 second. So he can slow a bunch of people slightly then pick a target to stick too, increasing the duration of the slow on that target by 1 second. 3.5 second slow is more up time and an answer to the cool down increase on E, particularly if it affords skarner an extra auto or two to decrease E cool down.
This is a good point, and we try to retain the core appeal of a character when they get reworked and in this case Skarner may have shifted a bit too far away from the CC tank role that he had before. Strengthening his CC seems like the right direction in the follow up to make Skarner closer to what his fans really loved about him before.


Skarner Rework Follow Through RiotScruffy
02-13-2014 09:24 AM
General Discussion

RiotScruffy

Game Designer

02-13-2014
6 of 29 Riot Posts

This thread is about follow up, and we are actively looking to improve the changes to make them something that Skarner players and all players are satisfied with.

The most frequent complaint I'm seeing is that a lot of the CC power that was core to Skarner's appeal is lowered, and that it's something that is core to his identity. So, with balance changes aside for the Skarner followup, this is something that we can try to bring back. His current pattern has more dueling/fighting power but less CC, but we can shift some of his power back into more impactful CC that will bring him closer to CC tank role that he had before.

Edit: if this is not a direction that you would like to see more of Skarner's power shifted into that is ok too, I'm collecting feedback.


Skarner Rework Follow Through RiotScruffy
02-12-2014 09:00 PM
General Discussion

RiotScruffy

Game Designer

02-12-2014
5 of 29 Riot Posts

Something that seems worth clarifying is that with the rework we wanted to introduce some windows where Skarner's opponent has the opportunity to escape, but we also have increased the power and versatility of Skarner's CC when he does use it. Skarner has a stronger slow that he can apply at range without having to build up on a target, when used correctly this can be a very powerful tool (especially when setting up for an ult in a team fight).

Further increasing the CC/Power spike when Skarner makes his play is probably the area that we would buff him in if we saw that he was undertuned.


Skarner Rework Follow Through RiotScruffy
02-12-2014 04:58 PM
General Discussion

RiotScruffy

Game Designer

02-12-2014
4 of 29 Riot Posts

I see a lot of concern over his early game power and just power level overall. This is exactly what we are going to address in the follow up patch. We are totally committed to making sure Skarner is viable if he is not.


Skarner Rework Follow Through RiotScruffy
02-12-2014 03:18 PM
General Discussion

RiotScruffy

Game Designer

02-12-2014
3 of 29 Riot Posts

Quote:
Originally Posted by mi ramfan View Post
I think what Skarner needs is the ability to cast Q while moving, because right now having to spam Q as well as autoattacking for damage just forces Skarner to stutter step everywhere and makes it next to impossible to chase anyone without the slow he used to have.

Just make Q and E able to cast while moving and see if that helps him. It certainly should make him able to keep up with champions like he used to without sacrificing DPS.
He is already able to cast Q while moving, but I have seen this suggestion for E from a few posters. I think that for a ranged slow like E, the cast time contributes to the risk/reward of the skill. A good Skarner that lands the E will catch up because of the payoff from the slow. This helps to differentiate skilled Skarners from the less skilled ones.


Skarner Rework Follow Through RiotScruffy
02-12-2014 03:16 PM
General Discussion

RiotScruffy

Game Designer

02-12-2014
2 of 29 Riot Posts

Quote:
Originally Posted by Ebonyjohnson View Post
Skarners early game feels bad. I think his W movement speed early levels should be increased and his E's cooldown reduced would make a worlds difference.
We only have a small amount of data, but it does suggest that he struggles in the early game. If he needs buffs, this may be a good place to look.


Skarner Rework Follow Through RiotScruffy
02-12-2014 03:04 PM
General Discussion

RiotScruffy

Game Designer

02-12-2014
1 of 29 Riot Posts

Hey guys,

We're really excited that the Skarner rework is now live, and we are playing him a ton on live, reading all of the forum feedback, and even watching a few Skarner streams. We are very passionate about him and we want him to be both a very healthy and fun champion for the game and a viable and balanced character.

These changes in 4.2 were targeted at improving his gameplay to accomplish a few goals:

-Enable more ability for Skarner and his opponents to display their skill
-Reduce the amount of feast/famine that Skarner can experience (when he's ahead he is unstoppable, when he's behind he has a hard time contributing at all)
-Create more defined windows of power for Skarner to create more interaction between him and his opponent

We think that the changes have accomplished these goals and we always try to release character in as balanced of a state as we can, but it is always a possibility that he is not balanced. The important thing here is that we are actively monitoring and planning on a follow up in the next patch for Skarner to make sure that he is balanced and viable.

Thanks again for all of your feedback, and we are always actively working to improve the game (and in this case Skarner). I'll come back to this thread to answer questions if you have any.

-Scruffy


Gotta hug 'em all! CaptnPanda
02-12-2014 12:15 PM
General Discussion

CaptnPanda

Player Support
Player Behavior Specialist

02-12-2014
12 of 28 Riot Posts

;D

-hugs-


Mods. I'm sorry. But this is more important than you realize. Poro'maw Picture. Riot Pwyff
02-11-2014 12:14 AM
Champ/Skin Concepts

Riot Pwyff

Player Communications

Follow Riot Pwyff on Twitter

02-11-2014
1 of 1 Riot Posts

While I can't in good faith make this a porority on the skins team, I'll at least make certain to let them know that we are trying very hard to shoehorn every champion into a poro skin.

Now someone make Poro'Gath.

Tha's all I got gang.


For the love....of SKINS Riot Nightblue
02-07-2014 03:01 PM
General Discussion

Riot Nightblue

Environment Artist

02-07-2014
4 of 4 Riot Posts

That dodgeball syndra. Riot please.


For the love....of SKINS RiotZeronis
02-07-2014 10:47 AM
General Discussion

RiotZeronis

Concept Artist

02-07-2014
3 of 4 Riot Posts

Wow these are really well thought out ideas!


Gotta hug 'em all! jaywatford
02-01-2014 10:18 AM
General Discussion

jaywatford

Lead Sound Designer

02-01-2014
11 of 28 Riot Posts

*Audio hug*


Maps Skin Ideas RiotForScience
01-30-2014 11:27 AM
Art Feedback

RiotForScience

Senior Environment Artist

01-30-2014
13 of 18 Riot Posts

Quote:
Originally Posted by Asatorrr View Post
winter SR, but with one change. POROS! POROS EVERYWHERE! POROS RIDING DRAGONS! POROS RIDING WOLVES! POROS SLEEPING IN BUSHES! POROS FALLING FROM THE SKY! POROS EXPLODING OUT OF TOWERS BECAUSE TOWERS SHOOT POROS AT YOU!

POWER THIRS-i mean-SUMMONERS RIFT: PORO EDITION!!!
I am not going to make any friends by saying this.. but I really think we need to keep poros just to ARAM. They work in that setting because of how much their cuteness contrasts with the grim darkness of the environment and play. In any other map they would not feel special.


Splash Discussion RiotSilver
01-29-2014 01:45 AM
Art Feedback

RiotSilver

This user has referred a friend to League of Legends, click for more information

Splash Artist

01-29-2014
2 of 8 Riot Posts

Dudes!

Sorry about the delay. I'm going to answer all of the questions I can and that the team is ready to answer. Feel free to continue posting feedback or questions. I will do my best to get back in here as often as I can to update.

----------------------------------------------------

Any word on old splash art for skins? eg: Nidalee and Master Yi.

For context. A while back around the time we reworked the Morgana/Miss Fortune/Tryndamere splashes, the splash team at riot reevaluated what our goals should be with all future splashes. We decided that if we were going to work on creating higher quality splashes, one of the things we would need to do is to start spending more time per splash. What used to be under two weeks (and in some cases, a day or two) has become 4 solid weeks of work per splash and while our team has grown to about 8 artists, this is still very limiting in the resources we have to make splashes. Because of this, we've had to prioritize what we are able to work on at any given time. Right now, our first and foremost priority as a team is to rework all of the base champion splashes that don't meet our standards along with new content. The main reason for this is that the base champions are both IP defining and also what the majority of our players see when they play the game. Once those are done we will prioritize skin reworks for the champions who have been relaunched where the skin splashes no longer accurately represent the in game model (champions like Trundle, Karma, and Nidalee) and after that we will work on the rest of the skin splashes but do not currently have a priority list defined for those. We are working on setting up a pipeline for working very closely with some of our trusted partners to begin working sooner on those relaunched skin splashes. No guarantees at this point, but we are constantly looking for ways to improve our process and get updated splashes for all of your and our favorite champions/skins as soon as possible. We've also been looking very hard to find artists (and passionate gamers) that are capable of hitting both the league of legends style and quality bar so that we can increase the rate at which we can achieve these goals.


Do you guys plan to replace the Chinese art you were putting in the game at one stage?

Yes this is an idea that was brought up a while ago and again very recently. We want to avoid communicating the idea that the splashes from the Tencent client, which we would theoretically substitute, would be a final solution for the current live clients in most other territories. This is because a lot of the older splashes created for the Tencent client no longer match the style of our new target. That being said, as a team we do agree there are probably splashes in our current client where the benefit of replacement might greatly outweigh the difference from our visual style. It's something we will explore.

Question regarding Pantheon's splash. In the Facebook page, as well as during the shifting portraits in the LCS, he uses a unique splash image that looks quite well. The one currently used is the one from the Chinese client years ago. What is the status on that splash, and what reasons were made for not using it in the game?

This splash was made a while back but we felt that it didn't accurately represent pantheon as a champion and what he is able to do in game. I would say however that the quality and personality of the splash overall is higher than the one currently live in game. We may revisit this one as a temporary replacement, since it is already done, until we can get to creating Pantheons actual reworked splash. I will talk to the team about it.

How much time is spent making ONE splash art?

For one person, 3-4 weeks depending on the artist.

How do you determine the position the champion will be in?

Our normal process is to make 3-5 quick black and white sketches exploring different compositions for the splash. We talk to writers, modelers and enthusiasts of the champion to find what we feel is the best version, pose and story for the splash. After we pick a direction we do a few color explorations and start to draw/render the final.

Is a splash art being "created" by one artist, or multiple?

Yes and no. Splashes are worked on primarily by one artist but we do regularly help each other to complete splashes based on each artists strengths and also to accommodate deadlines. On some cases, we have a large number of artists jump on one splash. (Elise's base splash was made in just under 4 days with 5 people working on it simultaneously.)


People point to Dragonblade Riven, the new Lux splash, the new Sivir splash, and others as being "humanly impossible to recreate". While I do see this and recognize the disjointed look that these art pieces give, I personally view these with a more open-mind because I give benefit of the doubt that there must have been some visual basis for these poses. If that is not the case, how did you decide on such postures?

You're right. I'm not sure if the team has previously highlighted our goals for the current target on splash art and if not, then that is probably my fault. We have a bunch of things that we want to aim for when creating new splashes and illustrations. For this question in particular, I'll talk a bit about character poses. The world of league of legends is a fantasy world where things and characters outside the realm of possibilities in the normal world exist. Furthermore, the game and stories are focused primarily on the champions, the super human characters of that world. Our champions have very exaggerated proportions, some can shoot fire from their hands, and some can jump 30 feet in the air. To be able to fully convey the force and impact of these moments, we employ a lot of the same principles used in comic books. We purposefully break anatomy to enhance the gesture of the pose and push characters into extreme perspective. The idea here is to communicate something that is not possible by a normal human. I will take some time in the near future to make a post where I deep dive into what we feel is our splash style. Right now the high level goals are: exaggerated comic book style character proportions and poses, high polish, but still hand painted rendering, limited color palettes with powerful accents, and dramatic lighting.

Are you guys responsible for splash-based animated login screens as well? if so how do you do it?

The illustration/splash team makes the 2d assets, separated out into Photoshop layers that the motion graphics team animates in programs like After Effects.

How do you guys feel about posting videos of the splash art creation process?

Unfortunately, around the same time that we reevaluated our splash style and timeline, we had to stop making process videos. Due to the larger file sizes, lag that's caused by the recording process, and the fact that we regularly collaborate on splashes, we found it unrealistic to continue to record the process. We are exploring new ways of opening the doors to our art process though. I'm not sure when things will be solidified, but its something regularly talked about at Riot HQ.

What other software are you familiar with for working on art? Which is your favorite? Do you use different software for different things?

Splash artists at Riot use Photoshop in tandem with other programs like Maya to create environment/prop models for reference, and a few internal tools. Photoshop is the medium that I've learned how to control the best, but Ive seen people make amazing works of art with MS Paint and Gimp. It all comes down to decision making rather than tools.


NA East Coast Server! Riot Foro
01-28-2014 03:59 PM
General Discussion

Riot Foro

Network Engineer

01-28-2014
4 of 6 Riot Posts

This isn't something we're ignoring - but it's not something that is going to happen overnight either. Can't divulge any plans right now but we're working on improving latency for everyone in North America (and of course, the rest of the world in their own regional enviroments).

PS @ O.P. - light travels at less than two-thirds of the speed of light through glass, and data through copper is even slower. Might want to readjust those calculations a little!


Maps Skin Ideas RiotForScience
01-28-2014 10:24 AM
Art Feedback

RiotForScience

Senior Environment Artist

01-28-2014
12 of 18 Riot Posts

I am pretty sure that map content is "Client side" which means that each player could pick their own map and see their own map as long as the gameplay information remains identical.