06/24 Emergency Server Maitance was not able to fix the problem

[Zyrusticae]

It's more than likely that the server hardware itself is inadequate to handle the sheer number of calculations being thrown at it at once.

This is not an easy problem to solve, mind you. For example, an often bandied-about solution is to increase the use of multithreading, but that requires that the code be programmed from the outset for parallelism - rewriting code to go from two, to three, to four threads is not at all a trivial task, and in some cases may simply be impossible (for example, in cases where data from one thread is dependent on receiving data from another, possibly causing stalling).

I have no idea how they handle this. It could be that each block is being handled by a single server shard, which would be a terrible idea and could be the cause of the heavy latency. But then, with 500 blocks out there, I can't say I'd blame them for taking that kind of route...

This is just speculation, mind you. I don't know the internal workings of Sega's server structure, I only know from experience with other, likely much smaller, server setups.

[Chik'Tikka]

It's more than likely that the server hardware itself is inadequate to handle the sheer number of calculations being thrown at it at once.

This is not an easy problem to solve, mind you. For example, an often bandied-about solution is to increase the use of multithreading, but that requires that the code be programmed from the outset for parallelism - rewriting code to go from two, to three, to four threads is not at all a trivial task, and in some cases may simply be impossible (for example, in cases where data from one thread is dependent on receiving data from another, possibly causing stalling).

I have no idea how they handle this. It could be that each block is being handled by a single server shard, which would be a terrible idea and could be the cause of the heavy latency. But then, with 500 blocks out there, I can't say I'd blame them for taking that kind of route...

This is just speculation, mind you. I don't know the internal workings of Sega's server structure, I only know from experience with other, likely much smaller, server setups.

+^_^+ I'd love to get a look at Sega's hardware+^_^+ I've been trying to get a peek at AT&T's local NOC for awhile, it's tucked behind this bar uptown in a squat gray building, but the things a vault (i heard rumors of water cooled servers in there, with all the cabling hanging from the ceiling and the water running underneath)+^_^+

[moeri]

I don't see the logic in how people think Text causes lag. I know you want an english patch badly but you can't just make up science. Text Files are so incredibly small.

hmm, though leaving all of that server side would atleast cause unnecessary stress on the servers

for example...

a simple quest description could easily be 1KB

lets say a million requests are made at the same time, this would be 1GB of data right off the start, that then needs to be transfered to the players

now, to tell the truth this is no big deal, the real problem is that with all the text on the server side, requests for the data need to be handled, which will put a lot more strain then the simple text files themselves, requesting the particular text, looking it up, handling the request, and such could easily be a big bottleneck, but they seem to have this handled...

putting the text client side would probably be a very good idea for them, but whether the actual source of lag is there, is shaky ground to stand on to say the least... I see no reson to keep the text server side, but meh, that is their choice

again, a bunch of text requests are probably nothing for sega's servers/ optimizations

[Polantaris]

Let's face it.

If they IP blocked the game from the start, they wouldn't have needed 3 maintanances already.

lolwhat?

Foreign connections don't cause lag. The servers handle requests the same regardless of source unless they implement something otherwise, which they have not.

[moeri]

lolwhat?

Foreign connections don't cause lag. The servers handle requests the same regardless of source unless they implement something otherwise, which they have not.

I" am pretty sure they mean that the servers would not have to deal with as many requests(as many people), therefore not causing a pile up of requests, causing lag

[Polantaris]

I" am pretty sure they mean that the servers would not have to deal with as many requests(as many people), therefore not causing a pile up of requests, causing lag

Except there's not that many foreigners playing when you compare to the amount of local players.

No matter how you want to look at it, foreigners are not the cause of the lag. Every single test has had serious lag issues (Minus possibly A1, I wasn't in that one), and in both A2 and CBT they admitted to some sort of server hardware issues.

[moeri]

Except there's not that many foreigners playing when you compare to the amount of local players.

Do you have numbers to back that up? If foreigners were even a small percentage of the population, then this might account for an unexpected number of people playing.

say... 10% would be enough to cause estimations of number of people playing to be far enough off to cause some issues such as the lag being experienced

Every single test has had serious lag issues (Minus possibly A1, I wasn't in that one), and in both A2 and CBT they admitted to some sort of server hardware issues.

Yes, but this could be due to an unexpected number of people? They really didn't limit the number of people that could play... atleast in CBT

"some sort of server hardware issues", could easily be a cover for, "wow we didn't expect this many people"

I am not trying to say that foreigners are the issue, but just knocking the possibility out of the way without proper testing/ analysis, would be a mistake

[Sizustar]

Do you have numbers to back that up? If foreigners were even a small percentage of the population, then this might account for an unexpected number of people playing.

say... 10% would be enough to cause estimations of number of people playing to be far enough off to cause some issues such as the lag being experienced

In order to make your statement true, you also need to provide evidence.

[moeri]

In order to make your statement true, you also need to provide evidence.

I am sorry let me correct my english

let us say... 10% would be enough to cause estimations of number of people playing to be far enough off to cause some issues such as the lag being experienced, maybe 10% of the population are foriegners

I am sorry, but with that example I was only trying to show how speculation can be thrown around, but without numbers to back up your speculations, they will forever only be speculations, I was only trying to show I could throw out a counter argument, as long as neither of us have hard numbers, both of our arguments are fairly weak and have no true value

Afterward, I went on to just reinforce the fact that a "possible" cause for a problem should not be dismissed without proper evidence, though I will say that this being that cause is very unlikely, but dismissing it entirely is the mistake.

[Zyrusticae]

Well, if we assume 90% of the foreigners on Ship 2 are playing in blocks 20->2x, right now we only fill up three and a half blocks.

The locals fill up all fifty.

So yeah, pretty sure we're a tiny, insignificant fraction. Furthermore, we do NOT play during peak Japanese hours (well, most of us, anyways). It would be disingenuous to even suggest we could be high contributors to this problem.