From MAILER-DAEMON@host27.websitesource.com Sat May 29 23:48:13 2004 Received: from localhost (localhost) by host27.websitesource.com (8.12.10/8.12.10) id i4U3mDx7025836; Sat, 29 May 2004 23:48:13 -0400 Date: Sat, 29 May 2004 23:48:13 -0400 From: Mail Delivery Subsystem Message-Id: <200405300348.i4U3mDx7025836@host27.websitesource.com> To: MIME-Version: 1.0 Content-Type: multipart/report; report-type=delivery-status; boundary="i4U3mDx7025836.1085888893/host27.websitesource.com" Subject: Returned mail: see transcript for details Auto-Submitted: auto-generated (failure) Status: R This is a MIME-encapsulated message --i4U3mDx7025836.1085888893/host27.websitesource.com The original message was received at Sat, 29 May 2004 23:48:13 -0400 from localhost.localdomain [127.0.0.1] ----- The following addresses had permanent fatal errors ----- (reason: 550 Host unknown) ----- Transcript of session follows ----- 550 5.1.2 ... Host unknown (Name server: istcalcc.com: host not found) --i4U3mDx7025836.1085888893/host27.websitesource.com Content-Type: message/delivery-status Reporting-MTA: dns; host27.websitesource.com Received-From-MTA: DNS; localhost.localdomain Arrival-Date: Sat, 29 May 2004 23:48:13 -0400 Final-Recipient: RFC822; davidford@istcalcc.com Action: failed Status: 5.1.2 Remote-MTA: DNS; istcalcc.com Diagnostic-Code: SMTP; 550 Host unknown Last-Attempt-Date: Sat, 29 May 2004 23:48:13 -0400 --i4U3mDx7025836.1085888893/host27.websitesource.com Content-Type: message/rfc822 Return-Path: Received: from 1stcallcc.com (localhost.localdomain [127.0.0.1]) by host27.websitesource.com (8.12.10/8.12.10) with ESMTP id i4U3mCx7025828 for ; Sat, 29 May 2004 23:48:13 -0400 From: "david" To: davidford@istcalcc.com Subject: post Date: Sat, 29 May 2004 23:48:12 -0400 Message-Id: <20040530034812.M24616@1stcallcc.com> X-Mailer: Open WebMail 1.81 20021127 X-OriginatingIP: 68.7.179.24 (david) MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Inverse Logic > General Discussion > Nice PATCH SoE :) << Prev Topic | Next Topic >> Author Comment Alcatrazeq You will not evade me %T! Posts: 897 (5/28/04 11:14 am) Reply Nice PATCH SoE :) ------------------------------------------------------------------------------ -- OK we notice some of this last night but here is what the DEVs are saying about 5-26 patch Posted from Grumbok-Dev Just wanted to provide what info I can about the collision system status. First, the bad news. It's still not completely fixed. We know this, and are continuing to work as hard as possible on correcting this. It's been a long road but I think we are nearly there. I wanted to list the major collision issues we're aware of and workarounds to minimize the effect these problems will have to your gameplay. Here they are: * Summoning a horse when near a wall or other object can result in you being unable to move. The good news is that you can simply deactivate your mount, move to a more open space, and summon again. * Getting summoned by an NPC that is smaller than you (such as a sarnak summoning an ogre) when the NPC is next to a wall will generally result in the player getting stuck in the wall. Please avoid corner tanking small NPCs that summon until we can fix this. I know this is a major problem and it is one of the top priorities to fix. Note that if the NPC has a collision volume larger than the player being summoned, this shouldn't be a problem. * Changing form, such as with illusions, can cause your collision volume to penetrate geometry which will leave you stuck. This isn't so much a problem when casting the illusion initially, but when your illusion fades and you suddenly become larger and you are next to a wall you will generally get stuck. Try to avoid using illusions that make you smaller. * Shrinking is pretty much the same thing as illusions. If you camp while shrunk, you will log in full size and if you were next to a wall or object you may become stuck. Try to avoid camping while shrunk. * If you are on a boat for a very long time, you will notice that you slowly move towards the back of the boat and in some cases can begin to get stuck in the boat geometry. If this should happen, just ride it out you will eventually push through it all. This can take a long time to even see this problem, but it's something folks should be aware of. So here's the good news. Like pretty much any patch day when it's far too late to make code changes, it seems that the NPCs & players sunk into the ground may finally be resolved. At least, we think we may have a fix for that problem and we are going to put it on Test soon to see if it helps. There was definitely a problem, and the upcoming change should correct that problem. Since the sunken npc/players bug is so hard to reproduce we're going to have to let it cook on test for a bit to see if it really helped or not. I'm optimistic at this point. This bug also had the effect of causing the find command to not always work correctly. We'll find out more after it's been on Test for a while. Thanks for your patience on this. It's been a very unfortunate thing that everyone has had to suffer with these problems as long as you have and I hope to be able to soon say that it is all fixed. Please continue to post any specific problems you may have seen related to collision in this thread. Please *DON'T* sidetrack this thread into a discussion about expansions, management decisions, or any other general EQ stuff. I *need* this thread to have data that can help solve collision related code problems because your feedback has been crucial in not only solving - but in some cases even learning about (ie PoK crash) - as many problems as we have. Thanks, Posted by Grumbok-Dev at 5-27 11AM This morning I was able to reproduce the problem with NPCs getting pushed in the walls and find what I believe is the reason why it was happening. We'll be pushing the fix to test ASAP, and this will also hopefully fix the sunken NPC problem but since there seems to be no way to reliably reproduce the sunken NPC problem we're going to have to wait for reports to either stop or continue to come in about them before we get a better feel for if that one is solved. I made a change to the code to allow our tester GMs to summon an army of NPCs to simulate a raid force pushing an NPC around and the problem with walled NPCs became apparent the first time it was used. Long story short, the server was using a collision volume that was different (smaller) than the client's and the two became out of sync resulting in a variety of problems. We'll be making an emergency patch as soon as we are sure these changes will help the situation. Posted by Grumbok-Dev at 5-27 4pm We're about to push the fixes for walled mobs & a potential fix for sunken NPCs/players to test. I was able to check, and in the current code, a gnome on a drogmor or horse has no problem with the steps in potranquility by the portal stones. There is something weird about the clip plane for gnomes which I'll look into after the critical issues are taken care of. Our QA verified this problem so it would seem the other changes have corrected this before I had a chance to identify what was the nature of that problem. Considering the code I just modified today, it's plausible that the bug was directly related to it. After the patch we'll check to see if short races still have issues with horses (they work fine on my dev boxes at this point). We do have a debug client that has a lot of features that helps diagnose. To reproduce these bugs *and* get useful information besides "it's still bugged", we have to have the zone server running in debug mode, and the client in debug mode, both attached to a debugger and *then* when the bug occurs we activate the debuggers and look at the states of everything. There's no doubt you folks have been seeing problems. It's just been difficult to get some of these problems to occur when we are set up for debugging. The army of pets worked pretty well for the walled NPC thing though, I'm pretty sure the walled NPC situation should be corrected at this point. Posted by Grumbok-Dev 05-28-2004 07:46 AM Well as much as I'd like to just push the code live and be sure it solved all the problems, it's simply not going to happen until it's verified to actually help. For one thing, the code on test right now turned out to not solve the problem with sunken NPCs. We're doing everything we can to solve these problems, I am sorry you folks are having to deal with these bugs. Going to close this thread for now, I'll start another when there's new information. So Basically right now we are screwed...A lot of the thread has people posting on where NOT to fight...which is basically everywhere -- ALC shogung Inverse Logic Member Posts: 1649 (5/28/04 11:44 am) Reply Re: Nice PATCH SoE :) ------------------------------------------------------------------------------ -- ever get that sinking feeling.....? Shogung (RIP) Battlecleric Hipo (still alive) Phyllobates (who let the frogs out? ribbit, ribbit) (RIP) Csonka1 Posts: 14 (5/28/04 12:14 pm) Reply ezSupporter Re: Nice PATCH SoE :) ------------------------------------------------------------------------------ -- I was having trouble with my UI last night. I and I do not use a custom on anymore. My Crap Mana Boy "Classes are not that out of balance" -AbsorEQ Aneadorn Registered User Posts: 44 (5/28/04 12:34 pm) Reply Re: Nice PATCH SoE :) ------------------------------------------------------------------------------ -- While I'm all for them fixing the walling sunken npc DX9 problem there is one thing I want to point out. Only certain mobs need to be cornered and even then there are other ways. The one thing about fighting a mob out in the open is that it forces everyone to pay attention. It forces your raid to really work as a team. There was a guild on my old server that did not corner mobs they where very skilled at controlling push and circling a mob. They also had great success on encounters with lower numbers than any of the other guilds on TP. Stormmaster T Registered User Posts: 51 (5/28/04 12:53 pm) Reply ezSupporter Re: Nice PATCH SoE :) ------------------------------------------------------------------------------ -- My eyes must be buggy. Their debugger is bugged? Its interesting the part about the not wanting to put untested or bugged code to the live servers. Hasn't seem to stop them before. As far as cornering mobs, do you "have" to corner mobs. Why not surround a mob and be extremely aware of your position and your push. We used to have main tank obviously in front of the mob, ONLY rogues behind the mob, hybrids and other tanks on the sides, pets would be spaced out to surround a mob and for the most part mobs stayed fairly well in place with minor adjustments. Stormmaster's Stuff Firion Wayfinder Inverse Logic Member Posts: 1468 (5/28/04 3:00 pm) Reply Re: Nice PATCH SoE :) ------------------------------------------------------------------------------ -- IL as a guild has struggled to master push for a long time. I would love to see us overcome it. Firion Wayfinder Forest Stalker Master Fletcher, Blacksmith, and Camp Cook Hunt 'em, Kill 'em, Cook 'em and Eat 'em --i4U3mDx7025836.1085888893/host27.websitesource.com-- From MAILER-DAEMON@host27.websitesource.com Sat May 29 23:51:14 2004 Received: from localhost (localhost) by host27.websitesource.com (8.12.10/8.12.10) id i4U3pEx7028653; Sat, 29 May 2004 23:51:14 -0400 Date: Sat, 29 May 2004 23:51:14 -0400 From: Mail Delivery Subsystem Message-Id: <200405300351.i4U3pEx7028653@host27.websitesource.com> To: MIME-Version: 1.0 Content-Type: multipart/report; report-type=delivery-status; boundary="i4U3pEx7028653.1085889074/host27.websitesource.com" Subject: Returned mail: see transcript for details Auto-Submitted: auto-generated (failure) Status: R This is a MIME-encapsulated message --i4U3pEx7028653.1085889074/host27.websitesource.com The original message was received at Sat, 29 May 2004 23:51:14 -0400 from localhost.localdomain [127.0.0.1] ----- The following addresses had permanent fatal errors ----- (reason: 550 Host unknown) ----- Transcript of session follows ----- 550 5.1.2 ... Host unknown (Name server: firstcallcc.com: host not found) --i4U3pEx7028653.1085889074/host27.websitesource.com Content-Type: message/delivery-status Reporting-MTA: dns; host27.websitesource.com Received-From-MTA: DNS; localhost.localdomain Arrival-Date: Sat, 29 May 2004 23:51:14 -0400 Final-Recipient: RFC822; davidford@firstcallcc.com Action: failed Status: 5.1.2 Remote-MTA: DNS; firstcallcc.com Diagnostic-Code: SMTP; 550 Host unknown Last-Attempt-Date: Sat, 29 May 2004 23:51:14 -0400 --i4U3pEx7028653.1085889074/host27.websitesource.com Content-Type: message/rfc822 Return-Path: Received: from 1stcallcc.com (localhost.localdomain [127.0.0.1]) by host27.websitesource.com (8.12.10/8.12.10) with ESMTP id i4U3pEx7028644 for ; Sat, 29 May 2004 23:51:14 -0400 From: "david" To: davidford@firstcallcc.com Subject: post Date: Sat, 29 May 2004 23:51:14 -0400 Message-Id: <20040530035114.M85234@1stcallcc.com> X-Mailer: Open WebMail 1.81 20021127 X-OriginatingIP: 68.7.179.24 (david) MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 > Nice PATCH SoE :) Author Comment Alcatrazeq You will not evade me %T! Posts: 897 (5/28/04 11:14 am) Reply Nice PATCH SoE :) ------------------------------------------------------------------------------ -- OK we notice some of this last night but here is what the DEVs are saying about 5-26 patch Posted from Grumbok-Dev Just wanted to provide what info I can about the collision system status. First, the bad news. It's still not completely fixed. We know this, and are continuing to work as hard as possible on correcting this. It's been a long road but I think we are nearly there. I wanted to list the major collision issues we're aware of and workarounds to minimize the effect these problems will have to your gameplay. Here they are: * Summoning a horse when near a wall or other object can result in you being unable to move. The good news is that you can simply deactivate your mount, move to a more open space, and summon again. * Getting summoned by an NPC that is smaller than you (such as a sarnak summoning an ogre) when the NPC is next to a wall will generally result in the player getting stuck in the wall. Please avoid corner tanking small NPCs that summon until we can fix this. I know this is a major problem and it is one of the top priorities to fix. Note that if the NPC has a collision volume larger than the player being summoned, this shouldn't be a problem. * Changing form, such as with illusions, can cause your collision volume to penetrate geometry which will leave you stuck. This isn't so much a problem when casting the illusion initially, but when your illusion fades and you suddenly become larger and you are next to a wall you will generally get stuck. Try to avoid using illusions that make you smaller. * Shrinking is pretty much the same thing as illusions. If you camp while shrunk, you will log in full size and if you were next to a wall or object you may become stuck. Try to avoid camping while shrunk. * If you are on a boat for a very long time, you will notice that you slowly move towards the back of the boat and in some cases can begin to get stuck in the boat geometry. If this should happen, just ride it out you will eventually push through it all. This can take a long time to even see this problem, but it's something folks should be aware of. So here's the good news. Like pretty much any patch day when it's far too late to make code changes, it seems that the NPCs & players sunk into the ground may finally be resolved. At least, we think we may have a fix for that problem and we are going to put it on Test soon to see if it helps. There was definitely a problem, and the upcoming change should correct that problem. Since the sunken npc/players bug is so hard to reproduce we're going to have to let it cook on test for a bit to see if it really helped or not. I'm optimistic at this point. This bug also had the effect of causing the find command to not always work correctly. We'll find out more after it's been on Test for a while. Thanks for your patience on this. It's been a very unfortunate thing that everyone has had to suffer with these problems as long as you have and I hope to be able to soon say that it is all fixed. Please continue to post any specific problems you may have seen related to collision in this thread. Please *DON'T* sidetrack this thread into a discussion about expansions, management decisions, or any other general EQ stuff. I *need* this thread to have data that can help solve collision related code problems because your feedback has been crucial in not only solving - but in some cases even learning about (ie PoK crash) - as many problems as we have. Thanks, Posted by Grumbok-Dev at 5-27 11AM This morning I was able to reproduce the problem with NPCs getting pushed in the walls and find what I believe is the reason why it was happening. We'll be pushing the fix to test ASAP, and this will also hopefully fix the sunken NPC problem but since there seems to be no way to reliably reproduce the sunken NPC problem we're going to have to wait for reports to either stop or continue to come in about them before we get a better feel for if that one is solved. I made a change to the code to allow our tester GMs to summon an army of NPCs to simulate a raid force pushing an NPC around and the problem with walled NPCs became apparent the first time it was used. Long story short, the server was using a collision volume that was different (smaller) than the client's and the two became out of sync resulting in a variety of problems. We'll be making an emergency patch as soon as we are sure these changes will help the situation. Posted by Grumbok-Dev at 5-27 4pm We're about to push the fixes for walled mobs & a potential fix for sunken NPCs/players to test. I was able to check, and in the current code, a gnome on a drogmor or horse has no problem with the steps in potranquility by the portal stones. There is something weird about the clip plane for gnomes which I'll look into after the critical issues are taken care of. Our QA verified this problem so it would seem the other changes have corrected this before I had a chance to identify what was the nature of that problem. Considering the code I just modified today, it's plausible that the bug was directly related to it. After the patch we'll check to see if short races still have issues with horses (they work fine on my dev boxes at this point). We do have a debug client that has a lot of features that helps diagnose. To reproduce these bugs *and* get useful information besides "it's still bugged", we have to have the zone server running in debug mode, and the client in debug mode, both attached to a debugger and *then* when the bug occurs we activate the debuggers and look at the states of everything. There's no doubt you folks have been seeing problems. It's just been difficult to get some of these problems to occur when we are set up for debugging. The army of pets worked pretty well for the walled NPC thing though, I'm pretty sure the walled NPC situation should be corrected at this point. Posted by Grumbok-Dev 05-28-2004 07:46 AM Well as much as I'd like to just push the code live and be sure it solved all the problems, it's simply not going to happen until it's verified to actually help. For one thing, the code on test right now turned out to not solve the problem with sunken NPCs. We're doing everything we can to solve these problems, I am sorry you folks are having to deal with these bugs. Going to close this thread for now, I'll start another when there's new information. So Basically right now we are screwed...A lot of the thread has people posting on where NOT to fight...which is basically everywhere -- ALC shogung Inverse Logic Member Posts: 1649 (5/28/04 11:44 am) Reply Re: Nice PATCH SoE :) ------------------------------------------------------------------------------ -- ever get that sinking feeling.....? Shogung (RIP) Battlecleric Hipo (still alive) Phyllobates (who let the frogs out? ribbit, ribbit) (RIP) Csonka1 Posts: 14 (5/28/04 12:14 pm) Reply ezSupporter Re: Nice PATCH SoE :) ------------------------------------------------------------------------------ -- I was having trouble with my UI last night. I and I do not use a custom on anymore. My Crap Mana Boy "Classes are not that out of balance" -AbsorEQ Aneadorn Registered User Posts: 44 (5/28/04 12:34 pm) Reply Re: Nice PATCH SoE :) ------------------------------------------------------------------------------ -- While I'm all for them fixing the walling sunken npc DX9 problem there is one thing I want to point out. Only certain mobs need to be cornered and even then there are other ways. The one thing about fighting a mob out in the open is that it forces everyone to pay attention. It forces your raid to really work as a team. There was a guild on my old server that did not corner mobs they where very skilled at controlling push and circling a mob. They also had great success on encounters with lower numbers than any of the other guilds on TP. Stormmaster T Registered User Posts: 51 (5/28/04 12:53 pm) Reply ezSupporter Re: Nice PATCH SoE :) ------------------------------------------------------------------------------ -- My eyes must be buggy. Their debugger is bugged? Its interesting the part about the not wanting to put untested or bugged code to the live servers. Hasn't seem to stop them before. As far as cornering mobs, do you "have" to corner mobs. Why not surround a mob and be extremely aware of your position and your push. We used to have main tank obviously in front of the mob, ONLY rogues behind the mob, hybrids and other tanks on the sides, pets would be spaced out to surround a mob and for the most part mobs stayed fairly well in place with minor adjustments. Stormmaster's Stuff Firion Wayfinder Inverse Logic Member Posts: 1468 (5/28/04 3:00 pm) Reply Re: Nice PATCH SoE :) ------------------------------------------------------------------------------ -- IL as a guild has struggled to master push for a long time. I would love to see us overcome it. Firion Wayfinder Forest Stalker Master Fletcher, Blacksmith, and Camp Cook Hunt 'em, Kill 'em, Cook 'em and Eat 'em --i4U3pEx7028653.1085889074/host27.websitesource.com--