| Pages: 1 2 3 4 [5] 6 7 8 9 :: one page |
| Author |
Thread Statistics | Show CCP posts - 6 post(s) |

Hugo Splat
Myrmidons
|
Posted - 2007.05.16 11:49:00 -
[121]
Originally by: Matthew
Originally by: Berek Bladesong CCP, people are tired of your lack of comittment to complete a job. The fact that you still rely on a daily downtime to resolve time related instability many years into the game is something you expect in the beta testing stages.
The DT would still be needed even if there were zero time-related instability. Lots of database maintenance activities and high-DB world-maintenance activities (e.g. refilling all the roids) are done during DT. Sure, they could be changed to run in the background of the live environment, but that would reduce the performance of the live environment, and lead to undesirable in-game effects such as people getting stuck between roids when the previously-popped belt suddenly reappeared around them.
There never used to be a downtime, yet roids still respawned...
|

OneSock
Silentia Mortalis
|
Posted - 2007.05.16 12:10:00 -
[122]
Originally by: Victor Valka
Originally by: OneSock Time to implement a pop cap per system ?
Time to stop acting like game/networking/database/hardware solution/programming expert and shut up? 
I've worked in IT systems long enough to realise that sometimes there is no point flogging a dead horse. CCP probably don't have an unlimited budget and while I have every confidence that they have the right people and ability to look into these lag issues, if the bottom line is lack of horse power and the funds are not there to provide it, your not going to get far.
I've been in the game since sept 2005 and since then lag has always been a major issue and despite new clusters and resources, the userbase just expands to fill it. Unpalatable as it may be to many I would rather have a pop cap imposed on each system to keep it playable, rather than just be stuck in a lag blackhole.
|

ry ry
StateCorp
|
Posted - 2007.05.16 12:24:00 -
[123]
i've had no problems logging in at work, where we have the proverbial fat pipes, but logging in at home (where my connection is a comedy throwback to the dawn of telecommunication) is a nightmare.
That suggests to me that connection speed plays a big part in the problems.
Interestingly, i've also noticed i have more trouble logging in after DT and early evening when everybody's trying to get in. are lower connection speeds are being deprioritised by the login server or something?
|

Solbright alt
|
Posted - 2007.05.16 12:29:00 -
[124]
Originally by: OneSock I've been in the game since sept 2005 and since then lag has always been a major issue ...
Expand on this please. Define symptoms of lag.
|

OneSock
Silentia Mortalis
|
Posted - 2007.05.16 12:57:00 -
[125]
Originally by: Solbright alt
Originally by: OneSock I've been in the game since sept 2005 and since then lag has always been a major issue ...
Expand on this please. Define symptoms of lag.
Ever been to Jita ? 
1 .Undocking do a black screen because of the lag loading all the ships outside station ?
2. Modules taking 10 seconds to activate in a busy mission hub ?
3. Warping into a mission and waiting while the grid loads, then finding your already down to half your armor ?
4. Of course the major one that pops up all the time is jumping into another system to engage an enemy fleet but once your loaded you find your in your pod/in clone station.
5. etc.
The first 3 examples are livable with to an extent, although I've almost lost ships to 2 & 3. Where as 4 is kinda game breaking really and CCP won't even replace your ship in that situation, which is why I avoid it like the plague.
|

Victor Valka
Caldari Archon Industries
|
Posted - 2007.05.16 13:03:00 -
[126]
Edited by: Victor Valka on 16/05/2007 13:02:03 Edited by: Victor Valka on 16/05/2007 13:01:40
Originally by: OneSock
Originally by: Victor Valka
Originally by: OneSock Time to implement a pop cap per system ?
Time to stop acting like game/networking/database/hardware solution/programming expert and shut up? 
I've worked in IT systems long enough to realise that sometimes there is no point flogging a dead horse. CCP probably don't have an unlimited budget and while I have every confidence that they have the right people and ability to look into these lag issues, if the bottom line is lack of horse power and the funds are not there to provide it, your not going to get far.
I've been in the game since sept 2005 and since then lag has always been a major issue and despite new clusters and resources, the userbase just expands to fill it. Unpalatable as it may be to many I would rather have a pop cap imposed on each system to keep it playable, rather than just be stuck in a lag blackhole.
I understand what you are saying, but I'd rather be stuck in the black hole of lag then be 155th in line for jump-in to a system with 6 gates.
Also. DDoS comes to mind.
"Don't want the opposing force from attacking your POS/Outpost?"
"No problem! Keep the system full of your guys!"
Population limit will never work in EVE - it's too dynamic and complex for its own good.
Originally by: Diana Marc Notice that BoB is agreeing with RA's concern. That's like Elrond and Sauron agreeing to reduce carbon emissions.
|

OneSock
Silentia Mortalis
|
Posted - 2007.05.16 13:24:00 -
[127]
Originally by: Victor Valka
I understand what you are saying, but I'd rather be stuck in the black hole of lag then be 155th in line for jump-in to a system with 6 gates.
Also. DDoS comes to mind.
"Don't want the opposing force from attacking your POS/Outpost?"
"No problem! Keep the system full of your guys!"
Population limit will never work in EVE - it's too dynamic and complex for its own good.
As opposed to the current situation where the enemy fleet just pile on until the node crashes, all the defenders cannot log back in to defend, but the enemy fleet can still jump people into the system from other nodes ?
pretty ****ty huh ?
I accept your point of view (if that's how you want to play) but don't necessarily agree with it. The issues you mention could be worked around. For example, sovereignty could play a part:
Popcap limit of 100 defenders having sovereign rights, plus 100 of a declared alliance/corp, plus a handfull of independants. These allocations could be dymanic to an extent but overall the popcap limits the system to what the hardware is capable of.
|

ry ry
StateCorp
|
Posted - 2007.05.16 14:51:00 -
[128]
my eve experience today...
downtime finishes at 12, so i start trying to log in. it's 2:45 before my connection is finally accepted.
i decide to do a couple of missions to kill time. warp in... lag... unlag deep into armour... try to warp out... in a pod.
close eve. watch a dvd.
i'm not fussed about the Hurricane and i'm not overly bothered to lose 3 hours of play time, but it's a bit crap that the game is completely unplayable recently. hurry up and fix whatever it is you broke!
|

Barbaro55a
Caldari Os Lobos Ethereal Dawn
|
Posted - 2007.05.16 15:18:00 -
[129]
Jumped into zz5 last night to kill a gate camp, lag blacked us out. Finally got back in atfer 2 hours to find an empty system. Luckily the gate campers had gone too. .......................................... We come in peace - My sig and your sig would make cute bay sigs. |

Miki Fin
Gallente Independant Union of Rangers
|
Posted - 2007.05.16 15:39:00 -
[130]
seems a lot of people just got dropped from metropolis 
|

Adaris
Dark and Light inc.
|
Posted - 2007.05.16 15:42:00 -
[131]
and 'other' areas * * * http://www.youtube.com/watch?v=euZ0j7vtKEQ
|

Callthetruth
Caldari Logical Logtistics
|
Posted - 2007.05.16 16:17:00 -
[132]
entire game atm i wish bob and others would finish the lag inducing war thats going on atm, its not like we are even at record numbers 34,000 concurrent users has dropped to 28000 at peak time
|

aliasdestructo
|
Posted - 2007.05.16 19:04:00 -
[133]
i say stop the trial acc system and clear game of all nubber ships floating all over the place *snip* Don't use offensive or deragatory language. -Rauth
trial accounts not likely to go away...and they shouldn't. "try before you buy" is a good marketing tool and a very old one. still, if eliminating abandoned rookie-ships would help reduce server load...why not eliminate even more clutter, in the form of ad-cans? there are some systems so filthy with ad-can clutter, that finding a place to launch a can for mining is a search unto itself. lets reduce the ad-cans...whether it helps with the lag or not!
|

Victor Valka
Caldari Archon Industries
|
Posted - 2007.05.16 20:32:00 -
[134]
Edited by: Victor Valka on 16/05/2007 20:30:01
Originally by: OneSock
Originally by: Victor Valka
I understand what you are saying, but I'd rather be stuck in the black hole of lag then be 155th in line for jump-in to a system with 6 gates.
Also. DDoS comes to mind.
"Don't want the opposing force from attacking your POS/Outpost?"
"No problem! Keep the system full of your guys!"
Population limit will never work in EVE - it's too dynamic and complex for its own good.
As opposed to the current situation where the enemy fleet just pile on until the node crashes, all the defenders cannot log back in to defend, but the enemy fleet can still jump people into the system from other nodes ?
pretty ****ty huh ?
I accept your point of view (if that's how you want to play) but don't necessarily agree with it. The issues you mention could be worked around. For example, sovereignty could play a part:
Popcap limit of 100 defenders having sovereign rights, plus 100 of a declared alliance/corp, plus a handfull of independants. These allocations could be dymanic to an extent but overall the popcap limits the system to what the hardware is capable of.
Okay.
The way I see it is like this:
People complain about lag because it interferes with their gameplay - their enjoyment of EVE experience - what we are all here for.
People will complain about population cap per system/constellation/region because it will interfere with their gameplay, just in a different way.
Both of these are undesirable meta game-mechanics in a sense that these are not really part of the game world but are there regardless due to the situation outside of it [game world].
Originally by: Diana Marc Notice that BoB is agreeing with RA's concern. That's like Elrond and Sauron agreeing to reduce carbon emissions.
|

Thargat
Caldari S-44 Coalition of Carebear Killers
|
Posted - 2007.05.16 22:17:00 -
[135]
I just don't get it. I read about people having a hard time logging in. People being lagged when running missions and being thrown out from the server. I've NEVER experienced that kind of problems, atleast not at the scale that some claim. Once or twice a month I get a disconnect or get really bad lag (and I can't honestly say that it's been CCPs fault with any degree of certanty). I play EvE for 4 to 8 hours every day on mixed times (2 hours before DT and 4 to 6 hrs at EvE evening time) with several characters from the North, through Jita down to deep 0.0 in the south. I go roaming 40-60 jumps about once a week and I participate in fleetbattles and smaller skirmishes and very rarely have I been in any serious LAG (last occation was when fighting PA in Empire and then it got a bit "slidshow" -ish.
Do some serious tests (log on SiSi and see if you have the same problems there) and THEN come back and whine.
If you don't like what I say, just ignore it, or admit I got to you. |

Hantsuro
|
Posted - 2007.05.16 22:40:00 -
[136]
I just got severe lag. So bad I couldnt even warp out of the mission.
|

Kruel
Beyond Divinity Inc
|
Posted - 2007.05.16 22:46:00 -
[137]
Originally by: Thargat
Do some serious tests (log on SiSi and see if you have the same problems there) and THEN come back and whine.
In other news, Thargat's farts don't stink.
|

Solbright alt
|
Posted - 2007.05.16 23:51:00 -
[138]
Originally by: OneSock
Originally by: Solbright alt
Originally by: OneSock I've been in the game since sept 2005 and since then lag has always been a major issue ...
Expand on this please. Define symptoms of lag.
Ever been to Jita ?
Lots. It's lag free. At least so close to it it don't matter.
Quote: 1 .Undocking do a black screen because of the lag loading all the ships outside station ?
I think that might be the client losing the plot. Prolly via packet loss. Normal time is only a few seconds. I relog if it goes blank for more than about 20 seconds.
Quote: 2. Modules taking 10 seconds to activate in a busy mission hub ?
Agreed, that is lag.
Quote: 3. Warping into a mission and waiting while the grid loads, then finding your already down to half your armor ?
That's not lag. That's stutter. Your client is frozen struggling with the all the input from the server.
Quote: 4. Of course the major one that pops up all the time is jumping into another system to engage an enemy fleet but once your loaded you find your in your pod/in clone station.
Again, stutter. You display is frozen. If this was lag then everyone would be in the same boat waiting for the server.
Hmmm, there is one more possibility. You have a connection problem. This can cause lag and packet loss. And it'll affect you but no others. This of course is not a problem with the cluster either.
|

Ender Ti'kai
|
Posted - 2007.05.17 09:43:00 -
[139]
One extremely simple thing that could to vastly increase enjoyability be done is the option to drastically decrease texture quality and a much lower resolution setting! In fact, an option to remove the images of ships from display altogether would help many people out a LOT, especially with larger and larger numbers being seen in major, decisive combat. Please, please, please implement this CCP!
|

Thargat
Caldari S-44 Coalition of Carebear Killers
|
Posted - 2007.05.17 11:33:00 -
[140]
Originally by: Kruel
Originally by: Thargat
Do some serious tests (log on SiSi and see if you have the same problems there) and THEN come back and whine.
In other news, Thargat's farts don't stink.
Kruel made a funny 
oh btw, your wrong 
If you don't like what I say, just ignore it, or admit I got to you. |

Matthew
Caldari BloodStar Technologies
|
Posted - 2007.05.17 12:09:00 -
[141]
Originally by: Solbright alt Won't happen. Emergency warps need a trigger. Non-activity from the remote end is not a good trigger in an event system. What might be a trigger is a low-level network connection loss. Dunno, this sort of detail is beyond of my knowledge. Another possible trigger is the server node overloading internally triggering auto warp.
I'm not talking about non-activity in an event system, obviously auto-warping you if you're sitting there doing nothing is silly. What I'm talking about is when the server sends the client something, expects a response, and doesn't get one. Even if the event being sent does not require a game-related response, the server will still be expecting an ACK response from the client for the packet of data that's been sent. In that situation, the server has no choice but to assume that it either did not receive the client's disconnect packet, or that the client has lost connection in a way that did not generate that packet. At which point it makes perfect sense for the server to close the connection and trigger the on_disconnect events (e.g. emergency warping).
Originally by: Solbright alt I'd say this goes both ways as neither the client or server are looking for a response to their respective events.
As I've said above, they may not require a game-related response to their events, but they will always require a network-related response to acknowledge the receipt of the packet. If those network-related responses are not being received, then there is no point trying to maintain game-level events over the obviously broken network pipe.
The reason the client and server behave differently when the connection breaks is because of how they behave. The client only sends events to the server when you do something with the client that requires it. If you've left the client afk not doing anything, it's not going to be generating any events. If it's not generating any events, it has no way of knowing if receiving no events is due to none happening or the connection being down, so stays open until it generates an event that is not responded to.
The server is much more proactive in sending out events to the client. If you're sitting in space it will send you position/velocity etc updates very regularly. Even if you're sitting in station afk, anyone entering or leaving local, the station guest list, any other chat updates, wallet/evemail flashes etc, will all generate an event to be sent to the client. This means that you are very unlikely to go any significant length of time without the server trying to send your client some sort of event. And if the server disconnects you due to no response, there's not much point it trying to send a disconnection event when it's disconnecting you on the assumption your connection has died anyway.
Technically you are right, both the client and server have the potential to not spot a connection loss immediately. But because the server sends events so much more often, the server is likely to spot the lack of response before the client does, especially an afk client, which is why you'll end up with zombie behaviour at the client end much more than on the server end.
e.g. 16:59:00 You go afk at a safespot 17:00:00 Internet burps and you lose connection 17:01:00 Someone jumps into the system, server sends you update to local list. 17:01:10 Server re-sends event due to lack of packet ack from client 17:01:30 Server assumes connection loss as no ack from client. Server drops connection and autowarps. Client still shows as online as has not received anything to suggest otherwise. 17:30:00 You come back from being afk and try to warp somewhere 17:30:10 Client has not received ack for the event packet, so resends 17:30:30 Client has not received any ack from server, so assumes connection loss and closes. ------- There is no magic Wand of Fixing, and it is not powered by forum whines. |

Matthew
Caldari BloodStar Technologies
|
Posted - 2007.05.17 12:11:00 -
[142]
Originally by: Hugo Splat There never used to be a downtime, yet roids still respawned...
When did there not used to be a downtime? I've been playing for several years and there's always been the regular daily downtime.
Besides, I didn't say that things like roid respawns couldn't be done live, just that there are reasons why it may not be desirable to do them live. ------- There is no magic Wand of Fixing, and it is not powered by forum whines. |

ry ry
StateCorp
|
Posted - 2007.05.17 12:31:00 -
[143]
Edited by: ry ry on 17/05/2007 12:29:56
Originally by: Thargat I just don't get it. I read about people having a hard time logging in. People being lagged when running missions and being thrown out from the server. [...] Do some serious tests (log on SiSi and see if you have the same problems there) and THEN come back and whine.
serious tests? what the **** are you talking about mungo?
i've tried logging in from 4 different machines with 2 different connections using 2 different accounts at a whole bunch of different times throughout the day, and compared that to the number of users on line and the approx. number of logins per minute. Short of breaking into the ccp server farm there's not a great deal else to test.
Besides, how on earth would logging into sisi tell you anything about lag on tranq when it's blatantly a capacity issue?
|

Demarcus Gainah
|
Posted - 2007.05.17 12:41:00 -
[144]
Originally by: ry ry Edited by: ry ry on 17/05/2007 12:29:56
Originally by: Thargat I just don't get it. I read about people having a hard time logging in. People being lagged when running missions and being thrown out from the server. [...] Do some serious tests (log on SiSi and see if you have the same problems there) and THEN come back and whine.
serious tests? what the **** are you talking about mungo?
i've tried logging in from 4 different machines with 2 different connections using 2 different accounts at a whole bunch of different times throughout the day, and compared that to the number of users on line and the approx. number of logins per minute. Short of breaking into the ccp server farm there's not a great deal else to test.
Besides, how on earth would logging into sisi tell you anything about lag on tranq when it's blatantly a capacity issue?
I can't believe it has to be said again, but really the issue isn't really capacity. It's certainly not simply capacity anyway. Concentrated capacity can affect in game lag in particular areas, because Eve is basically fractured into regional servers, if you will. However, general capacity should not affect overall lag everywhere, nor should it affect your ability to log on under normal circumstances. Theoretically Eve should be able to hold many, many more players without simple capacity being a major issue. Personally, I rarely, if ever have connection problems and I know many, many people can vouch to this. If the issue were really simple capacity then EVERYONE would be affected.
|

ry ry
StateCorp
|
Posted - 2007.05.17 12:47:00 -
[145]
yeah. you're completely right.
Originally by: "Wrangler" To address this we are running at decreased hardware capacity
oh no wait! everything worked fine for me until recently - comedy ingame lag aside, but then thet's not what i was whining about was it? you were infact completely wrong.
|

Solbright alt
|
Posted - 2007.05.17 12:52:00 -
[146]
Originally by: Ender Ti'kai One extremely simple thing that could to vastly increase enjoyability be done is the option to drastically decrease texture quality and a much lower resolution setting! In fact, an option to remove the images of ships from display altogether would help many people out a LOT, especially with larger and larger numbers being seen in major, decisive combat. Please, please, please implement this CCP!
Got my vote for this one.
|

Thargat
Caldari S-44 Coalition of Carebear Killers
|
Posted - 2007.05.17 12:52:00 -
[147]
Serious testing means that you should eliminate as many sources of error as possible before deciding it's all server related. It won't be easy and it won't be done in no time (more like a few days of tinkering and testing)but it'll be worth it I promise. Maybe not for you directly (it might all be hopeless for some players out there) but by eliminating every source of error from your own and your ISPs end (to a reasonable level) and the bug reporting the rest (with specs) your doing the community a favor. And since alot of people seem to be playing with minimal problems then "hey" maybe it isn't ALL CCPs fault that your experiencing problems.
If you don't like what I say, just ignore it, or admit I got to you. |

Demarcus Gainah
|
Posted - 2007.05.17 12:52:00 -
[148]
Originally by: ry ry yeah. you're completely right.
Originally by: "Wrangler" To address this we are running at decreased hardware capacity
oh no wait! everything worked fine for me until recently - comedy ingame lag aside, but then thet's not what i was whining about was it? you were infact completely wrong.
Were you talking to me? Pardon me then, as I thought you were referring to problems with connecting in general, not just the past day or so. I definitely wouldn't know anything about the reduced hardware capacity effects, and I'm thinking I misread your comment.
|

Solbright alt
|
Posted - 2007.05.17 12:54:00 -
[149]
Originally by: ry ry Besides, how on earth would logging into sisi tell you anything about lag on tranq when it's blatantly a capacity issue?
No way it's a capacity issue.
|

Solbright alt
|
Posted - 2007.05.17 13:00:00 -
[150]
Originally by: Matthew Even if you're sitting in station afk, anyone entering or leaving local, the station guest list, any other chat updates, wallet/evemail flashes etc, will all generate an event to be sent to the client.
Would cool to get confirmation on how the low level network interacts here. Ie: Does it really use ACKs?
You're prolly right though, it sounds like a worth while error checking function.
|
| |
|
| Pages: 1 2 3 4 [5] 6 7 8 9 :: one page |
| First page | Previous page | Next page | Last page |