JeanPaul Mesaj tarihi: Mart 7, 2003 Paylaş Mesaj tarihi: Mart 7, 2003 Offf offff offffff. Su anda gerci gitti Atlantic, Login Server filan ama gitmesine gerek yoktu, Atlantic ping 610 ms (ustelik de NYC ustunden cikmamiza ragmen) arti AoS lag de ustune bindi. Kosmayi birakin dogru duzgun yurunemiyor bile. Umarim yarin gece boyle olmaz. uo board'larda tartisilan konu su anki hali ile dial-up kullanicilari artik iyi baglanti yapamiyor , (dikkatinizi cekerim, amerika'daki dial-up kullanicilari), kapasite yetmiyor, adamlar her bir bilgiyi (item, vendor, vs) client'a gonderiyorlar, dolayisi ile oyundasiniz, etrafinizda kimse yok, 1 ya da 2 kisi girsin ekraniniza, aninda lag'e ugruyorsunuz. birakin vendor mall'larin yanindan gecerken ki durumu. OSI, net trafiklerini (data stream) onumuzdeki hafta test edecekler ve duruma bakacaklarmis ama ben pek umitli degilim. Eger duzeltmez iseler, amerika'daki 56K kullanicilar dahil olmak uzere ve bizler (kablonet bile olsa) artik oyunda bisi yapamiyacaz. pvm dahi zorlasacak. Bakalim gorelim.[hline]Ride to live, Live to ride... Link to comment Sosyal ağlarda paylaş Daha fazla paylaşım seçeneği…
JeanPaul Mesaj tarihi: Mart 7, 2003 Konuyu açan Paylaş Mesaj tarihi: Mart 7, 2003 Bir blackhole olayi. Yeni fix ile eger bir sekilde kalabalik bir yerden disconnect oldu iseniz ya da logoff filan yapti iseniz, tekrar logon oldugunuzda durumu anliyor ve sizi ve beraberinizdeki hayvani safe-zone gonderip ordan oyuna giriyorsunuz. boylece eskisi gibi kilitlenmiyorsunuz oyuna girerken. Ornegin ben brit-bank'de iken disconnect olmustum, beni tram-glow counselor guild house'a alarak baslatti. workaround diyorlar. gecici cozum.[hline]Ride to live, Live to ride... Link to comment Sosyal ağlarda paylaş Daha fazla paylaşım seçeneği…
Simofle Mesaj tarihi: Mart 7, 2003 Paylaş Mesaj tarihi: Mart 7, 2003 Benım makınayı aosun gelmesıyle bırlıkte jkasmaya basladı,mouse hareketlerınde yavaslama soz konusu oldu anlamıs degılım osi clientından kaynaklanıyor bendekı sorun[hline]Simofle [b]O 55 bin sosisliden biri -.-. --...--. .-.-.-.... ..-..-...-.-....-....-.-...-.... ---.-....--.-..--.- Link to comment Sosyal ağlarda paylaş Daha fazla paylaşım seçeneği…
JeanPaul Mesaj tarihi: Mart 8, 2003 Konuyu açan Paylaş Mesaj tarihi: Mart 8, 2003 1:20 ile bu saat arasi normal hizina kavustu.[hline]Ride to live, Live to ride... Link to comment Sosyal ağlarda paylaş Daha fazla paylaşım seçeneği…
deviL Mesaj tarihi: Mart 8, 2003 Paylaş Mesaj tarihi: Mart 8, 2003 ehe evet hoca amerika'dan dial up ile oynamak iskence... eskiden runebook acılmazdı (2 saat gumpı download ederdi) ona laf ederdim ama durum felaket simdi. Goodman'in bile (bisürrü vendor var ama yine de L shape) 2 ekran yanından gecince cokuyordu daoc gibi 3d mmorpgler de bile bu kadar lag yasamadım, oynayasım gelmiyor Link to comment Sosyal ağlarda paylaş Daha fazla paylaşım seçeneği…
JeanPaul Mesaj tarihi: Mart 18, 2003 Konuyu açan Paylaş Mesaj tarihi: Mart 18, 2003 En sonunda aciklama geldi: The data stream is being audited, just as I said it would be two weeks ago. This only started late last week because there were other fixes we deemed more important to get into production, among them the actual fix for the blackhole problem. Since the person working on the blackhole fix (Fertbert) is the same person who is working on reducing bandwidth utilization, these were not tasks that could be done in parallel. We know dialup is not fun right now. We are working to improve this situation. There are some fixes in a client patch that will be out this week, and a publish that MAY be out this week, that should make things BETTER, though not GREAT. An aggressive bandwidth optimization pass is in the works and this should make things considerably better for dialup players. As for UOASpeed, Tug has contacted us and we are discussing it. I can't predict right now how it's going to go, but I don't consider it anything to be alarmed about. MrTact ve AoS Bandwith Usage (modem users please read this) First of all, we're aware that some of you are experiencing lag since the publish of Age of Shadows. For that I'd like to apologize and explain what we're doing to improve things. While reading this explanation understand that bandwidth usage and lag are related... more bandwidth usage tends to create more lag for low bandwidth users. Here are the Age of Shadows features that do and will continue to use more bandwidth than Event 5 did: - Customized houses are about 500 to 2000 bytes each and are cached client side in a persistent cache - Item properties are about 20 to 100 bytes each and are cached client side in a temporary cache - Floating damage numbers are small number of bytes (< 20) Some details about what we're doing or have done to reduce the impact of these: - Customized houses are about as optimized as they're likely to get right now. Each house is broken into slices and encoded using one of several algorithms, then each slice is compressed. The only further optimizations that can be made with regard to housing is for us to gradually ship each slice to the client instead of sending all the slices at once, and to turn off the standard packet compression/encoding for housing packets. Since I have yet to read one single complaint about customized houses causing undue lag, I have no immediate plans to implement those further optimizations. Further more the majority of the house-related lag appears to be due to the lockdowns and vendors… - Items properties are the current target of our optimization efforts... these appear to be the major cause of the increased lag, and are what have been causing people to be blackholed. The size of the average item properties packet caught us by surprise, we had not originally intended to send so many properties, so our initial impact analysis was way off. We have a fix in QA right now to address this problem. We're now throttling the rate at which these packets are sent to the client so that when people enter vendor malls or recall to a popular place their available bandwidth is not overwhelmed. - Floating damage numbers do not appear to have a significant impact upon lag, are are already as optimized as they're likely to be. Those same features that can cause lag, can also cause the client to be completely overwhelmed with data resulting in a character becoming blackholed. We have a patch in QA to address this problem. I have refactored the server side buffering code to behave better in these situations. Once this patch goes live, naturally occurring blackholes at champion spawns, banks, and vendor malls should be rare. Also in this patch is a fix for the packet corruption problem that plagued many users (especially modem users.) This corruption is likely the cause of the majority of the recent random disconnects, seeing the character select screen in the middle of gameplay, and other miscellaneous weirdness. The reason why this bug appears to be more prevelant now is that it is only triggered when one of the server side buffers in near capacity… as we’ve been seeing lately with the increased bandwith usage of Age of Shadows. In development is a fix that should address the famous multiple door problem (as it applies to regular buildings in UO) as well as a related cause of blackholes around places with too many candelabra, communication crystals, the Delucia bank, etc. Right now I am focusing on more item properties optimizations, and weeding through out data stream to remove duplicate information. I anticipate having another batch of fixes in Publish 18. Thank you for your patience in this matter, Fertbert[hline]Ride to live, Live to ride... Link to comment Sosyal ağlarda paylaş Daha fazla paylaşım seçeneği…
JeanPaul Mesaj tarihi: Mart 18, 2003 Konuyu açan Paylaş Mesaj tarihi: Mart 18, 2003 4.0.0e Update. Brit Bank'de az lag ile kosabildim. Sanirim olacak bu is: http://update.uo.com/fyi.html :)[hline]Ride to live, Live to ride... Link to comment Sosyal ağlarda paylaş Daha fazla paylaşım seçeneği…
Öne çıkan mesajlar