Can We Conduct Testing Before Rolling Out Updates? Dont We Have Hero Beta?

[spiral root]

AI-translated. Some terms may be inaccurate—your understanding is appreciated.

I noticed this bug just one and a half minutes after the update went live. Why is it that I can summon two ECs but only be able to renew one of them?

Cant We Rolling Out Updates on Hero Beta first to test the bugs, or open up few WC servers for players to test?

Regarding this incident, I can conclude that the management team has not fully tested the revised game at all.

[vocal zinc]

I believe their reason for not using Hero beta was because it costs money to run the servers but I do agree with your point of absolutely needing to playtest things

There's so many bugs that we can say for sure not a single thing was playtested

[tranquil dome]

Just wait untill he learns about 1minute timer EC

[twin pier]

not like we can playtest if things like raids outright don't work or just hidden cough wardrobe wonders aspects

[amber hamlet]

i dont think this fix to what happened with this update is to make every update a hero beta update

and I dont think that hero beta wouldve even fixed this, because (lack of knowledge) I would guess this was just an issue with pushing the update to production, as the other explanation is that QA didnt finish a single raid during testing which doesnt seem quite possibly

[spiral root]

It’s true that launching Hero Beta for every single update isn’t realistic. However, I think involving players in some of the testing phases would also relieve the current pressure on the development team. Another viable option is to open up a few select WC servers for players to test.

[keen ermine]

this is the beta

[velvet iron]

It is important to know that often times I see with updates seemingly random things break from being pushed from gm servers to live servers, so there is only so much testing that can prevent that.
A good example is the ghost slab blocks on the path to Church in SE. I have heard they work perfectly fine on gm servers to but here they break, and no one knows why

[spiral root]

Mysterious bugs💀

[pseudo linden]

pretty sure theres still some ghost blocks on the pathway in Cinfras

[spiral root]

Perhaps rolling out the update to a selected number of live servers in advance for testing could resolve this kind of issue. This is just a guess, though—I don’t know if this method is feasible or if there are any technical obstacles.

But I can already picture the development team scratching their heads in front of their computers over this.

[faint tiger]

🥀

[amber hamlet]

theoretically this is what a staging environment is for but i do understand that shit just happens

[pseudo linden]

shit just happens every single update

[hoary owl]

at this point u might as well wait for the next day to play

[pseudo linden]

I heard raids were voiding items, probably best you wait a bit after any update

fruma bugs boutta to crazy

[tidal lark]

Just like league client aged spaghetti tends to have magic within

[tidal lark]

Get ready for the bingo sheet

[velvet iron]

I mean yeah wynn has had this going on for years. “Wynn spaghetti code” type stuff

[velvet iron]

I dont know what would be a solution as i am only a moderator and dont work on the game
However I have been around long enough for a few update cycles that sadly stuff just breaking every update is common, so for people who arent aware its more of a shock

[velvet iron]

Yeah i think wynns backend being a bit on the older side may or may not be contributing to part of it lol

[pseudo linden]

I got it ready

[verbal pebble]

you guys aren’t sacrificing to the machine spirit enough

whenever something goes wrong with my code i just put some candles around my laptop and say a prayer

usually works

[velvet iron]

we must feed it spaghetti to keep it satiated for a time, however we dig ourselves deeper into spaghettification of the code… such is the vicious cycle we are forced in

[ivory fern]

why isn’t anything tested on a normal server

[velvet iron]

Wdym "normal server"

[ivory fern]

and it doesn’t matter what code causes what problem the problem is that the problem isn’t even known about and problems shouldn’t be pushed if no explanation known

[ivory fern]

can push to one server at a time (unrealistic yes, but do smth similar like basic testing on one or two before pushing to all)

[velvet iron]

I am sure that it is actually this simple and easy of a fix that no one ever mentioned all this time
i mentioned earlier how maybe part of this is due to wynns backend being ancient so who knows what happens back there

[ivory fern]

bro what backend if ur changing smth with acolyte use one spell for 10 seconds

it doesn’t matter what causes it the issue is there

[velvet iron]

I am sure you haven't been around to know, but for example wynns backend for certain things is 1.8.9 or so. Ie builders really want to use new blocks for builds, but cannot due to the block pallet being limited by this. That is what I mean by outdated backend

[ivory fern]

yeah ik but this isn’t what i’m talking about

[velvet iron]

Again, it is never that simple for certain things. You can read my example above with the SE Church stairs, if it was this easy it would of already been fixed lol

[ivory fern]

idc what the restrictions are or the causes i’m talking about the reality they cause. the SE. church example yes hard to catch but no one would be mad abt that. some of thee are like caught in nanoseconds tho and aren’t unrelated to what is being changed

very frustrating

[amber hamlet]

enderclaw, are you aware if in the update process things are tested in a staging server

or if its just straight from the gm server to production

[velvet iron]

I understand it is frustating but if these "mystery" bugs coming to live were to be fixed the backend would prob need to be uprooted and redone to make it better or something but that sounds like a giant task for devs

[ivory fern]

then don’t release it

[velvet iron]

I do not know those details. The only reason I know enough about the hero beta to like live stuff is because I have been around several of those and seen it

[amber hamlet]

yeah this guy is ragebaiting

[velvet iron]

Well you are free to voice your concerns to them

[ivory fern]

?????? if there is a bug u don’t still release it because it could be potentially hard to change

do u know what ragebaht means?

[velvet iron]

I am only here to educate and give information about this kind of stuff since it is a bit shocking to newer players in Wynncraft

[velvet iron]

You are missing the point, half of these changes were (hopefully) not found in gm testing, and were a suprise when they were put into live

[amber hamlet]

so if you used your reading glasses, the entire thread is saying "yeah the bugs wasnt there in gm server but was in production", in this scenario, you dont know there is a bug before you release it

[velvet iron]

I don't think they were due to a lack of QA testing. There have been plenty of things that just break whenever a changelog happens that aren't related to the update at all

[ivory fern]

if u read what I said then you would know I had an answer

[amber hamlet]

what pushing to certain servers to start?

is that not what happens like every single update

[ivory fern]

HMMMMMMMMM

and if u read what ender is saying ender is not saying what u claim he is saying / he is not representing the rest of the thread and I am not talking to the rest of the thread

[amber hamlet]

oh my god he cannot read

[velvet iron]

Guys why are we aruging it isn't that deep :(

[ivory fern]

and testing on those servers

[amber hamlet]

the fix is in the staging server (or lackthereof)

[ivory fern]

this guy thinks he’s cool and I have to defend myself cause oeople tend to take the side of the aggressor that claims general failures

[amber hamlet]

must be rbing

[velvet iron]

This is what I agree with. Pushing a small amount of live servers for testing feels a bit complicated and prone to many variables that could screw things up. Especially if it is "new" content, or new items that could be brought to older worlds

I am not however qualified enough to know specifics but thats just my opinion based on my experience of going through many of these

[ivory fern]

I thought the problem was that the staging severs / testing severed (gm server? idk what that means but that’s what I assumed) were not representative of the actual server and too hard to make actually representative

[velvet iron]

That is not the issue as far as I am aware

[ivory fern]

^ maybe I can’t read?

[velvet iron]

Maybe some parts are, who knows, I don't. What I can make an educated guess on however is just how live tend to screw things up for seemingly no reason. I mention the Church SE stairs example, but there are plenty of smaller examples of just stuff that breaks that would be pretty obvious to gm people if it existed there

[ivory fern]

and this

[velvet iron]

Like heretic mask passive walk speed breaking on scrolling/classing

As something a bit recent

Oh also hummingbirds*

[ivory fern]

yeah so the solution is to do some testing on main (yes more running cost, but if u can’t fix the root, at least do this)

[amber hamlet]

from what i gather QA is just ran on the gm server which isnt the best play

[velvet iron]

The tricky part is besides 2x the testing, your sort of blindly wandering in the dark "hoping" to catch something that may be different from gm or whatever

I don't know if the qa would even want more work like that for only the possibility of more bugs being caught, let alone costs

[ivory fern]

yeah true and doing all testing on the main might be a little weird especially if things can get deleted and people cant spawn things in

[velvet iron]

Correct

[amber hamlet]

i mean to my eyes ct's would do what they think should work, send all the changes to a staging server, and let QA and others test out the changes and see if everything works properly

[velvet iron]

Mmh

[ivory fern]

CT should do basic testing before they send to QA tbh QA shoudl test the unexpected things

[amber hamlet]

i mean yea someting like testing if an ability works on gm would make sense

[ivory fern]

QA is quality assurance not quality maker checker quality quality everything

[amber hamlet]

but testing if raids give rewards correctly is something for QA

[hoary owl]

is tna berry room fixed

[spiral root]

[amber hamlet]

out in action,,,

[visual root]

hey iknow tha guy

[tender gate]

Testing raids is always kinda awkward because you need to grab people who are working on other stuff and tell them come do this for an unknown amount of time

To be clear we can force a raid but I’m sure there are bugs for being the only person in a raid

[queen fern]

But ya gotta at least do each raid once like how is it that an entire raid (or all 4) are broken on a patch

Like it happens pretty regularly that some gamebreaking bug appears that deletes items or makes the game unplayable

[amber hamlet]

yeah i dont expect the smaller raid bugs to be caught all the time (ex a bug that makes tna golem get stuck in certain conditions), but atleast all rooms that have been changed in any fashion should be ran atleast once