All Worlds Emergency Maintenance (Mar. 17)

[DragonCloud64]

* An issue where puppetmaster does too much damage will be addressed.

[Lokim]

But SE's testing and QA has always sucked, they don't even test the stuff they directly change properly. See: 2h weapon update.

It's impossible to say without actually knowing what SE was thinking but I'd bet the 2H buff wasn't a mistake. Blizzard recently gave a talk at GDC that referred to their techniques for adjusting and introducing jobs in a beta program. Though this isn't beta the same principles apply, and the logic goes like this:

1. Whatever job or action needs to be adjusted almost certainly has a stigma attached to it (it's gimpy, useless, etc).
2. A measured buff that brings this job or action in line with where it's supposed to be will almost certainly be dismissed by the community because of simple psychology.
3. You can't possibly test all of the edge cases as well as your 500K+ community can.

So, rather than doing one perfectly targeted buff or a series of smaller buffs spread out over time, you do one massive buff that's intentionally over-powered. This grabs the community's attention and causes tons of people to level that job or skill up that action and use the crap out of it. They'll quickly reveal exactly how over-powered it is, at which point you can slowly back the changes down to a more appropriate level.

Using this technique you get the community aware of the useful changes you've made to the job or action. When you back them down people will complain a little but they'll generally agree that the "nerf" was necessary, but they'll still see the usefulness of the buff. You end up at the exact same place you would have been with a series of smaller buffs except this way your players actually notice the changes.

This is almost exactly what they've done with scholar, too - rather than make a series of smaller changes they made one big fuck-all change and, if it turns out to be over-powered, they'll deal with that when it comes. But this way they've gotten the community's attention and that's what you really need.

[Tricen]

* An issue where puppetmasters were happy has been addressed.

[kenshyn]

MM is just broken 4k from a puppet is just insane could easily break the game in every way puppet master uses Vent. Thfs steals puppetmasters hate and flees its broken to no end period.....

[arshesney]

Either SE does not test updates of their servers need some runtime parameters to be passed them, there are many glitches that really shouldn't be there.

How can you explain the (now fixed, thank god) Blood Pact Recast gear not working anymore?
They added a smn ability, they HAD to test it with smn, yet they did not notice gear not working as supposed...

Get better testers! (or just not lazy ones)

Giving us players a chance to (willingly) beta-test would be nice, a fair shot at a non-broken AV even better!

ninja edit: booobooo 4 hrs maint. booobooo screwing us EU, fk SE, dammit, diaf, etc

[Popsiclestick]

ninja edit: booobooo 4 hrs maint. booobooo screwing us EU, fk SE, dammit, diaf, etc

You get a couple ground kings in your prime time in the coming days . NA is screwed pretty much for a week.

[Wizerd]

But SE's testing and QA has always sucked, they don't even test the stuff they directly change properly. See: 2h weapon update.

It's impossible to say without actually knowing what SE was thinking but I'd bet the 2H buff wasn't a mistake. Blizzard recently gave a talk at GDC that referred to their techniques for adjusting and introducing jobs in a beta program. Though this isn't beta the same principles apply, and the logic goes like this:

1. Whatever job or action needs to be adjusted almost certainly has a stigma attached to it (it's gimpy, useless, etc).
2. A measured buff that brings this job or action in line with where it's supposed to be will almost certainly be dismissed by the community because of simple psychology.
3. You can't possibly test all of the edge cases as well as your 500K+ community can.

So, rather than doing one perfectly targeted buff or a series of smaller buffs spread out over time, you do one massive buff that's intentionally over-powered. This grabs the community's attention and causes tons of people to level that job or skill up that action and use the crap out of it. They'll quickly reveal exactly how over-powered it is, at which point you can slowly back the changes down to a more appropriate level.

Using this technique you get the community aware of the useful changes you've made to the job or action. When you back them down people will complain a little but they'll generally agree that the "nerf" was necessary, but they'll still see the usefulness of the buff. You end up at the exact same place you would have been with a series of smaller buffs except this way your players actually notice the changes.

This is almost exactly what they've done with scholar, too - rather than make a series of smaller changes they made one big fuck-all change and, if it turns out to be over-powered, they'll deal with that when it comes. But this way they've gotten the community's attention and that's what you really need.

Brilliant, if you ask me.

[Mytiwar]

I think the best proof of their testers sucking ass or being non existent is when they implemented Raptor's mazurka.


Put Harlequin's horn in and VOILA permanent flee speed.

[arshesney]

ninja edit: booobooo 4 hrs maint. booobooo screwing us EU, fk SE, dammit, diaf, etc

You get a couple ground kings in your prime time in the coming days . NA is screwed pretty much for a week.

Too many faggots botting on Valefor, and not even worth wasting 3 hours in Aery/BD/VS old content is ooooold

[Ksandra]

WHY DOES SE ALWAYS PICK MY DAY OFF TO DO AN UPDATE?!










WHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHY ;_;

[Soprano]

I'm guessing those that complain about the testing, have never worked on any major software projects. Most often only the changes are tested, and the expected areas those changes will affect. Ie, they add a new spell and see how well it'll work and do what they want. But what they don't test is how all the different possible effects that could influence that spell, both the positive and the negative effects. Think how long it would take to test EVERY possible gear, attribute, buffs/debuffs, weather/day, and anything else that influences a spell or ability. Now that's something they can control. When you look at things like the Mog House bug, there could be a lot that's influencing than we could know about. Especially if not everybody is experiencing it.

I know this from experience. I worked as a maintenance programmer for 5 years. Can't tell you the number of times I made a change in one area of our primary software, and I ended up breaking something in what I thought was an unrelated area. It got past QA b/c the changes were tested, but not all the effected systems b/c we didn't know that they were affected. But when it hit the offices, we quickly found them b/c we had 100+ offices using it. And yes, we had several 1 or 2 day releases, even a 1hr release.

edit: fixed english, I hope.

So what you are saying is, your company still does not understand what change management is? Or parallel testing or full regression testing? Seems like all you do is unit testing.

[Lordender]

I don't know how they fucked up the 2H adjustment when it was first added. I don't understand why they couldn't of taken 6 people of a merit set up, slap god mode on and brought them to their favorite merit camp and let it rip before they released that shit to the public.

I understand that no one releases perfectly coded stuff on their first try but there is a difference between slightly bugged, bugged and what the fuck bugged which seems like SE releases shit on the "what the fuck bugged" scale way too often.

The way 2H update SHOULD of been handled was like this:

God mode 6 people take them to a merit camp, pimp them all out without relic weapons and have three 2H melee, drk,drg,sam then a dualwield war + brd+rdm.

Run numbers, parse to see how far ahead the dual wielder was before changing any formula then plug in some more DD's like a mnk and a GAXE war and see how the mnk does then take all that data and begin to adjust the formula accordingly.

[evilpaul]

So the servers are back up at 4PM east coast time?

Also, the open testing server idea sounds interesting, but it would kind of fly in the face of SE's apparent policy of keeping people from cashing in on update changes by knowing about them beforehand.

[SephYuyX]

Yes, and an NDA would work.

[Kyreth]

Yes, and an NDA would work.

It doesn't, usually. All it'd take is a copy of the update and the usual suspects to .dat read the thing. Anonymous and highly effective.

List of new stuff/quest text before you could say "Check files." for the actual update, and it's oh, so exploitable. How many people would have dodged buying an adman ingot if they knew it'd be a quest as of the update? Or a Merzaq knowing there'd be an upgrade (possibly knowing it'd be cheaper)?

[Osede]

Inadvertently breaking some seemingly unrelated code does happen, and is excusable. But SE's testing and QA has always sucked, they don't even test the stuff they directly change properly. See: 2h weapon update.

SE >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> bliz.

nothing perfect. Go play wow and you will see that SE is way too better when coming into fixing stuff and patching stuff.

[Toonshady]

Inadvertently breaking some seemingly unrelated code does happen, and is excusable. But SE's testing and QA has always sucked, they don't even test the stuff they directly change properly. See: 2h weapon update.

SE >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> bliz.

nothing perfect. Go play wow and you will see that SE is way too better when coming into fixing stuff and patching stuff.

i played both for a while and it's about the same. at least Blizzard listened to their customers.. and had torrents for patches instead of the clusterfuck most FFXI updates are for the first day.. and had beta tests for new patches..
oh fuck sorry for turning into another SE vs Bliz comparison -.-

[Paegan]

Inadvertently breaking some seemingly unrelated code does happen, and is excusable. But SE's testing and QA has always sucked, they don't even test the stuff they directly change properly. See: 2h weapon update.

Do you have have any idea the scale of programming that goes into a game of this scale, and how interdependent the programming is? As for the 2h weapon update, the original problems with it were fixed, the adjustment that came later was a necessary one, but not as predictable as you might think.

Yes, I do. The same scale that goes into other MMORPGs whose developers don't constantly fuck up and release updates full of major bugs and untested adjustments.

It's impossible to say without actually knowing what SE was thinking but I'd bet the 2H buff wasn't a mistake. Blizzard recently gave a talk at GDC that referred to their techniques for adjusting and introducing jobs in a beta program. Though this isn't beta the same principles apply, and the logic goes like this:

1. Whatever job or action needs to be adjusted almost certainly has a stigma attached to it (it's gimpy, useless, etc).
2. A measured buff that brings this job or action in line with where it's supposed to be will almost certainly be dismissed by the community because of simple psychology.
3. You can't possibly test all of the edge cases as well as your 500K+ community can.

So, rather than doing one perfectly targeted buff or a series of smaller buffs spread out over time, you do one massive buff that's intentionally over-powered. This grabs the community's attention and causes tons of people to level that job or skill up that action and use the crap out of it. They'll quickly reveal exactly how over-powered it is, at which point you can slowly back the changes down to a more appropriate level.

Using this technique you get the community aware of the useful changes you've made to the job or action. When you back them down people will complain a little but they'll generally agree that the "nerf" was necessary, but they'll still see the usefulness of the buff. You end up at the exact same place you would have been with a series of smaller buffs except this way your players actually notice the changes.

This is almost exactly what they've done with scholar, too - rather than make a series of smaller changes they made one big fuck-all change and, if it turns out to be over-powered, they'll deal with that when it comes. But this way they've gotten the community's attention and that's what you really need.

This isn't a beta, the same principles do not apply. A beta is where you do your testing, not the release version.

Yes, I could understand if it was some obscure edge case, but it wasn't. It took one SAM spamming 1-1.5k WS's on the first faf to pop after that update for this forum to realise exactly how broken that was, and how unbalanced 2h vs 1h now was. If they intended that then they really don't have a clue how their own game works.

SE >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> bliz.

nothing perfect. Go play wow and you will see that SE is way too better when coming into fixing stuff and patching stuff.

I do play WoW, do you? Blizzard have plenty of their own faults, but anyone who thinks SE's update/patching/testing process is better is just another fanboy for whom SE can do no wrong.

[Karbuncle]

I wonder if they're ever going to fix the fact Moo Ouzi the Swiftblade shows up in Campaign Battles as "Moo Ouzi the Swiftblad"
The chatlog says Swiftblade, but above his head is Swiftblad. i'll post pics when i upload them o-o

[Katlan]

I think the best proof of their testers sucking ass or being non existent is when they implemented Raptor's mazurka.


Put Harlequin's horn in and VOILA permanent flee speed.

Way to make me sad again. Dev's shoulda let that one slide and gave bard's some fun... And aside that.. im curious what they r gonna change w/this update. in before the campaign QQ