PlaneShift
Support => Technical Help: IN GAME bugs (after loading world) => Topic started by: Persiffon on August 27, 2005, 03:18:22 pm
-
Hello All,
Recently I have been refocusing my attention on the true reason we are all here, to work the system and find some bugs. Maybe help the devs fix a few things.
Lately in my endevor to do just that I was told by a player or should I say \"warned\" that my attempts to recreate a bug , that was in wide use, that I was using and teaching someone else that bug or \"exploit\"? Here is my question how can you exploit something when? A) in order to exploit something you must gain something perminent to form an advantage over the rest of the player base. B) there will be wipes of the player base in order for the devs to have a level playing field over the rest. Wipes will happen because the devs will not be able to track every little thing the each and everyone does. That is the fact and thats why the stickies are there to infor the rest that it is inevitable.
This is a Tech Demo. It states it very clearly in several places. The devs use this fact to test the software for future releases and \"we\" are all testers.
Now to get to the main subject of this post. On several occasions I have been harrassed by other play testers because I was trying to reproduce the bugs to see if it was reproducable and what it takes to reproduce the bugs. I was and have been a long time tester of Planeshift. I probably always will be. I am proud of what this open source game represents in both time and technology that the devs have made. I would like to think that the community has degraded to a game player base instead of being what the devs need. Which is input and help testing the code. Instead we have a bunch of players wandering round reporting people for little infractions like spawn camping (which any other pay service don\'t bother with). Camping is a fact no matter where you go in this genre. To boot someone right now in the current state of the software is alittle like putting a bandaid on a mortal wound.
so what I usually do is move onto the next area and let someone else take over the spawn. I don\'t report them because i feel like causing problems for the GM\'s.
With that being said I was trying to get as much info about why the mobs loose thier focus on a player and I think I did that. I even tried to get others to test it out and this is what I found.
The process: About 20-30 game feet start a normal attack. The mob will come towards you. When the mob gets about 5 feet from your character start spamming the attack button while nudging backwards. When the mob stops, which it will eventually, gently nudge forward until your character starts attacking and walla you can kill the meanest mob on the world without it attacking.
Now this is my question: Why are the basics not being adressed? This bug is a major issue and should be worked on asap to facilitate better testing in the future. All to many time we loose focus on the current issues only to facilitate newer features. But if those features are marred with previous Bugs how can we acurately test the newer features. This flaw is making the tech demo look bad because alot use it to even the playing field in a tech demo full of imballances.
Please devs if you do anything adress the basics like combat, abilities and anything that forms the building blocks of your vision before adding things like magic and siege weapons. Just an example.
It would help us play testers in helping you with ferreting out the bugs if we have a solid core to run from.
Thanks for your time,
Persiffon
-
If you report your findings, there\'s nothing wrong with trying to recreate a bug. Exploiting is something different, that\'s when you continuously \"recreate\" a bug for your own gain. I think that\'s lame. And spawn killing, well, technically it\'s allowed, and there\'s nothing you can really do about it, except for continuously harassing the GM\'s, so when I see someone spawncamping, I simply move to another monster, very simple.
And I think this community is what the devs need, people are enthousiastic and remember that it would be very boring to just test if some new bugfixes work. People like changes and new features. If nothing would change in PS except for bugfixes and changing the very basics a bit, a lot of people would stop playing PS. At the moment there is a perfect balance between features and bug fixes in my opinion. First they bring out a patch with new features, and then they fix all the bugs, then they start working for new features and after that they start working on fixing bugs again.
-
So far as a GM i personly have not had a single complaint about people spawn camping and really if a person did complain i would ignore them.
I have seen PLENTY of spawn places for ALL classes. There\'s rats in Sewers, Akkio, Arena and more.
If you have found a bug and want to reproduce keeping in mind to try and resolve it, thats perfectly fine.
If you want to report your findings goto here (http://planeshift.oodlz.com/bugtracker/index.php) instead of posting them in public ;)
-
Spawn camping dosen\'t bother me but like I have seen and read on the forums it does bother some. However IMHO right now should be dedicated to helping the devs with getting things straight. Obviously There are other GM\'s who have taken issue with spawn camping and imho goto the extreme. I won\'t name names but I believe at this time this software is in no condition to support the claims of \"Exploiting\" and \"spawn camping\". Seeing that wipes are gonna control what the devs see as such. Planeshift is in it\'s infancy and I see great things for the future. On the other hand there are a great many people who are doing wonders to help (GM) (testers) and the likes.
So cudos to you guys and gals for helping make this tech demo a great and wonderous setting for our fatasy world!
Thanks for letting me say my peace,
Persiffon
-
Hello,
First of all I must say that I am astonished of your devotion to PS and I am really happy - personally, I don\'t speak for others - about players like you, Persiffon.
As ThomPhoenix says the problem is that not everybody enjoys testing all the time, therefore it is normal that the things have somehow degenerated mainly in a almost standard game and not pure testing. This also depends from the fact that, even if the game is in a test phase, Planeshift is not just merely searching for testers, but players as well.
The difference? I think the fact that the community is growing up, and that some \"pillars\" have been already created. These pillars are players that are important for the game.
Of course, if everybody would test and help find the bugs and fix them, this would be a real test phase, as it is intended in more \"company oriented\" environments, but nobody of the team has ever forced a player to not \"play\" the game, inventing RPG actions or enjoying \"powering up\" with skills and money. Also the dynamics of the game are being tested, if you think, in this way.
This doesn\'t mean I want to stop you from testing in the way you do. I think it is a really good thing that there are players devoted as you are. But of course, considering the amount of people that has created an account, it would be hard - even if wanted - to pretend that everybody would be here to test and find the problems in the code.
From my experience, I can say that when I needed somebody to test some bug I am working on, I have always found people willing to help. And this is a good sign :-)
-
I have found in this as in all other mmorpg\'s which I have played that in order to effectively explore the game enviornment and the capabillities of it\'s engine, a certain ammount of powergaming isw neccessary.
My pleasure in any game, but especially in one under development is in exploration. But before I can wander safely across the whole worldmap or play around with the capabillities of the game, it is absolutely necessary that I build up my charachters. In this endeavor, \"bad rp-etiquette\" activities such as spawncamping are helpful.
If I were truly here to be a playtester pure and simple, I would have the power to simply edit my char. to have whatever attributes/skills I needed for a given playtest session rather than being asked to earn them. Playtest in a setting such as this where I am asked to test the game on its own terms is necessarily a hybrid thing, I wouldn\'t agree with anyone who accuses powergamers of breaking faith with their role as playtesters on a prima facia basis.
-
However early in development we may be, this is still a game, and it still has rules. Exploitation being one of the things we do not tolerate.
As said previously in this thread, exploiting is continuing to reproduce a bug when you already know how it is caused and how it happens, yet you do it anyway because it gives you an advantage. When a bug is continually replicated and not reported (or, reported but then exploited over and over untill it\'s fixed), that is considered exploiting.
Testing for bugs and reporting what happens is absolutely encouraged. And there is a very easy way to see whether someone is testing or exploiting something: watch them. They do it once or twice, good for them. They do it ten times, call the GMs. :)
-
Wow, with this thread I realised that testing is more important (insert childish flame here)
But your completely right, to test bugs we need to train our characters, maybe with a lower level character we can discover that we can whack at something for 10 minutes until its dead by merely using the combat bug and not being hit...
Also while I\'m here, the outside walls of Hydlaa and anywhere else need serious help, it would be best to maybe put a lot of trees there and make it into an unexplored forest or something, give it more of an RP feal until the next version of the game (SB steel blue) is created when you can add more areas and dungeons, then off for more bug tracking.
But what I want to point out, is that fixing the bugs after creating the new features, is a waste of time because eventually with all new stuff being added, that bug will work its way back up, like the black glitch (but this time its worse due to the different colored ones) Don\'t you think that as you add to the Oja road and other places that the hole might re-open in the road?
THIS is where the fly glitch would be handy, that shouldn\'t have been removed until the last points of the game\'s testing time and the main game is almost implimented. This would allow people to explore new, bugged areas and discover even more exploits and so on, though most people would abuse it and jump on the statue of laanx (like I used to do :/) it won\'t matter because after new bugs are discovered, you can remove the fly glitch, and fix the very little bug traces that left around.
See? Helpful isn\'t it? :P
-
Originally posted by Moogie
Testing for bugs and reporting what happens is absolutely encouraged. And there is a very easy way to see whether someone is testing or exploiting something: watch them. They do it once or twice, good for them. They do it ten times, call the GMs. :)
As a programmer (not on this game, but at work), and often testing code, I have to disagree here. Once or twice is seldomly enough to figure out how exactly how to recreate a bug, narrowing it down can take lots of tries. And the more precise descriptions we can give the developers, the easier it will be to fix.
-
True, but most of the time you can verify the problem quickly on the main server, then recreate it to find out more on your own test server. Besides, it\'s best to do the testing on a server compiled from the CVS, just in case the area in question has been changed since the main server\'s build. Now, this isn\'t always possible, because test builds lack the full database. In the case that something is the db\'s fault, yes you\'d have to do allot of stuff on the main server. (or get access to the db yourself)
-
I guess my main reason for starting this is somehow from an ex-developers point of view IMHO there isn\'t enough to planeshift mechanics to support game rules (keeping in mind wipes specifically) to warrent GM attention right now.
Like when I was testing the NPC Combat AI Problems which just isn\'t a leave the range of attack, I found out. This action is a major flaw in the combat on the server side. I tried to submit a bug report and it got turned into a repeat problem but it isn\'t the same thing as leaving the range of combat bug. This is repeatable almost every time. But I thought I submitted it with full details on how to to it.
Maybe I wasn\'t clear in my submision. Well in the process of getting info about how I can stop a npc dead in his tracks and kill him while he stands there I got this warning about how I was exploiting a known bug. Now I don\'t know about you but I havent taken the 2 weeks that it would take to go thru all the bug reports and the forum to see if there are like bugs in them. And even if it was in there I wouldn\'t see the same reason for this bug. You can overload an npc with spamming the attack buttons and cause him to stop attacking. Now don\'t get me wrong I have seen both bugs clearly at work here. Not just one that looks like the other. I think it has something to do with attack ranges that are not equal or something along those lines for the out of range bug causing the npc to loose thier interest when I leave the threat range. The one bug I was looking at was keeping the npc in range but spamming the attack button in the process. The npc will halt and if you run up before it moves again he will stay frozen there open to attack and not attacking back.
But since there are rules both official and unofficial I digress from testing any further and meerly become another \"player\".
I have come to this descision because I like Planshift alot and I don\'t want to loose my connection to it either by banning or even getting the label of exploiter. If the makers of Planeshift realize that if they want the \"Testers\"
to continue there cannot be a \"Exploiter\" rule because thats what we are trying to figure out what are the exploits? How to recreate exploits and if those exploits have been fixed?
I got a return email from the bug tracker and it had some interesting comments on the bottom. Like: \"I thought this was fixed?\"
Rules are one thing for a game but when you impose those same rules to a tech demo then you are limiting your real testers to playing and not testing.
Persiffon
-
Actually, with respect to the bug you\'re talking about, it seems I wasn\'t clear enough. What you described, is a variation of the MOST commonly executed bug in the game:
Get an NPC to follow you; get close enough for it to stop, with you at the border of its attack range; hit the attack button a few times within range, then move slightly out while hitting; move slightly back in and attack; the NPC will not react.
I, and may others know the distance to be at so well, that they can just wiggle and do it perfectly, taking no damage. This is what I was talking about in bug 968. Stfrn posted that \"there are many variations on this\", and mentioned that you can just attack as it\'s approaching and can forget to attack. (basically, the method I listed above forces this) What you posted in bug 1058, and mentioned in the previous post here, is this same general flaw in the NPC client. It\'s basically dumb as a brick, that\'s all... Khaki posted in bug 968 that it\'d be fixed in the next server update (I think by extending NPC ranges), and because I (and many others) have been trying to clean up some of the extraneous posts on bugtracker, I noted that in 1058 and marked it as a duplicate. (I have yet to look at that area of the code, and was just quoting Khaki, really.) It\'s really just semantics whether you consider variations on a theme duplicates or not.
This is a well known problem, and testing it is a mute point till someone gets around to posting a fix for it. If you want to help test the bugs, that\'s wonderful. That\'s what you\'re supposed to be dong. Just build your own server from the CVS, and do your rigorous testing there. (that\'s what it\'s for) Major bugs are not to be tested repediately on the Laanx server.
-
I haven\'t been able to reproduce this bug yet, I thought extending NPC range would prevent any duplicates or associated bugs from cropping up but from this thread it looks like it hasn\'t worked.
Thanks especially to Persiffon and DaveG for describing it much more precisely than most of the other posts I\'ve read have. It\'ll definitely help a lot in helping us fix the issue and the right one this time. I\'ll keep you guys posted though if I make a breakthrough, luckily it\'s probably just a NPC client fix.
-
Originally posted by DaveG
True, but most of the time you can verify the problem quickly on the main server, then recreate it to find out more on your own test server. Besides, it\'s best to do the testing on a server compiled from the CVS, just in case the area in question has been changed since the main server\'s build.
I am not talking about the developers testing, as I am confident that GMs will not ban or otherwise punish developers for \"cheating\", no matter how bad it looks. I\'m talking about the rest of us testing something to figure out the exact way to reproduce a bug, to be able to submit a good bug report that will allow the dev to think \"oh, that\'s line 17243\" immidiately, without having to do their own testing, thus letting them fix it in three minutes instead of three weeks. We can\'t test on our own test server, because we don\'t have our own. We only have the main server.
-
PlaneShift is open-source. If you don\'t have a test server, feel free to make one. :)
Click here for the CVS Docs (http://cvs.sourceforge.net/viewcvs.py/planeshift/planeshift/docs/#dirlist)
Various build guides can be found at the location above. If you want some additional help, you can come to the IRC room at: irc://freenode/#planeshift-build (http://irc://freenode/#planeshift-build)
Build the client, server, and NPC client on your machine. Load \'em up, and test to your heart\'s content. ;) If you\'re serious about testing PlaneShift, this is something to do. Especially, because rigorous testing should really be done on the most recent code if possible, not the last released client. If the problem is a db issue (the CVS build only comes with a limited db, for testing only) you\'ll need access to the main db. (which you won\'t have) In that case, yes I agree, the main server is where you\'ll have to do your testing. If that testing involves the repeated exploitation of some major flaw, then I guess the best thing to do is to talk to a GM or dev about it first.
-
Not everyone is able to compile and get their own server running, even though they may be great testers. Programmers aren\'t the right people to test every aspect - they are great at testing every combination that makes sense, but for testing the combination that don\'t make sense, you really need what some people call \"lusers\".
I can\'t even compile the client, how would I ever get the server working? The PS code is very picky about what compiler one has installed.
And if the main server is not for testing, and it\'s not for playing (\"This game sucks, way too much is not working\" - \"you\'re here to test, if you want a finished game go buy UO\"), then what is it for?
-
I.. don\'t have a clue how to compile a server, and don\'t you need a \"server\" to compile a server?
Increase the attack range?? its already twice as large as ours and it isn\'t helping, and besides its unrealistic to see the rogue hitting you from across the room.
-
is it that far? i havent really had much chance for powerlevling over the last weeks or so...
but i must say i prefer that over the people killig then easy way thing except im going to guess that it can cause problems with people getting killed through walls especially with more aggressive monsters...
-
Well yeah I ran awayfrom a rogue and even though it was a bit of lag probably he hit me from about 5 meters away. literally. I died btw. X(
Yeah and the through walls thing.. actually people can attack through walls too they should change that later..
But the AI is really horrible as of now anyway, they can\'t talk, follow or hit.. :rolleyes: SO we can fix all at once :D
-
Originally posted by Neryam
I.. don\'t have a clue how to compile a server, and don\'t you need a \"server\" to compile a server?
I compiled the server on my laptop just fine. If you want to serve to many people at once, then you\'d need a more powerful computer, but for testing you\'re not doing that.
If you\'re not able to get your own build working, that\'s fine. (Though, it\'s probably not as hard as you think it is. Ask on IRC and people will help.) Most of the bug finding can be done on the main server, without too much of a problem. Though, if you\'re only testing on the main release code, you often end up reporting problems that have already been fixed. (if you think that may be the case, ask someone with a CVS build to check for you) It\'s just the repeated testing that becomes an issue.
-
i see what you mean about the range now.
(http://i7.photobucket.com/albums/y273/drey7/rNGE.jpg)
and there was quite a long pause between the blows so i dont think it was lag or something. i had moved back from the rogu, then stopped for potions then BAM dead.
-
I just made a fix to NPC combat behaviour today, hopefully they should act a bit more naturally. Can someone let me know if this is the case? i.e. is it still possible to glitch some npcs to non-responsiveness as you could before?
Thanks
-
i found two ulbernauts and a tefusang in ojaroad refusing to give up chasing someone who quite clearly was nowhere near them anymore, they were stumbling against a steep hill, moving them back shortly
-
yeah, they dont seem to give up the chase to easy, like if i run into the wrong type of tefusang in the arena, they will still be waiting for me if i drop in by them again a little later on.
and khaki, not sure when you made your fix but around 1pm our time people were still able to stick NPCs.
-
Ok the problem should finally be fixed now. Post here as usual if you can still bug NPCs.
-
I managed to bug a Trepor, but this time its a bit different, and I only managed to do it once. Theres a VERY small distance bug, if you stand in the right spot behind the Trepor, then they will not attack and you can attack them. Your AI fix stops NPC\'s attacking a certain range and stops them from moving, then after backing a little further they chase after you, you simply need to line up in the perfect distance where the NPC is froze, but you aren\'t too far away to attack.
Also they seem to use this against me, as I get the too far away too attack message, they are still hitting me, so clearly this fix has made huge distance glitches...
EDIT: Only used this on a Trepor, you could possibly reproduce this bug on other NPC\'s too, but since Trepors are shaped like bugs and have a wide space, its fairly easier to measure the distance point where you can exploit them