|
On March 27 2011 18:56 Paradice wrote:Show nested quote +On March 27 2011 18:18 MasterOfChaos wrote:On March 27 2011 08:23 Liquid`Jinro wrote:
When people talk about latency, it always brings to mind an article or quote I read somewhere, about how what matters the most is really not absolute latency, but rather stability. As long as you can get used to it, its not gonna be too bad for your game I think. I assume you're referring to the article about the AoE netcode: For RTS games, 250 milliseconds of command latency was not even noticed -- between 250 and 500 msec was very playable, and beyond 500 it started to be noticeable. It was also interesting to note that players developed a "game pace" and a mental expectation of the lag between when they clicked and when their unit responded. A consistent slower response was better than alternating between fast and slow command latency (say between 80 and 500 msec) -- in that case a consistent 500 msec command latency was playable, but one that varied was considered "jerky" and hard to use. 1500 Archers on a 28.8: Network Programming in Age of Empires and Beyond Damn it you beat me to it! I love that article. I would also recommend, if anyone is actually interested in learning about how different genres of games communicate with each other (hopefully to shut people up who have been comparing SC2 latency to Counter Strike ?!?), read this article: (the Peer To Peer Lockstep model is relevant for SC2): What every programmer needs to know about game networking
You see, this is how SC1 worked, and latency was the same for both players. If the same applied for SC2 you would have the same latency as well. The idea that is being put forward is that the overseas player has more latency than the local. So SC2 does NOT use the peer to peer lockstep! The funny part is that this model is by far the best I can see for using in multiplayer RTS. So Blizzard must have gone to great lengths to implement something different for the purpose of discouraging cross-regional play.
|
United States33406 Posts
I don't think this is in any way the final word on the NA-KR lag issue, but it's good to have a balancing opinion from someone who says KR-NA is definitely playable against those who claim it's some incredible disadvantage that makes master leaguers lose to people in silver.
The truth is, we don't know exactly what kind of conditions the players were playing under. We know NA-KR lag can range from quite playable to downright horrendous, but trying to guess where it fell through VODs of a game is fruitless and pointless speculation.
What I can say though, is that TSL players were offered the opportunity to delay their game and play on a different connection if they found the lag to be unplayable. As a last resort, we offered anyone to play at the oGs-TL house where we had confirmation from TL players that the lag was definitely in the playable range. Of the Korean players, only Nada (who does not normally reside at the oGs house) chose to exercise this option. Everyone else showed up, and just played their game.
|
guess that means nada wins today :o
|
Thanka a lot Jinro, great post
|
Imo people are underestimating the perfomance drawbacks some slight lag can have, let alone a whole second delay. Did someone say 2 seconds delay as tweeted by Boxers wife? I'm not surprised as if you watch the replays you can see obvious reaction delays that normally shouldn't be made by pros. Not to mention Boxer got disconnected, didnt exactly give me a good impression of stable connection.
Hell, a half second delay 500 ms latency is killer for perfomance.
SC2 is a fast game. Its not Call of Duty, but fast in a different way. Reaction times are just as important.
Teamliquid will do their best to underrate the effects of lag because otherwise this whole tournament will be considered useless wich is bad PR for Teamliquid. And I understand completely they have every right to protect the games.
|
So what you are saying is that to play with lag u need to practice with it. And i doubt the koreans would bother practicing with lag only for TSL. So, since they most likely never practiced with lag, lag was a deciding factor.
|
On March 27 2011 23:57 Nizaris wrote:
So what you are saying is that to play with lag u need to practice with it. And i doubt the koreans would bother practicing with lag only for TSL. So, since they most likely never practiced with lag, lag was a deciding factor.
If they don't bother practicing for TSL don't you think that's a problem?
|
Germany25657 Posts
Its their own fault then, they agreed to play i a tournament and did not prepare for it accordingly. TL even told them about possible lag, so yeah, maybe next time they will come prepared :D
EDIT: basically what redmark said ^^
|
I haven't read through the entire thread so this may have been mentioned but Boxer's wife or girlfriend or w/e mentioned that Boxer was playing with 2000ms in his games against Nightend and that's no small little difference that you can get used to. I assume that she may have been talking about it spiking that high but honestly I don't really know for sure.
I play for Team EG for another game so I've done a lot of LAN / Online E-Sports type tournaments and as Jinro said you can get used to latency worse than normal by playing on it, but what you can't get used to is inconsistent latency. If someones latency is spiking to 2000 during a match if it occurs at an important time even if it's only bad for 1-2 seconds it can easily decide the outcome of a match whether it's lining up a head shot or microing mid-battle.
The last thing I'll mention is that from talking to Korean players at LAN events and trying to play with them online their latency is so insanely good in Korea that even the slightest delay felt unplayable and disgusting to them. My one Korean friend said he got around 5ms at home in most games and the thought of playing with 100 was unthinkable to him so they didn't even want to try. So all in all the latency might not be that big of an issue for those of us used to playing with it but little things can really get in your head while playing in tournaments, I've seen people not be able to play right because they're sitting a few CM lower than they're used to or some other tiny little thing doesn't feel right and it gets in their head. So if that's the case with the Koreans then it's really unfortunate because it can totally throw you off.
That said the NA / EU players played better in those games and the Koreans likely didn't prepare simply because like everyone else they assumed it'd be easy wins. If there was massive ping spiking then it's hard to say who would've won but if it's just a bit of latency then obviously the people who won deserved to it's the Koreans fault for not preparing. Either way it kind of sucks because it's lose / lose for the tournament and the players since so many people end up blaming results on lag it discredits the players and the tournament and who's to say that they wouldn't have won those games against the koreans playing at a LAN?
|
In WC3 if I would play some games on LAN latency and then play ladder, my first few games would always be awful, just because I'd wrongly anticipate some tricky micro timings. If you're not prepared to deal with them, the first sign of lag causing you to play bad enforces a bad mindset where you're not focusing on the games. I imagine this is why the Korean players were all playing below potential and making mistakes that don't seem to be caused by lag: it's because their focus was off due to being distracted.
|
I remember when I used to play DotA with my old crappy computer, I had 1-2 seconds delay before the command was executed. The more and more and I played, I started to be more and more accustomed to the lag, until I totally forgot about his existence. One day, a friend came over and we played a couple of games. He was never able to perform well due to the delay, even tho I didn't felt any.
When I switched computers, I almost didn't notice a difference either, except that all my actions were a few miliseconds ahead. I barely noticed it. I would say that if the koreans were not accustommed to the lag, it is their fault. If they never experienced it and start playing a tournament game... well blaming lag for your loss seems like a good option.
|
This two second lag that keeps getting brought up is an obvious exaggeration, the game would freeze to sync if there was really that much latency involved. While I can't speak for every player, I know the connection from New York to Korea is 240ms. Bnet US is in California, so subtract 70ms from this for approximate good condition Bnet US to KR latency. Please stop with the hyperbole.
Regarding the earthquake that people think has affected connections, it is true that some routes were affected and the re-routing was less than optimal, causing 600+ms latency over affected routes. To the best of my knowledge, the affected routes were only affected for a period of several days shortly after the quake, during which time no TSL games were played. In addition, 600ms is borderline unplayable and would have hopefully been brought up during any testing.
|
Sweden33719 Posts
On March 27 2011 23:56 Paincakez wrote:
Imo people are underestimating the perfomance drawbacks some slight lag can have, let alone a whole second delay. Did someone say 2 seconds delay as tweeted by Boxers wife? I'm not surprised as if you watch the replays you can see obvious reaction delays that normally shouldn't be made by pros. Not to mention Boxer got disconnected, didnt exactly give me a good impression of stable connection.
Hell, a half second delay 500 ms latency is killer for perfomance.
SC2 is a fast game. Its not Call of Duty, but fast in a different way. Reaction times are just as important.
Teamliquid will do their best to underrate the effects of lag because otherwise this whole tournament will be considered useless wich is bad PR for Teamliquid. And I understand completely they have every right to protect the games.
There isnt even a 2 second delay from KR to EU -_________- Hell, the EU delay isnt even 1 second so wtf.
People need to stop putting up such ridiculous numbers, because numbers that big are actually physically unplayable and nobody would agree to a tournament under those conditions.
|
I am just really amazed Jinro took the time to post and address all the comments that others have been posting. From what I have read in online articles regarding the tremendous amount of practice time pro-gamers spend everyday perfecting their skills, I am greatly impressed by Jinro's efforts at helping everyone understand this latency issue. Much thanks!
Edit: Sorry I have nothing else to add because they have all pretty much been answered by reading the previous posts in this thread.
|
this forum has been Jinrolled once again! great post. I sure hope its gonna cooldown some trolls/rager/hater/ppl who dont know what they are talking about. its always good to have an opinions from the inside. ty Jonahtan and best of luck in next GSL!
|
It's a sad realisation that still in 2011, there is latency to magnitude that people consider it the biggest contributor to the results, it simply shouldn't be this way.
There's little that can be done on the lower levels (i.e. the people affected by this), other than preparation for the match, knowing that the Koreans would be required to play on different servers, and having the privilege of flawless connections prior to this, they should have become accustomed to some form of latency before playing.
But, you really cannot blame them, because they shouldn't have to prepare for latency in the first place, it's pretty ridiculous. And it is true what people are saying, even the slightless fluctuation in connection can really throw a player off, it's constantly irritating you, always in the back of your mind that the connection is playing up.
That's not to take away from the foreigners in the TSL though, atleast, most of the games, because they merely just played alot better than their opponents, especially during the games we saw last night. Though it does dis-credit the victories slightly, knowing that this controversy is going on, for people will always be wondered and saying to themselves: "Well, would this same result be the case if there were no lag?", to which we just cannot answer that question.
TL;DR, Battle.net 2.0 fails, other countries with stable economies need to invest far more into improving their high-speed internet, to atleast compete with the speed over in Korea. This will happen in 4-6 years or something, unfortunately.
|
Dominican Republic913 Posts
+ Show Spoiler +On March 27 2011 13:55 usethis2 wrote:Show nested quote +On March 27 2011 11:03 Maurader wrote:I'm going to get banned again, but this post is worth it. I love jinro and am a huge fan, but jinro knows what effect lag has on the game as evidenced when he played vs select in the gcpl. These are the jinro quotes after that game, when lag wasn't as bad as it was pre earthquake. Jinro doesn't know how bad the lag was for these players, since he wasn't actually sitting with them when they played, but does know the frustration it causes to himself to deal with lag. Also tyler knows how lag effects the game as well cause shortly after the earth quake he was playing on korean server. He was in a base trade with a zerg and lost and then even he acknowledged that the lag was terrible and the reason he lost. http://www.teamliquid.net/forum/viewmessage.php?topic_id=189994¤tpage=28#548On February 04 2011 16:03 Liquid`Jinro wrote:
Microing vikings in lag is a fucking joke -- http://www.teamliquid.net/forum/viewmessage.php?topic_id=189994¤tpage=29#564On February 04 2011 16:19 Liquid`Jinro wrote:On February 04 2011 16:10 gaston116 wrote:On February 04 2011 16:03 Liquid`Jinro wrote:
Microing vikings in lag is a fucking joke -- What did you think of SjoWs play? Uhm, he played fine of course? Its not like I went into either game expecting anything else. Select played great, but its fucking annoying to play vs the way he plays when it lags this much. Its like, in perfect situations his playstyle is annoying to deal with, then you add lag and it makes you want to kill yourself. http://www.teamliquid.net/forum/viewmessage.php?topic_id=189994¤tpage=29#574On February 04 2011 16:30 Liquid`Jinro wrote:On February 04 2011 16:27 Ezze wrote:On February 04 2011 16:10 gaston116 wrote:On February 04 2011 16:03 Liquid`Jinro wrote:
Microing vikings in lag is a fucking joke -- What did you think of SjoWs play? Ouch man. But yea I kind of agree. I didn't see any lag issues in the first game. I would probably give some credit to SeleCT there Jinro instead of blaming the loss on lag. Lol, why do you think it took THREE scans to kill 1 banshee in the first game? That doesnt mean I think they are bad, but theres just no way Im going to beat a player who is as good as me in this lag -- Im not gonna lie and say "oh the lag was fine", that doesnt mean they didnt play good and might very well have won anyways.I can also mention that a big reason I won game 1 was because my hellions luckboxed and saw his dropship going out. Making excuses sucks, it makes you look terrible but I m too frustrated to not bring it up. Yes, I knew it was gonna lag when I agreed to play but that doesnt change anything about how you feel after games like these.
User was banned for this post.
Why is this guy banned? He simply quoted someone else's posts (which were not deemed ban-worthy, obviously). Confused here. Is TL so closed-minded that it doesn't want to hear anything unpleasant, even though it's presented in civil manners? Not that I think his post is most polite one nor agree with what message his post presumably intends to convey, but the free speech means nothing if you simply block what you don't want to hear.
This is the most Beautifull post i ever read in TL
|
Braavos36375 Posts
2GRe-Play-, usethis2 was not banned, Maurader was banned for martyring.
Relevant post:
On March 27 2011 14:15 Hot_Bid wrote:Show nested quote +On March 27 2011 14:05 usethis2 wrote:On March 27 2011 13:59 Myles wrote:
Martyring yourself - aka saying anything like 'this will get me banned' - will always get you banned here regardless of what else is in your post. Interesting. Is it in the ToS (Term of Service) of this forum? Yes. Show nested quote +TL 10 Commandments
Don't be a drama queen martyr. TeamLiquid isn't colonial America or Tienanmen Square. You do not need to preemptively declare "I'm going to be banned for this..." or "ban me, I don't care." Our Moderators ban for good reasons. If they read someone acting like a martyr, they'll simply treat it as a request to be banned. Martyring could be a huge problem for our mods because it's a lose-lose situation. If you ban martyrs it looks like they were right about your "biased" moderation. If you don't ban them then you encourage more martyring and passive-aggressive jabs at biased moderation. So we simply remove the choice, and state that every martyr is an auto-ban regardless of what they say. If you offer constructive criticism we always try to receive it fairly.
Summary: usethis2 asked why Maurader was banned, I explained it had nothing to do with his criticism but rather breaking the martyring rule.
|
On March 27 2011 23:57 Nizaris wrote:
So what you are saying is that to play with lag u need to practice with it. And i doubt the koreans would bother practicing with lag only for TSL. So, since they most likely never practiced with lag, lag was a deciding factor.
If a tennis player always played on a grass court, then entered a tournament which used a clay court, the different ball bounce could be a deciding factor. Doesn't change the fact that the player is at fault for never practicing on a clay court.
|
![[image loading]](http://www.pingtest.net/result/37716012.png)
400-450 ping is fine, though borderline, for an RTS.
An FPS, on the other hand...
|
|
|
|
|
|
EPT
