Would you like to see Aetolia have a coder specifically for bugs/typos/ideas?

Haven · April 2013

Can't forget the Queen of Bugs! @Moirean.

Moirean · April 2013

Haven said:

Can't forget the Queen of Bugs! @Moirean.

I know it's kinda a forum meme, but this seems a good thread to talk about it. I don't just magically spawn bugs. I worked in video game QA for 4 years (and if I ever move to Seattle or back to California, I'll likely work in it again), and that has given me a somewhat abusive innate approach to trying out new things in games. The first thing I tend to do when I see something new is try out limits.

A large majority of bugs are not really anything broken so much as outlying conditions not being coded for. In QA a big chunk of testing is just finding that, ideally before users do. For example, a quick pass before release might include using a Carnifex attack with a bastard sword instead of a bardiche, or using the new upturn command with all of the various possible container types. Obviously, trickier bugs also emerge, such as stress tests (what happens if we get 20 people in a room with vibes, ents and pit ticking?) or memory sinks due to bad lines in code overloading things when certain things (old catapult chains) are repeated, but a surprising number of issues - I'd even say most - can be caught with systematic pre-production testing.

I don't know what Aetolia's QA process is, but if there aren't test plans, I highly suggest that something like that be considered. A test plan is basically a generic template and systematic approach you apply to testing before it starts. For example, let's say a new skillset was released. You'd want to start at the highest level of testing and then break down the various possible conditions of the skill's use. Just gonna give some examples.

Category: Skillset X

Area: Skill Y

Sub-area: Test basic functionality of Skill Y

Test 1: Test unhindered skill use

- Test against unhindered target

- Test against hindered target

- Test against prone target

- Test against webbed target

- Test against paralyzed target

- Etc

Test 2: Test hindered skill use

- Test while prone

- Test while webbed

- Test while paralyzed

- Etc

Test 3: Test weapons with skill Y

- Test with all weapons skill is intended for

- Test with all weapons skill is NOT intended for

- Test without weapons

- Test wielding a non-weapon

- Etc

While following the test plan, you'd also halo test (test in the same general area as the specific testing assignment) to see if any related conditions bring up errors. This gives a semi-directed window for exploratory testing, letting testers do weird, less common stuff, while still focusing on the current test goal.

Like I said, I'm not sure what Aetolia's QA process is, but if test plans aren't used, it probably couldn't hurt to write up a few basic templates to apply when checking new skills and for use when helping to pinpoint where errors are coming from.

Also, to help out coders, we could probably revise our bug reporting system. Currently, we just type bug blah blah blah, but a slightly more in-depth setup could help admin figure out issues at a glance - user UI, especially for reporting, can help shape the kinds of responses you get.

For example, give players a way to enter these following options and then let liaisons tweak and edit them, meaning that coders have a lot of time saved when it comes to classifying and sorting bugs. You could use MXP for several of these to also speed up the entering and sorting.

BUG FULL - bring up the full report template

BUG: short summary of the issue here, eg ssl is not doing damage to fish NPCs

CAT: category, eg Savagery

TYPE: what type of bug is it, eg skill use, personal (only affects you), communication (issue with says/emotes/etc), items, website, combat, etc

RATE: rate the bug's severity on a 1-5 scale, eg 1 is making the game crash, 3 is a headache, 5 is a typo

EXPECTED: How should things normally function (eg, spinslash should hit mobs when I use it)

STEPS: Detail as many steps as possible to produce this bug. This not only makes things easier for coders to reproduce, it helps the person reporting the bug review what they are doing (in some cases, it shows them that they are just confused!). Eg, 1. Wield halberd (not bardiche), 2. Use pole spinslash against any npc who is of the type fish (for example spinelash fish in Nal'jin or salmon)

RESULT: Elaborate on the actual result here. Eg, "spinslash will give an error message about Slyphe preventing you because you suck" or "blank lines will be sent instead of the command"

Perhaps that is a bit too in-depth, but at the least a FEW more fields would help a lot with reporting more clearly. I admit, I get lazy and just fire off a few lines and don't bother to look at repro steps - but a more meaty bug report means that users will put a little more time into reviewing what causes the bug, and save investigative time, as well as make it easier for high priority bugs to be seen. I know there is already a classification and rating system in place, but afaik, bugs are mostly sorted by skillsets, versus more logical groupings of what is causing them. A bug dealing with item interaction probably has a different type of code than one which handles says and emote colors.

Anyways, ramble about QA aside...

I don't think Aetolia is that bad off, bug-wise. Avalon is currently unplayable due to bugs; Aetolia has some minor ones (pulverize curing aside, grrr) and they tend to be handled fairly quickly. I see constant turnaround on bugs, and that is always awesome. A full-time paid staff coder (which would include bug fixes) would of course be great, so obviously I'd like to see that if the budget could handle it. Hiring someone just for bugs? No, that's a waste of resources. But hiring a coder for general coding would be great, of course, and it would mean we'd see faster content releases. That's a money decision, though - balance the income from more releases vs the cost of employing the person - and not something for players to decide, since we don't know what the game's finances are.

Zun · April 2013

If we're expanding the bug system, a space to put pastebin logs wouldn't go amiss.

Serrice · April 2013

@Moirean: Liaisons should be getting a pile of work soon as mentioned by Raz in his March post, so that systematic plan you posted would be suuuuuper helpful. Thanks!

Moirean · April 2013

Also, give liaisons access to a pool of dummy accounts. These would be secondary accounts at a range of levels with access to liaison arena, enabling liaisons to log in on the secondary attack to have a dummy to test things against, such as damage taken, how skills are functioning, etc. It seems pretty inefficient and unwieldy to have to require a second actual player to just be your crashtest dummy.

Arbre · April 2013

Moirean said:

I don't just magically spawn bugs.

ftfy

Osmond · April 2013

Honestly, there are bigger fish to fry than typos and minor bugs. IRE needs dedicated coders that are fixing the current imbalance amongst the classes (which from I have read in the announce, they are working on that, awesome) and giving some fresh content. The game is getting kinda stale. The Shamans and the Teradrim were a good step, but we need some meat and potatoes here. A ship system, a new badass war system, something to create some new conflict.

Serrice · April 2013

Some people would argue the exact opposite, that all the new stuff is just bread and circuses and not resolving all the broken things.

Moirean · April 2013

What severely broken things are we talking about here? Aetolia seems very stable and bugs are minimal and addressed fairly quickly (except for pulverize AHEM!). Are there some huge glaring bugs that are lurking right now?

To give a counter example, before I came back to Aetolia, I was playing Avalon. On any given day, says/tells might stop working, you couldn't pick up items you dropped, you'd get stuck in a permanent web you couldn't struggle out of, you wouldn't be able to use your main class attack, your prompt might vanish, you could happily run a script to get infinite herbs and poisons and guild tutors would run around attacking newbies on sight.

Aetolia is stable and awesome. -_-

Serrice · April 2013

If that question is directed at me, I'm just offering a counterargument to Osmond's assertion. If it's not, oh well.

As for myself, I'd rather see all the bugs that I see fixed just out of principle, before new content that might also introduce new headaches is released.

Moirean · April 2013

Well it was mostly a general question, since there is a poll/thread about the state of bugs. I was wondering what severe ones there were riddling the game.

Haven · April 2013

Moirean said:

Well it was mostly a general question, since there is a poll/thread about the state of bugs. I was wondering what severe ones there were riddling the game.

I think he meant that the combat imbalances between the classes in the game are / should be a more pressing issue than the abundance of 'minor' bugs.

Lanira · April 2013

I'm going to disregard the topic of bugs mainly and talk about what I think the admin should be focused on.

I'm of the opinion that most of the bugs in the game aren't really an issue. I also don't think class revamps should be the main priority either; Though, they are very important. I believe there's still a gaping hole in the game when it comes to org vs org conflict. I liked the war system for it's ability to create tension in the game, it formed alliances and bitter enemies while also adding life to each city's inner politics. Obviously, it was also very griefy and, in one way or another, everyone was forced into it. I know we have focii now, and I think it was a great addition to the game; But, I also feel it's missing a sense of continuity and progression, they're really just disconnected, individual sets of team fights.

I want something to conquer, to defend, to strategize over, and to care about. But, it needs to feel epic without taking over the normal world. I think the new (lol) continent would be the place for this. This could also allow for new factions and allow people to not just be defined by to which city they may or may not belong (let being cityless actually be an option). Whether it's a continuous system of conflict or a series of long/medium/short conflicts, it needs to take into consideration that not everyone will want to participate, and those who do participate need to sleep. I think a good place to start would be to look at the other IRE games and figure out what is working for them, not to copy them, but to further understand what motivates and gets players excited.

Daskalos · April 2013

I agree in the sense that while class revamps are nice, I think we're a bit too caught up in 'Achaea has it and we shouldn't' - I much prefer the way classes like the Luminary have evolved - instead of scrapping it, it's been slowly reforged over time into something way different from the other games, but still retains some elements. In the end, you've got a really well rounded class. The new classes ARE a lot of fun, but considering the average online size of our guilds is, what, maybe 20 people a day total in the biggest guild? You're only affecting a small portion of the playerbase.

Don't get me wrong, I'm really, really happy with the administration, I guess if it were me, for everytime we're doing something that benefits a certain organization (the Dominions, a class revamp, et cetera) I'd also have someone else working on something that benefits everyone else (more ways to gather ylem, improve the Sect, et cetera).

That being said, the two things I'm looking forward to most are supposedly happening, so patience is a virtue they say. Come on new ylem ways and factions!

Howdy, Stranger!

Quick Links

Categories

In this Discussion

Would you like to see Aetolia have a coder specifically for bugs/typos/ideas?

Comments