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

Joshua Calvert
|
Posted - 2005.08.14 07:38:00 -
[31]
Sod the jump to option.
Just fix evemails.
LEEEEERRRRRRRRRRROOOOOOOOOYYYYYYYYYYYYYYY! |

Professor McFly
|
Posted - 2005.08.14 09:14:00 -
[32]
Now if only you could still transport dreads inside freighters, this wouldn't be so bad...
"enemy fleet sighted!" "tell em to wait a minute, we're still unpacking the dread. Someone get me a screwdriver!"
|

Eight Ace
|
Posted - 2005.08.14 10:24:00 -
[33]
My commisserations to the QA department.
I know a hard sell to get "we need an exact copy of our production environment to catch stuff like this" past the beancounters.
People who don't know QA should keep their opinionated waffle to themselves. Anyone who disagrees can post a comprehensive test plan that will catch the next cat 1 defect before it ships to production.
~Shipping bugs for 15 years and still smiling~
|

JoCool
|
Posted - 2005.08.14 10:42:00 -
[34]
Lol, class 
|

Thrak
|
Posted - 2005.08.14 10:46:00 -
[35]
Edited by: Thrak on 14/08/2005 10:47:35
Quote: People who don't know QA should keep their opinionated waffle to themselves. Anyone who disagrees can post a comprehensive test plan that will catch the next cat 1 defect before it ships to production.
Well, I admit I'm no expert. But
Quote: a couple of GMs tested jump drives on TQ.
doesn't seem that comprehensive. As for "an exact copy of our production environment", you might not have noticed but since the cold war patch GMs have had bountiful (23/7)access to TQ, which is er, the EXACT production environment.
Thats several weeks, to test jumpdrives, on TQ, across several nodes if needed. Wasn't done.
But I do agree with some of your sentiments.
Quote: People who don't know QA should keep their opinionated waffle to themselves.
This applies to you.
|

Discorporation
|
Posted - 2005.08.14 11:21:00 -
[36]
This made me rofl ;p
[23]
|

Eight Ace
|
Posted - 2005.08.14 11:46:00 -
[37]
It's a nasty miss for sure but I have to feel some sympathy seeing as it's what I do for a living as well.
As we have been told the error is caused because "the events about gang members having active fields are not making it over node boundaries"
You need a pretty low level of test to catch that. If it wasn't such a complex environment-sensitive system I would be thinking this should have been caught at the unit test phase by the developer, but of course at that level the events are always gonna make it because they most likely won't be crossing node boundaries on the dev's desktop.
You would need to have a fairly in-depth knowledge of the development methodology, and capabilities of the QA team within CCP before you could confidently start assigning responsibility for this.
Actually that's a bit of an oversimplification. The responsibility of course lies with the QA manager, the FAULT is the bit that's sometimes hard to pin down.
If the test plan for this was simply to let some GM's play with the code on the live system then yes that's pretty shocking.
Whatever they did on tranq, most likely they would be doing this in very quiet space so every man + dog wasn't there gawping at the new ships, unfortunatley this could well be connected to the issue itself and you need every man+dog to engineer the conditions in which the defect would manifest itself.
If this was my team I would be mightily embarassed but given the nature of the defect I wouldnt be breaking out the P45's.
Sorry if I am going on a bit. I get upset with "OMGWTFBBQ QA SUXXOR" kinda rants. Understandable as a paying customer but I also appreciate the other side.
|

Nafri
|
Posted - 2005.08.14 11:54:00 -
[38]
since my dad is QA-Mananger he always told me: Always consider, if there is a slight chance for a failure, this failure will happen.
This node thing must have been know... but still funny 
Your bla bla hit bla bla for bla bla damage. Wanna have some bubu now? |

Viceroy
|
Posted - 2005.08.14 11:54:00 -
[39]
I believe this is a case of severe "lollerskates"  __________________________
Finite Horizon Your end is our beginning.
|

Eight Ace
|
Posted - 2005.08.14 11:56:00 -
[40]
If you press your Dad a bit further he will probably also say that it almost never happens for the people who are meant to find it  |

Faramir
|
Posted - 2005.08.14 11:58:00 -
[41]
Yeah, what are the odds one would travel from 1 node to the other? 
|

Nafri
|
Posted - 2005.08.14 12:04:00 -
[42]
Edited by: Nafri on 14/08/2005 12:07:49
Originally by: Eight Ace If you press your Dad a bit further he will probably also say that it almost never happens for the people who are meant to find it 
But thats where he makes plans to find all possible plans, unlike in Computer games, other companies wont take a few days delays couse of a "mistake".
If such thing happens his company would been dead before you can say "lollercoster"
A test plan for Eve would be like:
Feature X, based on Y
Test server: test function of X -> if it doesnt work -> test it depending on Y -> still not working -> back to programming
if function X works -> test it depending on Y -> works like intendend -> bring it on live server
live server: repeat above
Your bla bla hit bla bla for bla bla damage. Wanna have some bubu now? |

Valar CCP
|
Posted - 2005.08.14 12:15:00 -
[43]
Before it was deployed on TQ, this feature was tested in a multinode enviroment on Singularity, and then again last night. However, all attempts to node jump on Singularity were successful. So this bug is not consistently reproducable. Even during my tests on TQ last night, I once managed to do the jump. But it seems that the event if not getting through in all cases.
------ Valar Quality assurance department CCP games |

Del Narveux
|
Posted - 2005.08.14 12:23:00 -
[44]
You guys might want to do some sort of in-depth testing of all the TQ nodes (maybe during a DT or something) to make sure theyre all working together and whatnot. Something like this, makes me wonder if it could be connected to the UI bugs, server lag, and other random stuff associated with potentially cross-node stuff...like one of the nodes isnt communicating right or something.  _________________ [SAK] And Proud Of It! -- aka Cpt Bogus Is that my torped sig cloaking your base?
Originally by: Wrangler Well, at least we have forum PvP..
|

Eight Ace
|
Posted - 2005.08.14 12:44:00 -
[45]
Actually that's interesting,
Valar says this thing isn't consistently reproducible, but on the front page Bubbles is saying they have not had a sucessful jump in two days. That sounds reproducible to me.
Again you would need to have in depth knowledge of the code and exactly how this is implemented to be able to nail it down. If I wasn't at work right this second testing fairly complex software I would be well pleased to log onto singularity and help out.
You may want to start by pulling up the locations of all the dreads in existence on live right this minute, porting the data over to the test environment and jumping those about on the test environment - starting with Bubbles.
For the cynics there's also always the possibility that the actual root cause of the defect does not lie within code maintained by CCP and some kind of work-around is needed. |

Dukath
|
Posted - 2005.08.14 12:55:00 -
[46]
Does anyone know what happens when the ship creating the cynosurial field gets destroyed?
Since the dread jumps to member... will he end up in yulai if the guy making the field gets podded while the dread jumps? does the dread end up at the cynosurial field or at the guy creating the field?
Did quality control try out these options?
|

Amthrianius
|
Posted - 2005.08.14 13:43:00 -
[47]
you cant shoot a person generating the field can you? At least thats what i thought... ---------------
|

Dukath
|
Posted - 2005.08.14 13:59:00 -
[48]
You cannot lock him when he has a field up. If you have a lock and he sets up a field the lock stays and you can activate modules on him (didnt have gun fitted but the nos did take away his cap)
I really hope the quality control didn't make assumptions like that since assumptions are often wrong :)
|

Faramir
|
Posted - 2005.08.14 14:00:00 -
[49]
Originally by: Valar CCP Before it was deployed on TQ, this feature was tested in a multinode enviroment on Singularity, and then again last night. However, all attempts to node jump on Singularity were successful. So this bug is not consistently reproducable. Even during my tests on TQ last night, I once managed to do the jump. But it seems that the event if not getting through in all cases.
May i ask how we are to determine a "route" to move Dreadnoughts?
It seems the information on this is worse then "sketchy"... From rumours i heard it was 2,5 lightyears as base jump-distance. But the best part is that the map does not show any indications of lightyears.
Also Dreadnoughts have no "jumprange" indicators...
This means you can "guestimate" how far you can go, but you won't be certain till you have a cynosural field activated. However if the field is too far away, you are waiting another 10 mins before you can find an alternate route/system...
I think this seems pretty much unfinished and rushed....
When will we actually now how, where and be able to use jumpdrives properly?
|

Redblade
|
Posted - 2005.08.14 16:09:00 -
[50]
Originally by: Valar CCP Before it was deployed on TQ, this feature was tested in a multinode enviroment on Singularity, and then again last night. However, all attempts to node jump on Singularity were successful. So this bug is not consistently reproducable. Even during my tests on TQ last night, I once managed to do the jump. But it seems that the event if not getting through in all cases.
I know u like to be restrictive with new stuff in sisi but tbh i think stuff like this could be caught quite easy if people where allowed to use the ships before release as new stuff will be tested by people who realy want to check the usefulness and what the limits are.
So in short, if dreads where avalible on sisi before launch u can bet your ass on that alot of people would have jumped around all over the place and found this bug.
This ofc with some help with skill boosting.
Killboard |

Hllaxiu
|
Posted - 2005.08.14 16:31:00 -
[51]
Originally by: Redblade
So in short, if dreads where avalible on sisi before launch u can bet your ass on that alot of people would have jumped around all over the place and found this bug.
That is of course assuming that the bug existed on singularity - Valar's comments seem to indicate it wasn't. If he is correct and it only happens sometimes on the multinode live environment then anyone that wants to move a dreadnaught is going to have to spend many hours to get that one lucky jump. 
Proud member of Elite Academy. |

Redblade
|
Posted - 2005.08.14 16:38:00 -
[52]
Originally by: Hllaxiu
Originally by: Redblade
So in short, if dreads where avalible on sisi before launch u can bet your ass on that alot of people would have jumped around all over the place and found this bug.
That is of course assuming that the bug existed on singularity - Valar's comments seem to indicate it wasn't. If he is correct and it only happens sometimes on the multinode live environment then anyone that wants to move a dreadnaught is going to have to spend many hours to get that one lucky jump. 
Sisi = multinode so the issue shuld be there aswell so the more people jumping around the biger chance of finding the bug, and i think alot of people if given the chanse would have moved dreads around quite alot of jumps.
Killboard |

Hllaxiu
|
Posted - 2005.08.14 16:51:00 -
[53]
Originally by: Redblade Sisi = multinode so the issue shuld be there aswell so the more people jumping around the biger chance of finding the bug, and i think alot of people if given the chanse would have moved dreads around quite alot of jumps.
Valar said that the bug didn't appear on SISI though, and theres apparently a node line very close to pf-346...
Proud member of Elite Academy. |

Redblade
|
Posted - 2005.08.14 17:02:00 -
[54]
Originally by: Hllaxiu
Originally by: Redblade Sisi = multinode so the issue shuld be there aswell so the more people jumping around the biger chance of finding the bug, and i think alot of people if given the chanse would have moved dreads around quite alot of jumps.
Valar said that the bug didn't appear on SISI though, and theres apparently a node line very close to pf-346...
"So this bug is not consistently reproducable. Even during my tests on TQ last night, I once managed to do the jump."
That makes me assume that it could just aswell be on sisi hence why my post sugest what it sugests.
Killboard |

Letouk Mernel
|
Posted - 2005.08.14 17:47:00 -
[55]
Heh, hindsight 20/20 vision, of course now that the bug has been discovered, a dozen things could have been done to prevent it. They couldn't have very well designed a procedure to detect this bug without KNOWING about the bug, like y'all do.
We laughed, it motivated them to fix it and justify their actions, done. Let's move on.
Tell you what they need... they need to set up the end-all macro program, one where they can program 3-500 different actions that a player might take (any interaction with the UI), and then this macro can do them repeatedly and in a random order. Would totally eliminate any assumptions about players doing things in a certain order, and maybe find combinations that crash the game (hypothetical: open market while self-destructing your pod? crash! heh).
|

Eight Ace
|
Posted - 2005.08.14 17:59:00 -
[56]
Said macro program is called Compuware QA Center / Rational tool suite ... and you really don't wanna know how much these suckers cost, you could be looking at a hundred grand before you even blink, expecially on the kind of kit CCP are using.
Usually that's an even harder sell than 'can we have an exact duplicate of the live environment please'
|

Dukath
|
Posted - 2005.08.14 18:45:00 -
[57]
Don't think you can test thing like that with a macro. There is intelligence needed for testing like this, its not a gui but a concept you are testing.
What happens when the guy creating the cynosurial field tries to jump out or dock? I know you can't activate a module but is jumping or docking the same?
|

Redblade
|
Posted - 2005.08.14 19:38:00 -
[58]
Originally by: Dukath
What happens when the guy creating the cynosurial field tries to jump out or dock? I know you can't activate a module but is jumping or docking the same?
That's what im talking about =)
I most defenatly think that some one would have tried that if it was avalible on sisi before launch and som other dude might have tried to start the cynosurial field while he deploy drones,attack a target with the drones and cloak while trying to dock, not because it would do him any good, people just try strange stuff when new toys are given to them to see what works and what doesent.
Killboard |

Zungen
|
Posted - 2005.08.14 19:38:00 -
[59]
Originally by: Dukath You cannot lock him when he has a field up. If you have a lock and he sets up a field the lock stays and you can activate modules on him (didnt have gun fitted but the nos did take away his cap)
I really hope the quality control didn't make assumptions like that since assumptions are often wrong :)
no but they can smartbomb you :P had that happened to me lol
|

bUBbLeS
|
Posted - 2005.08.15 14:17:00 -
[60]
are we there yet?
Julius ceaser : "operor vos volo MCCCXXXVII laganum bUBbLeS?"
|
|
|
Pages: 1 [2] 3 4 :: one page |
First page | Previous page | Next page | Last page |