|
I'm reading through the thread for the first time (this is the first time i've actually cared enough to read through an entire thread THIS long) and I just wanna post to say, holy poo poo this is soul crushing
|
#
¿
Aug 31, 2020 18:22
|
|
|
|
| # ¿ May 14, 2024 03:09 |
|
|
Lodin posted:Lol what a poo poo company. Still using tiny CRTs in 2004 or whenever this timeline takes place. This is from a few months ago but fun fact: I worked a QA testing job in LA for a major AAA game company in 2010, and they still used CRTs for the vast majority of the testing stations. It even looked quite a lot like this: Except each station also had a computer, and a VCR. PS3 and Wii consoles didn't have nicely built-in recording software like Xbox 360 did, so the cheapest way for the company to enable us to record footage was to literally record it on VHS then upload them to a computer. e: Tohru's interview process is fairly accurate as well, at least compared to how things were done over here at the time. In-person talky interview, then they'd take you into a room with a lovely CRT connected to a console, they'd give you an old build of a recently-released game, and have you physically write bugs down about it. e2: the inaccurate part so far is that the second job for the tiny shovelware company pays crap. Generally smaller game companies have only a handful of QA testers, and thus pays them fairly well since they need to be more skilled. It's the big corporations that have teams of several dozen testers that pay them like poo poo cus who cares, we're going to replace you the first time you break a plate anyways. CodfishCartographer fucked around with this message at 22:08 on Aug 31, 2020 |
|
#
¿
Aug 31, 2020 21:32
|
|
|
Oh gently caress now i'm caught up on the thread and don't know what to do anymore. Uh, post more NEET!
|
|
#
¿
Aug 31, 2020 23:05
|
|
|
I was sad after the revelation that Everybody's Gorilla = Everybody's Golf, the translation didn't change to Hot Shot's Gorilla
|
|
#
¿
Sep 1, 2020 02:30
|
|
|
BattleMaster posted:Yeah NEET's reaction is that she has a "pretty" name but is astoundingly ugly, not that it's necessarily a weird name or that she called herself "princess" as in the title. On the other hand, Hime doesn't read as a "pretty" name in English. I think Princess works - it comes off as just a weird name in general, but also ill-fitting due to her looks (in neet's opinion)
|
|
#
¿
Sep 1, 2020 05:27
|
|
|
The stunning revelation that you can freeze vegetables
|
|
#
¿
Sep 3, 2020 15:25
|
|
|
Gwyneth Palpate posted:That's actually pretty clever -- keep around a build with bugs that you know about to train new employees. Unfortunately, this is a fictional story; QA departments don't actually exist in real life. The AAA studio I worked QA at actually did this. The first week was spent entirely on training using the old build and an insulated bug database for us to write lovely bugs in.
|
|
#
¿
Sep 6, 2020 16:50
|
|
|
C! ... Or G!
|
|
#
¿
Sep 7, 2020 08:09
|
|
|
Lodin posted:Since when do you need that much training in writing a bug report? You have forms that ask: CodfishCartographer posted:The AAA studio I worked QA at actually did this. The first week was spent entirely on training using the old build and an insulated bug database for us to write lovely bugs in. Its likely all this training is more for how to actually test than writing reports, but there easily could be a full day of training on bug reports. Major publishers spend a lot of time on training QA. Professional testers need to be able to accurately reproduce bugs and be able to whittle down the steps that do/n't cause it to occur, and then clearly explain those steps so anyone can follow them. This sounds easy enough, but a) people are REALLY BAD at explaining things clearly, and b) it can take a lot of critical thinking and puzzle solving to be able to narrow down reproduction steps.
|
|
#
¿
Sep 7, 2020 16:55
|
|
|
Crazyweasel posted:I’ve read so much NEET that I’ve finally reached enlightenment. I went to the store and had some fleeting mundane thought about why one of the automatic doors was broken and the entrance was closed that it struck me how the comic is created. NEET then proceeds to buy way too many frozen meals, filling his cart with them (and also bags of chips)
|
|
#
¿
Sep 8, 2020 20:00
|
|
|
WHAT
|
|
#
¿
Sep 9, 2020 00:02
|
|
|
NEET gunna have a fuckin heart attack
|
|
#
¿
Sep 10, 2020 00:12
|
|
|
NEET hasn't ever thought this much in a single day before
|
|
#
¿
Sep 10, 2020 04:01
|
|
|
We Kaiji now
|
|
#
¿
Sep 18, 2020 01:48
|
|
|
RoeCocoa posted:Is this the first time we've had NEET ~in color~? Nah quite a lot of panels have had red in them, usually to highlight moments
|
|
#
¿
Sep 18, 2020 02:27
|
|
|
Devils Affricate posted:So they're secretly recruiting speed runners on the cheap, nice grift I've mentioned in this thread before, but this depiction of QA testing is VERY accurate for a AAA studio, this sort of thing included. When on a tight schedule and you get a new build every day, you want someone who can play through the critical path of the game as fast as possible to ensure the devs didn't break anything along the way. Not necessarily world-record speed running, but it really helps to be able to beat the main story in a single shift.
|
|
#
¿
Sep 21, 2020 06:09
|
|
|
Aino Minako posted:Literally no writing and only the need to form a marginally coherent thought. Also consistent with the tester work experience? I know this is just a joke about QA lyfe, but since I can't stop writing about QA testing: it depends on the place, from my experience. Most places would use something like this as a teaching opportunity though, like "okay, let's walk you through how to search the bug database and how to write up a new ticket" sort of thing. If they really just want to focus on the newbies learning the game, or they're pretty tight on schedule, I could see them just passing the bug-writing off to someone experienced enough to knock it out quickly on high-priority bugs like that. Don't want a super important bug like that to read "It crashed I guess idk" instead of "Game crashes when you X Y Z" However, some companies (the lovely ones) have bug quotas, and if this place is that kind of company then this dude could in theory be tricking the newbies into giving him (or his friends) more bugs to pad out their numbers. With how sinister people are in the criminal timeline, I'm used to being paranoid like this with new characters, but the lead dude seems on the up-and-up so I doubt he's trying to be a dick. Hopefully.
|
|
#
¿
Sep 23, 2020 03:51
|
|
|
Fumaofthelake posted:Krillin is the strongest human who has ever existed, put some respect on his name 
|
|
#
¿
Oct 1, 2020 01:07
|
|
|
RandomFerret posted:No, that is expressly forbidden in point 2. It really is heaven lol as if that ever stops anyone microwaving fish
|
|
#
¿
Oct 2, 2020 02:33
|
|
|
Oh boy
|
|
#
¿
Oct 4, 2020 17:47
|
|
|
Honestly, "use every skill and spell on every monster" is a fairly antiquated way to test. While yes, there could be a bug when you use a specific spell against a specific monster, it's so unlikely that the time spent checking all that is probably better spent elsewhere. The logic of "but you need to check everything!" leads to an endless rabbit hole: well, we've checked every spell on every monster, why not check every spell on every monster with every party member in the group? And then in every area? And then after using every item? Etc. Instead the more sensible way to test things like this is to test single individual instances of each of the things that could break. In this case, you'd probably have someone test all the fire spells against a single enemy weak to fire to ensure they all are actually flagged as the correct element, then do the same with every ice spell against a different enemy weak to ice, then every lightning spell, etc. That confirms all the weaknesses work and that each spell is flagged properly. A separate test would be to test a single fire spell against an enemy neutral to fire, one resistant to fire, and one that absorbs fire, then repeat for other elements - that way you check that each of the elemental interactions work. Maybe run a third test to check all the elemental weaknesses of each enemy, but only using a single spell of the relevant element per enemy to speed things up. Honestly though a "fight all enemies" test may be superfluous because you can expect all enemies to be fought through other testing such as checking different areas, it's smart to give multiple areas of focus at once - so "check each enemy weakness while testing collision of area 5" or whatever. Yes this method of testing leaves gaps where bugs could exist. Maybe the devs hosed up setting up one specific enemy somehow, but it's such an edge case that spending hundreds and hundreds of man hours looking for it is just wasting time that could be better spent testing features more likely to break (or more likely to have a bigger impact if they do break, a spell not working properly against a random enemy will likely just result in it not doing the intended amount of damage, boohoo) All that being said, poorly made test plans are absolutely run this way, and NEET may very well be working at a place being poorly ran like that. Especially 10+ years ago, the mentality of "we need to check every combination of everything!!" was quite common.
|
|
#
¿
Oct 6, 2020 15:28
|
|
|
Shadow0 posted:According to the timestamp, this panel is from 2008, which is 12 years ago. NEET needs to stop living in the past. Yup. I also realize that "test everything lol" also makes for a more compelling story / comedy for our poor dumbass NEET so I'm cool with it
|
|
#
¿
Oct 6, 2020 16:11
|
|
|
bad posts ahead!!! posted:don’t say this ?
|
|
#
¿
Oct 6, 2020 17:27
|
|
|
Aino Minako posted:I strongly suspect the answer to this is “no,” but could they be using dev hardware that could at least save the system state, allowing them to revert before a battle and making it unnecessary to find the same enemies over and over? lol I haven't tested on PS4 or XBone but test PS3s and 360s didn't have any built-in functionality like this. However, many games have debug tools to help smooth things over, but it varies wildly from game to game, even from the same studio. Sometimes you'll get awesome debug tools that let you access almost everything in the game quickly and easily, from playing cut scenes on demand to jumping to any location to spawning weapons and armor. I haven't tested a jrpg, but I would expect it to at least have the ability to jump to specific areas / dungeons and manually adjust exp and gold. Maybe the ability to encounter specific enemies if NEET is very lucky. Unfortunately, save states / rewinding are pretty rare for debug tools, however often there will be ways to pull/push a save from a previous build. Often this means the qa team will have a library of save files that were made "naturally" and can be pushed to more quickly access areas for test. Using FF7 as an example, they could have a save file after the first reactor, after the second, after leaving midgar, etc. The downside of this is these saves are usually from earlier test builds, and thus can occasionally result in false positive bugs or can just stop working. Of course, later in a game's lifestyle (usually in the last 3ish months before release) debug will often be stripped out of a game so it doesn't release with debug enabled (this is absolutely happened lmao) so if this company is that late in the life cycle (which they could be, often that's when lots of hiring occurs to crunch) then testing has to happen the old fashioned way, which means for a jrpg getting assigned an endgame area must suck hard lmao E: in NEET's situation I would expect them to provide him an endgame save file with every area and ability available, then he just gets to go through them all one by one. But as the above poster said its very possible they'll just make him grind that out on his own CodfishCartographer fucked around with this message at 00:54 on Oct 7, 2020 |
|
#
¿
Oct 7, 2020 00:46
|
|
|
Brother Make posted:
Gotta say, one of the best parts of being a lead tester is finding some lovely repetitive task that you hate doing and telling the coworker you hate to do it.
|
|
#
¿
Oct 7, 2020 18:13
|
|
|
nullEntityRNG posted:No. Obviously he will do a few hours of work, have a mental tirade about "This isn't efficent and besides no one actually would try to use x on y" and fill out the rest as passed. It's surprisingly easy to slack off on a list of test cases only to just bulk Pass them at the end. If something turns out to be broken, just say it was working fine when you checked it!
|
|
#
¿
Oct 13, 2020 07:10
|
|
|
Hatsune Mike posted:Make siblings:
|
|
#
¿
Oct 13, 2020 23:27
|
|
|
End of Shoelace posted:Honestly, this job is still piss-easy. I'd take it in a heartbeat. Yeah doing checklists like this is generally really easy, especially when the debug tools are as nice as it seems the ones they have are. The problem comes from the monotony - sure, doing a couple of these kind of spreadsheets aren't bad, but when you have to do them once or twice a week for months (or years) then it starts driving you insane. NEET may be able to struggle through this one, but I'm guessing he'll have some kind of existential crisis (or just quit lol) when he realizes this is what he'll be doing every day for the foreseeable future. I also wonder how NEET will handle the part of testing that actually takes some kind of skill - investigating a bug to try and puzzle it out. I wonder if he'll enjoy it or just fail miserably
|
|
#
¿
Oct 16, 2020 17:35
|
|
|
Podima posted:Hope all is well, Sister Make! Sister Make also quit weirdly suddenly
|
|
#
¿
Oct 16, 2020 20:32
|
|
|
He's had to do an annoying repetitive test for one day
|
|
#
¿
Oct 18, 2020 01:07
|
|
|
"Stupid buggy game" he says, resetting the console before returning to what he was doing without writing anything or telling anyone
|
|
#
¿
Oct 18, 2020 18:18
|
|
|
Gwyneth Palpate posted:They must not have Jira in the NEET timeline. JIRA was definitely around back then, but that was also early enough in its life that it wasn't the de-facto bug tracking software yet, and you occasionally would get companies using other programs. HardDiskD posted:Now this is an odd procedure I assume you mean having a computer at a separate desk? That is odd, but I have worked QA before that had neutral stations like this for testers to use. The job I worked at it was used as a capture station to take your VHS tapes to, since PS3 didn't have any built-in recording software. I haven't heard of a place with a dedicated "here is where you write bugs" station but I could see it. Might be more common in Japan?
|
|
#
¿
Oct 19, 2020 16:41
|
|
|
Skelejohn
|
|
#
¿
Oct 20, 2020 04:46
|
|
|
projecthalaxy posted:If nobody else bugged it, it's probably not important, Tohru. Better to ignore it.
|
|
#
¿
Oct 20, 2020 18:17
|
|
|
Holy poo poo the physical bug reports weren't just a thing for training??? That seems insane to me, I wonder if that's common for Japanese QA
|
|
#
¿
Oct 21, 2020 02:17
|
|
|
NEET even if the bug is a dupe, you should always check to see if you can add additional context or info with a comment. You amateur. You absolute imbecile.
|
|
#
¿
Oct 21, 2020 23:38
|
|
|
Straight White Shark posted:
That's also terrible bug writing. There aren't even any repro steps!!
|
|
#
¿
Oct 22, 2020 20:24
|
|
that's not what a softlock is! Tohru you're going to make your career go up in flames this way!
|
BigBadSteve posted:
I know this is a joke but: It's generally not terrible to try for a few repros before searching the database for a bug, especially big ones like this. Having more info can help when searching for the bug - for example, the game hard locks when NEET here uses Meteon on John Skeleton, but that crash may occur when using ANY skill on ol' John skelly, or when using Meteon on any enemy. If NEET just looks for "Meteon on John Skeleton" then he could miss and write up a dupe. Although there's always some lovely tester who never fully investigates a bug and writes up a dupe like that though Also while the "correct" thing to do is actually reproduce a bug 10 times, lmao nobody actually does that poo poo unless the bug is super inconsistent. Most testers will just get to 3/3 and call it good, while doing a couple halo tests for similar issues.
|
|
#
¿
Oct 23, 2020 01:20
|
|
|
big cummers ONLY posted:Hey this is a great bug write up tohru! You're a natural. Do you want to get your penis hosed, Tohru? I'm not sure which I think would be funnier: her giving him the mildest feedback possible and him taking it like her saying he's worthless, or her saying it's a good bug and him thinking it means she's in love with him
|
|
#
¿
Oct 23, 2020 20:22
|
|
|
|
| # ¿ May 14, 2024 03:09 |
|
|
I love donuts! Jelly-filled are my favrotie!
|
|
#
¿
Oct 26, 2020 05:11
|
|