Does gw2 need a PBE? — Guild Wars 2 Forums

Does gw2 need a PBE?

PBE= Public Beta Environment

This would mean that Arenanet can Let players test out balance changes and pull off data from the PBE to determine the impact of the changes.

Think about a PBE in PvP and an instanced raid for example to differentiate between pvp and pve

Does gw2 need a PBE? 142 votes

Yes
51%
ReaverKane.7598Matt H.6142Neural.1824ButterPeanut.9746BinaerHamsteraer.7396Vlad Morbius.1759Khailyn.6248Karmapolice.4193DietPepsi.4371Ashantara.8731zealex.9410Alatar.7364Spurnshadow.3678Asum.4960AncientYs.8613Nimrod.9240Carighan.6758Ardid.7203Game of Bones.8975circuitnerd.5863 73 votes
No
48%
WhatLiesBeneath.9018Shayne Hawke.9160Jahroots.6791Cynn.1659TheGrimm.5624Rauderi.8706IndigoSundown.5419Nate Anabe.6091derd.6413mosser.7695DeanBB.4268Inculpatus cedo.9234Ayrilana.1396Gulesave.5073Mea.5491Hevoskuuri.3891lukey.8951Glox.5942CETheLucid.3964Konig Des Todes.2086 69 votes
<1

Comments

  • zenleto.6179zenleto.6179 Member ✭✭✭

    Maybe. It could be something.

  • Arzurag.7506Arzurag.7506 Member ✭✭✭

    Absolutely, they need to test stuff before they bring it on the live-servers.
    Though, I guess it´s not worth the effort due them releasing a balance patch once in a blue moon.

    "I´m not big on sermons, Broken bones teach better lessons."

  • Ashantara.8731Ashantara.8731 Member ✭✭✭✭
    Yes

    @zenleto.6179 said:
    Maybe. It could be something.

    It would definitely be better than having skill updates being shoved down our throats without them going through some practical testing beforehand.

  • zenleto.6179zenleto.6179 Member ✭✭✭

    @Ashantara.8731 said:

    @zenleto.6179 said:
    Maybe. It could be something.

    It would definitely be better than having skill updates being shoved down our throats without them going through some practical testing beforehand.

    Probably yeah.

  • Malafaia.8903Malafaia.8903 Member ✭✭✭✭
    Yes

    I never play on test environments, but i know there's people willing to do it, so... WHY NOT?

  • Hoodie.1045Hoodie.1045 Member ✭✭✭✭
    Yes

    ArenaNet would benefit a lot if they allowed the community to help them with the balance patches and giving them feedback, both for PvE and PvP.

    Karras The Engineer

  • Imperadordf.2687Imperadordf.2687 Member ✭✭✭✭

    Yes, if it's not gonna be "anyone can enter". Needs to have criterias, needs to be elitist.

  • Dami.5046Dami.5046 Member ✭✭✭✭

    No because people will just want it all nerfed, no point.

  • The V.8759The V.8759 Member ✭✭
    Yes

    @Mea.5491 said:

    @Etheri.5406 said:
    The community is 95% clueless about balance;

    This. It would be like "I main this profession so it needs buffs" lol.

    Well thats not true. Playing on a PBE doesnt mean that your opinion matters roughly said. ArenaNet can still take a lot of data from the server to see what the effects of changes are

  • Etheri.5406Etheri.5406 Member ✭✭✭✭

    @The V.8759 said:

    @Mea.5491 said:

    @Etheri.5406 said:
    The community is 95% clueless about balance;

    This. It would be like "I main this profession so it needs buffs" lol.

    Well thats not true. Playing on a PBE doesnt mean that your opinion matters roughly said. ArenaNet can still take a lot of data from the server to see what the effects of changes are

    So we get more spam with the vast majority of players claiming stuff they don't grasp. I'm sorry if I want wacko balance ideas i'll just read these forums.

    Anet needs to ignore it's casual community and start talking / listening to high-end guilds for the respective gamemodes. I have more faith in the couple hundred top end pvp players than in the thousands of silver and bronze players with strong opinions. Same for PvE. Same for WvW.

    The loud mediocrity just demands to feel relevant; but in truth they just shoot themselves in the feet time after time.

  • Game of Bones.8975Game of Bones.8975 Member ✭✭✭✭
    Yes

    Data needs to be collected from the widest possible swath of players.
    Players shouldn't give opinions of new skills only play the skills as presented. There should be random instances where solo or PUGs take on a group as the devs design.
    The devs can then determine how the average player uses the skill, how often it is used, and if it needs to be modified.
    On the downside, I think the majority of the people who would be interested in how skills are tweaked are the "elite" players who modify builds for the best possible outcome. The "average player" follows what the elite players have said what works for them.

    "That's what" -- She

  • TheQuickFox.3826TheQuickFox.3826 Member ✭✭✭✭

    I don't care too much. I would not 'waste' my time on beta testing because I need all my play time to achieve my goals on the live server.
    On the other hand: if many player would like to preview stuff or beta test, why not?

    In the end, it will be relevant how much time and resources such a thing would require off the company, and if this is worth delaying other things for that have also be requested by the community members.

    Ascalon Will Prevail!

    GW Wiki user page | GW2 Wiki user page

  • Airdive.2613Airdive.2613 Member ✭✭✭
    No

    @Game of Bones.8975 said:

    @Etheri.5406 said:
    I have more faith in the couple hundred top end pvp players than in the thousands of silver and bronze players with strong opinions. Same for PvE. Same for WvW.

    If 90% of your customers are "silver and bronze players" those are the people you need to get your information from. On a bell-curve, the high-end players are the outliers.

    You don't necessarily need to get any information at all.
    It might be valuable, or it might undermine the developer's creative inspiration. Ultimately ANet has every reason to develop their game to the best of their abilities, not their playerbase's. They are the content creators, not consumers.

  • Danikat.8537Danikat.8537 Member ✭✭✭✭

    I'm not sure which answer I'd pick but I'm leaning towards no. Just having a public test server isn't a guarantee that a game will be bug free or that most players will like the balance changes, there's a lot more that needs to go into it and even with a really good process it's not fool-proof. Plenty of games have test servers and they have exactly the same problems as GW2, sometimes even worse.

    The biggest problem with test servers I've seen is that a lot of people who use them couldn't care less about testing. They use it to get a preview of the new content ahead of time for their own interest, or to practice so they have a headstart on new builds and completing new content (e.g. spend 2 weeks practicing on the PTS, claim "worlds first" raid clear on the first day after release) or just to mess around with tools provided to facilitate testing like free instant level-ups or infinite free currency. I'm as guilty of this as anyone else - I used to love playing on Ultima Online's test servers because I could boost my character to ridiculous levels and get infinite gold (in my defense I was a young teen, and I'd been told they were for the developers to test the server and they just needed lots of people on it, it was only later on I learned that wasn't the case). I've also used the Elder Scrolls Online test servers to try out new builds or cosmetic changes to characters. But none of that helps the developers.

    On a related note I absolutely disagree that what you want is all "elitist" players testing. If you only get the best players (or the ones who consider themselves to be the best) giving feedback on changes it will almost certainly lead to an extreme version of what we saw with HoT - a vocal minority crying out for harder content because everything in the game was too easy then a lot of people crying for it to be changed after release because what they actually wanted was better rewards (which they associate with content advertised as being harder, because of other games) which they could farm easily and other people crying for it to be changed because they're not that good and find themselves incapable of completing the new content.

    But it also needs support from Anet to monitor and collate feedback and pass it on to the right teams, and time for those teams to review it and work out how to act on it. They do have an entire QA department so I assume they do that already but with more testers there's more feedback so it takes more time.

    Finally there's the timing problem. An update needs to be nearly finished before players can test it, ok maybe the voice acting isn't done and some animations and things aren't in, but it all needs to be functional and basically finished. Which means they're going to be close to the intended release date and what they have time to fix will be very limited. Big changes like re-writing the story are probably out of the question. So you end up with the company "ignoring" lots of feedback and even bug reports because there simply isn't time to fix it before release and lots of negative feelings from people who feel like their time and effort is going to waste because the feedback is never used.

    I think a PTS can work, but it needs to be done right and that's easier said than done and needs a lot of time and resources from the company and buy-in from everyone at the company to be flexible enough to make the process work.

    "Hard knocks, bad luck, been knocked down,
    You got back up, rise up, shine on, keep on fighting, the war is almost done...But then I hear you're gone.
    I feel, when the lights go down, you are still here, all you hold dear remains.
    Your star never fades."

  • Erik.6951Erik.6951 Member ✭✭
    No

    No. It would further divide the community, and people would still get just as frustrated when things don't work on the PBE anyway. "I wanna do this event, ANET should have made it work before releasing it" even though it's not "released", etc.

  • Game of Bones.8975Game of Bones.8975 Member ✭✭✭✭
    Yes

    @Etheri.5406 said:

    @Game of Bones.8975 said:

    @Etheri.5406 said:
    I have more faith in the couple hundred top end pvp players than in the thousands of silver and bronze players with strong opinions. Same for PvE. Same for WvW.

    If 90% of your customers are "silver and bronze players" those are the people you need to get your information from. On a bell-curve, the high-end players are the outliers.

    Yes; and 90% of your customers don't have balance issues. They have L2P issues. Anet doesn't need to change the game to make it more balanced for them.

    You need your high-end scene because of what is essentially trickle-down economics in video games. Your high end players determine the meta; rely on balance, provide the majority of external content, competition, goals for others to aspire towards, ...

    You cannot in any way balance for the "90% silver and bronze players". You cannot in any way design the game around these players either. At best; you can try and have healthy learning curves for most champions and the game overall. Look at ANY competitive game or game with a healthy pvp scene. Not a single one balances for its majority; all of them balance for the high-end scene.
    Only insane outliers (low skill high reward type of stuff) gets reworked because it's difficult to balance for both high and low end. And if they are in fact too difficult to balance for both; they're often just left as setup while banished from high-end meta. Which is ok.

    You can keep going "90% of the players is more of the players!!!" but that doesn't mean they have any idea what they're on about. 90% of the players have no idea what they're doing; yet you want to put them in charge of balancing :trollface:

    I agree with most of what you are saying. People look to the upper players for advice, I know I do.
    That is one of the reasons why I think individual instances would be the most helpful. Run things past the best players first, make mods to skills, then allow the rest to experience the already modded skills in their own instance.
    I didn't mean to say that you should rely solely on the "average, low-information" player, I didn't word my post as well as I could have.

    "That's what" -- She

  • Etheri.5406Etheri.5406 Member ✭✭✭✭

    @Game of Bones.8975 said:

    @Etheri.5406 said:

    @Game of Bones.8975 said:

    @Etheri.5406 said:
    I have more faith in the couple hundred top end pvp players than in the thousands of silver and bronze players with strong opinions. Same for PvE. Same for WvW.

    If 90% of your customers are "silver and bronze players" those are the people you need to get your information from. On a bell-curve, the high-end players are the outliers.

    Yes; and 90% of your customers don't have balance issues. They have L2P issues. Anet doesn't need to change the game to make it more balanced for them.

    You need your high-end scene because of what is essentially trickle-down economics in video games. Your high end players determine the meta; rely on balance, provide the majority of external content, competition, goals for others to aspire towards, ...

    You cannot in any way balance for the "90% silver and bronze players". You cannot in any way design the game around these players either. At best; you can try and have healthy learning curves for most champions and the game overall. Look at ANY competitive game or game with a healthy pvp scene. Not a single one balances for its majority; all of them balance for the high-end scene.
    Only insane outliers (low skill high reward type of stuff) gets reworked because it's difficult to balance for both high and low end. And if they are in fact too difficult to balance for both; they're often just left as setup while banished from high-end meta. Which is ok.

    You can keep going "90% of the players is more of the players!!!" but that doesn't mean they have any idea what they're on about. 90% of the players have no idea what they're doing; yet you want to put them in charge of balancing :trollface:

    I agree with most of what you are saying. People look to the upper players for advice, I know I do.
    That is one of the reasons why I think individual instances would be the most helpful. Run things past the best players first, make mods to skills, then allow the rest to experience the already modded skills in their own instance.
    I didn't mean to say that you should rely solely on the "average, low-information" player, I didn't word my post as well as I could have.

    I know you didn't say you only listen to the majority. The issue is for a game as GW2; you cannot listen to the majority in general.

    In games like league, dota, even overwatch you still have the trickling down in sufficient amounts. The high end players are a very small fraction of the votes; but they influence a large fraction of the votes through various media. Reddit, twitch, discussions, guilds / clans / communities / ...

    All of this? Doesn't exist anymore for GW2. Casuals don't watch streams or read discussions on twich. Many don't even consider top tier players as better, just as "toxic elitists". And clearly; they know balance better, because in silver DH is really really OP. The popular feedback you get from the GW2 community is usually... not good feedback. That is the truth.

    Side thought : The argument of XXX are the majority so they are important is genuinely a very, very dangerous and malicious argument. It was made when WvW was neglected; but you just kicked out a major aspect of the game. Same for PvP; another aspect of the game gone. Now it's being made for raids and difficult fractals, after all casual players are the majority.

    Something will /always/ be the majority; but if you ignore the rest then soon your majority will still be a whole lot smaller. And casuals, even if they don't realise it, do benefit a lot from having a large and healthy veteran population. You know, for guilds and making groups and content and all that stuff. WvW without veterans will be ... not even close to the same. PvP without veterans will be a major clownfiesta with nearly no "close" games. PvE with insufficient veterans compared to casuals still leads to elitism because you can't rely on the average player to carry their weight.

    And the smaller the veteran population gets, the more elitist they will be. After all; it's the only way they can continue their style of play. There are "QQ stop elitism" posts in both pve and wvw subfora; but it won't stop nor get better. It's mostly because... well, we do it to ourselves :)

    So at the end of the day... majority groups very frequently told minority groups that they aren't important or relevant; but in due time they simply made the game worse and less inviting for themselves. And they further harmed the development of these modes, too. Which is something that is most likely not reversible and the direct result of our toxic yet friendly community.

  • PyrateSilly.4710PyrateSilly.4710 Member ✭✭✭
    edited July 24, 2018

    No
    There are things/bugs that do not turn up until you have it on the live server, just the nature of the beast.
    Most will just see what is coming up and stock pile materials just to make more gold. They will only try out new things just to be the first do them when it goes live. Various things like these is what has happened on so many other beta servers and that's why most just don't do them anymore.

  • Sinful.2165Sinful.2165 Member ✭✭✭
    Yes

    I’ve always thought they’d be well served by a public test server.

    So many changes go live that players spot and Anet fixes, currently the test environment is the live server. It would only benefit the game to put a buffer between anets rather low quality internal testing and the live servers.

    Consider the deadeye change for example where they removed stealth from kneel and put it at the beginning of dodge. The player should shoot someone and dodge to stealth only to be immediately revealed when their projectile hit the target. LOLWUT? That would have been caught by any half decent player in 10 minutes of testing but it went live?? Do they even have an internal test team?

    So yes, absolutely. Anet, let some of your enthusiastic player base do some free testing for you.

  • Revilrad.1962Revilrad.1962 Member ✭✭
    edited July 24, 2018
    No

    No because people who participate would start wanting "special" , "prestigious" rewards for their effort and the whole thing will become just another recurring Celebration Event.

  • TheGrimm.5624TheGrimm.5624 Member ✭✭✭✭
    No

    One thing to consider here is the value of dedicated testers versus on/off again testers and whether or not people play all the classes. Quite often players will be biased since a change being made will be on the class they "main" if they don't play all the classes and therefore the feedback you might receive might be biased resulting in errors in testing. On/off again testers may also use different metrics and not be consistent again resulting in skewed results.

    Envy the Madman his musing when Death comes to make fools of us all.
    De Mortuis Nil Nisi Bonum.
    TheGrimm PoTBS/GW1/WAR/Rift/GW2/MWO/ESO/WoT/WoW/D2/HoTS/CU/AoC

  • Yes

    Yes yes yes you have three yes's from me! Make guildwars2 great again!

  • Dreadshow.9320Dreadshow.9320 Member ✭✭✭
    No

    So that players testing things can go to Reddit or the official forum and complain that the bug or issue they reported several times still made it too live? Over and over again ? Nah we get those already no need for more.

  • CETheLucid.3964CETheLucid.3964 Member ✭✭✭
    No

    @PyrateSilly.4710 said:
    No
    There are things/bugs that do not turn up until you have it on the live server, just the nature of the beast.
    Most will just see what is coming up and stock pile materials just to make more gold. They will only try out new things just to be the first do them when it goes live. Various things like these is what has happened on so many other beta servers and that's why most just don't do them anymore.

    That's the sum of it. Way too much for way too little. It wouldn't accomplish anything and generally be a waste of resources.

  • FrizzFreston.5290FrizzFreston.5290 Member ✭✭✭✭
    No

    Considering the nature of such things I think a closed beta experience would be better.

  • Inculpatus cedo.9234Inculpatus cedo.9234 Member ✭✭✭✭
    No

    ArenaNet already has test servers; they just aren't as public as some may desire. You can sign up to be a volunteer test player if you should so desire.

    Good luck.

  • The V.8759The V.8759 Member ✭✭
    Yes

    @Revilrad.1962 said:
    No because people who participate would start wanting "special" , "prestigious" rewards for their effort and the whole thing will become just another recurring Celebration Event.

    Why would they do that? theres plenty of games with PBE's, havent seen a single one giving rewards. If a players wants to test in pbe, its their choice. You know there is no reward

  • Westenev.5289Westenev.5289 Member ✭✭✭✭

    I feel it's a non-issue, since few people here has incentive to give unbias opinions on balance. Personally, I'd prefer more frequent hit-and-miss balance patches.

  • ReaverKane.7598ReaverKane.7598 Member ✭✭✭✭
    Yes

    @Airdive.2613 said:

    @Game of Bones.8975 said:

    @Etheri.5406 said:
    I have more faith in the couple hundred top end pvp players than in the thousands of silver and bronze players with strong opinions. Same for PvE. Same for WvW.

    If 90% of your customers are "silver and bronze players" those are the people you need to get your information from. On a bell-curve, the high-end players are the outliers.

    You don't necessarily need to get any information at all.
    It might be valuable, or it might undermine the developer's creative inspiration. Ultimately ANet has every reason to develop their game to the best of their abilities, not their playerbase's. They are the content creators, not consumers.

    Have you been playing this game?
    Seriously? Have you? because it feels you haven't...
    They seriously need a PBE, there hasn't been an instance of balance where they have got it right and that didn't have unforseen consequences when it hits live.

    Most people i've read voting no, don't seem to grasp the concept of how a PBE works, and what it's for.

    First of all... They already have something like a PBE. Except it's restricted to content creator partners. That's why there's been issues with players leaking recipes and dealign with insider trading. They were content creators that had access to that early build servers and exploited the information. That's been dealt with though. Just serves as proof it exists.

    So the groundwork is done already.

    Second, like currently, access to such servers is usually highly vetted, not just anyone gets access, so usually the feedback will be more on point, it's not just a early version of every balance patch.

    Third, having 1000 players play-testing a build is probably already tens of times more people than their internal testing teams can provide. And while they can simulate concurrent players for some errors, it's impossible to simulate player choice and it's impossible to emulate the "hive-mind" of millions of players pushing a new meta after a release. Even 1000 players in a PBE would be a spec of dust, but already a great improvement. And even without any written feedback from those players just the data stream from a PBE environment alone would be enough to show how effective and ineffective changes are.

    Fourth, GW2 has a catastrophically slow balance pace, we can't really afford to spend 3 months with broken metas after each balance patch (which tends to happen frequently). If their pace was monthly or bi-weekly, then i'd say it would be frequent enough that balance would be restored after a failed attempt. But with such a slow pace, they really need that pre-balance-release feedback so that live servers don't have to be ruined by their own failures constantly.

    Seriously, just look at the market out there. Most mobas and similar games have a PBE, have more frequent balance updates, and still they have instances of broken metas. If they want to keep the quarterly pace, Arena Net should really stack the deck in their favour and get a PBE environment to make sure their planned changes don't fizzle or blow back in our faces has is common.

  • Airdive.2613Airdive.2613 Member ✭✭✭
    edited July 24, 2018
    No

    @ReaverKane.7598 said:

    Seriously, just look at the market out there. Most mobas and similar games have a PBE, have more frequent balance updates, and still they have instances of broken metas.

    Say what? So they spend tons of resources on PBEs and there are still broken metas. Why keep spending them, then, except for PR?
    Also, balancing GW 2 professions is inherently much harder than, say, League champions because those have about 4 abilities each with the same list of passives to choose from, whereas the number of theoretically possible combinations in GW 2 is mind-boggling. It is best they don't bother too much and just roll out stuff for us players to explore and enjoy.

  • ReaverKane.7598ReaverKane.7598 Member ✭✭✭✭
    edited July 24, 2018
    Yes

    @Airdive.2613 said:

    @ReaverKane.7598 said:

    Seriously, just look at the market out there. Most mobas and similar games have a PBE, have more frequent balance updates, and still they have instances of broken metas.

    Say what? So they spend tons of resources on PBEs and there are still broken metas. Why keep spending them, then, except for PR?
    Also, balancing GW 2 professions is inherently much harder than, say, League champions because those have about 4 abilities each with the same list of passives to choose from, whereas the number of theoretically possible combinations in GW 2 is mind-boggling. It is best they don't bother too much and just roll out stuff for us players to explore and enjoy.

    Yeah, no... They mess up sometimes, but guess what? They mess up a lot less consistently in those games than in GW2. Also they fix them more readily than GW2.
    Is it mind boggling? Is it greater than League of Legends for example? Because GW2 balances classes mostly in a vaccum. LoL balances their interaction, which is a different league ahead of what we get.

    Yeah, 4 abilities, and a magnitude more synergies and item combinations than GW2 has. Just look at the latest changes, they're having to balance the game because not of a champion being too strong but because people were palying it a bit differently than usual (the funnel strat) which made champions grow way faster than intended (because extra money) and it was breaking the game. So they had to fix that.

    @Illconceived Was Na.9781 said:
    It turns out to be hugely expensive to maintain a public test environment, without generating much useful intel.

    They already have a server that is used for similar purposes. Just not as widespread.

    Too many people who play beta do so to learn and prepare for the upcoming content, including theorycrafting, exploring new areas to figure to efficient farming techniques, speculating on where they might make a killing on the trading post (or its equivalent). And all too few people who report issues do so in a way that makes it easy to action them, e.g. "this didn't work" instead of "the skill say X, but when I used it, I got Y." There's also a lot of confusion between, "I don't like this" and "it's broken."

    They can do what they already do in live servers, collect data and dismiss most of player feedback. Or they can have only vetted players join (which is what they do already, just need to expand it a bit) and use only their specialized feedback.

    Worse, running a test environment means more work to keep surprises hidden, to manage economic changes, to avoid spoilers for those who don't "test," and to prevent people from burning out even before the update. And finally, it adds to the behind-the-scenes costs, because it requires keeping track of one more 'build' of the code & keeping track of differences. Besides what we see 'live,' there's also going to be the 'next to go live', the 'testing now', and 'developing now' versions of the code (among others). It's part & parcel of software development to have robust tools to keep track of each and 'promote' only the approved changes, but each additional 'official' environment adds to the complexity.

    They already allow Content creators early access to content. And it doesn't get spread out. There was a lot of issues a few years back about people using the insider knowledge obtained from such servers to profit, by stockpiling mats that would be used in the "to be released" recipes and then selling them after their price inflated. This was already handled.

    And to top it off, even in environments that include public betas, plenty of bugs go unmentioned.

    Yes, lets not get a vaccine for a disease because some people can't be vaccinated. Lets not disinfect water because some people are immune to chlorine.
    Even if a few bugs slip through, less will.

    Putting it all together: it's a lot more effort without it preventing a lot more bugs.

    The majority of the effort was already undertaken.

    I share the frustration of the OP that this game sometimes gets hit by what seems like easily-preventable issues (and I'm sure, in some cases, ANet would agree: there's no good reason for those). All the same, while it sounds like a good idea in theory, in practice... not so much. Which is why ANet has actively considered it (more than once) and concluded (more than once): it's not worth it.

    I just love how you're so wrong with such confidence.

  • Ephemiel.5694Ephemiel.5694 Member ✭✭✭
    Yes

    Even if they had it, they would pull a Blizzard and constantly say "we're listening to your feedback" while ignoring it completely.

  • @ReaverKane.7598 said:

    @Illconceived Was Na.9781 said:
    It turns out to be hugely expensive to maintain a public test environment, without generating much useful intel.

    They already have a server that is used for similar purposes. Just not as widespread.

    No, those aren't interchangeable tools. Setting up a PBE is nothing like the work to set up a private test server for specific purposes, with people under Non-Disclosure, with people specifically chosen because of their relationship to the game, the type of feedback they can offer. Under those conditions, they can use the same test environment that already exists. A public server has to go through a different amount of vetting, because the average person expects more stuff to work more often (just look at the sorts of reports we got from the official BETA preview for GW2's launch, for HoT).

    With a private test environment, they can boot people who don't provide valuable feedback; with a test server, they can't tell which feedback is useful or not until they read it. That's a lot of reading, without any clear indication of whether the content will provide any actionable intel. Much like the video linked here: we're expected to watch 20 minutes without any clue as to whether there's anything interesting.

    "Face the facts. Then act on them. It's ...the only doctrine I have to offer you, & it's harder than you'd think, because I swear humans seem hardwired to do anything but. Face the facts. Don't pray, don't wish, ...FACE THE FACTS. THEN act." — Quellcrist Falconer

  • Zaklex.6308Zaklex.6308 Member ✭✭✭✭
    No

    I said No, they've had an Alpha team in the past, and for all I know they still might have one...there's plenty of testing, it's just that you can't test for everything a player base might do. When you have a million players you have the potential for a million different things to be done...and that is impossible to test for.

    Yes...no...maybe...what do you want, can't you see I'm busy saving the world...AGAIN!

  • Danikat.8537Danikat.8537 Member ✭✭✭✭

    @ReaverKane.7598 said:
    Second, like currently, access to such servers is usually highly vetted, not just anyone gets access, so usually the feedback will be more on point, it's not just a early version of every balance patch.

    That's not a public test server then. That's the system they already have where periodically groups of play testers are invited to try unreleased content under a non-disclosure agreement. I can't find the link now but there is/was a thing on the Anet website (www.arena.net, not the GW2 one) where you could sign up to be considered for future tests.

    But if it's not publicly available it's not a public test environment. It's basically inviting players (or sometimes non-players) to be volunteers in the QA department.

    "Hard knocks, bad luck, been knocked down,
    You got back up, rise up, shine on, keep on fighting, the war is almost done...But then I hear you're gone.
    I feel, when the lights go down, you are still here, all you hold dear remains.
    Your star never fades."

  • ReaverKane.7598ReaverKane.7598 Member ✭✭✭✭
    Yes

    @Danikat.8537 said:

    @ReaverKane.7598 said:
    Second, like currently, access to such servers is usually highly vetted, not just anyone gets access, so usually the feedback will be more on point, it's not just a early version of every balance patch.

    That's not a public test server then. That's the system they already have where periodically groups of play testers are invited to try unreleased content under a non-disclosure agreement. I can't find the link now but there is/was a thing on the Anet website (www.arena.net, not the GW2 one) where you could sign up to be considered for future tests.

    But if it's not publicly available it's not a public test environment. It's basically inviting players (or sometimes non-players) to be volunteers in the QA department.

    It can be publicly available, but they can still veto who accesses it. League of Legends places restrictions on who accesses it.
    GW2 can do the same. They can for example launch a PvP test build, and only open the server to players with 80+ rank on PvP and x amounts of matches, or plat+.

  • @ReaverKane.7598 said:
    It can be publicly available, but they can still veto who accesses it. League of Legends places restrictions on who accesses it.
    GW2 can do the same. They can for example launch a PvP test build, and only open the server to players with 80+ rank on PvP and x amounts of matches, or plat+.

    That still comes with all the issues of a public server, and none of the benefits of a private one. People with Plat+ or 80+ rank don't automatically become good testers, good at giving feedback, or even able to offer criticism that applies generally rather than their specific point of view. We have many such examples already of skilled players with myopic views about the game.

    "Face the facts. Then act on them. It's ...the only doctrine I have to offer you, & it's harder than you'd think, because I swear humans seem hardwired to do anything but. Face the facts. Don't pray, don't wish, ...FACE THE FACTS. THEN act." — Quellcrist Falconer

  • ReaverKane.7598ReaverKane.7598 Member ✭✭✭✭
    Yes

    @Illconceived Was Na.9781 said:

    @ReaverKane.7598 said:
    It can be publicly available, but they can still veto who accesses it. League of Legends places restrictions on who accesses it.
    GW2 can do the same. They can for example launch a PvP test build, and only open the server to players with 80+ rank on PvP and x amounts of matches, or plat+.

    That still comes with all the issues of a public server, and none of the benefits of a private one. People with Plat+ or 80+ rank don't automatically become good testers, good at giving feedback, or even able to offer criticism that applies generally rather than their specific point of view. We have many such examples already of skilled players with myopic views about the game.

    Again, player feedback is as useful as they find it. The true virtue of such a server is data (not player feedback, but server logs and statistics) that approximates closer to how it will behave on live, as opposed to how it behaves when being tested exclusively by the QA team.

  • @ReaverKane.7598 said:

    @Illconceived Was Na.9781 said:

    @ReaverKane.7598 said:
    It can be publicly available, but they can still veto who accesses it. League of Legends places restrictions on who accesses it.
    GW2 can do the same. They can for example launch a PvP test build, and only open the server to players with 80+ rank on PvP and x amounts of matches, or plat+.

    That still comes with all the issues of a public server, and none of the benefits of a private one. People with Plat+ or 80+ rank don't automatically become good testers, good at giving feedback, or even able to offer criticism that applies generally rather than their specific point of view. We have many such examples already of skilled players with myopic views about the game.

    Again, player feedback is as useful as they find it. The true virtue of such a server is data (not player feedback, but server logs and statistics) that approximates closer to how it will behave on live, as opposed to how it behaves when being tested exclusively by the QA team.

    And again, you're only looking at it from a player perspective. Yes, absolutely, it's valuable to find out that a particular skill or chat code only bugs out on a full map or that the stress-testing scripts missed a key factor of 100 humans playing together (or 10000 logged on at the same time). But it's an enormous amount of work to make it possible for the occasional situation in which that is the aggravating factor.

    The point is that the amount of useful data is limited, and it's not favorable to the costs. There are set up costs, maintenance costs, and change & code management headaches. The hardware is the easy part; it's all the human resources that go into it that make it a bad deal overall.

    Put another way, public test servers do prevent some issues, they create new ones, and they change expectations in unrealistic ways. We know that ANet has considered public testing, and rejected it. And from the remarks, it's likely that someone internally brings this up about as often as we do on Reddit or the forums, and each time, after consideration, they make a business decision that the game wouldn't be enough better off to make it worthwhile.

    (Incidentally, there are many other things that ANet probably could change internally that would also result in fewer issues. Some of them are as simple as having the right people in change/code management positions. But like most process-oriented changes, they are more easily said than done.)

    So by all means keep suggesting it. I think ANet should hear how unhappy people are with bugs, and especially those that seem to be preventable. Just keep in mind that, on their side of things, there are other considerations — it's a business decision whether it's the best way to deal with bug prevention. Make the suggestion and then let them decide whether its' "cheap" or easy or not.

    "Face the facts. Then act on them. It's ...the only doctrine I have to offer you, & it's harder than you'd think, because I swear humans seem hardwired to do anything but. Face the facts. Don't pray, don't wish, ...FACE THE FACTS. THEN act." — Quellcrist Falconer

  • The V.8759The V.8759 Member ✭✭
    Yes

    @Illconceived Was Na.9781 said:

    @ReaverKane.7598 said:

    @Illconceived Was Na.9781 said:

    @ReaverKane.7598 said:
    It can be publicly available, but they can still veto who accesses it. League of Legends places restrictions on who accesses it.
    GW2 can do the same. They can for example launch a PvP test build, and only open the server to players with 80+ rank on PvP and x amounts of matches, or plat+.

    That still comes with all the issues of a public server, and none of the benefits of a private one. People with Plat+ or 80+ rank don't automatically become good testers, good at giving feedback, or even able to offer criticism that applies generally rather than their specific point of view. We have many such examples already of skilled players with myopic views about the game.

    Again, player feedback is as useful as they find it. The true virtue of such a server is data (not player feedback, but server logs and statistics) that approximates closer to how it will behave on live, as opposed to how it behaves when being tested exclusively by the QA team.

    And again, you're only looking at it from a player perspective. Yes, absolutely, it's valuable to find out that a particular skill or chat code only bugs out on a full map or that the stress-testing scripts missed a key factor of 100 humans playing together (or 10000 logged on at the same time). But it's an enormous amount of work to make it possible for the occasional situation in which that is the aggravating factor.

    The point is that the amount of useful data is limited, and it's not favorable to the costs. There are set up costs, maintenance costs, and change & code management headaches. The hardware is the easy part; it's all the human resources that go into it that make it a bad deal overall.

    Put another way, public test servers do prevent some issues, they create new ones, and they change expectations in unrealistic ways. We know that ANet has considered public testing, and rejected it. And from the remarks, it's likely that someone internally brings this up about as often as we do on Reddit or the forums, and each time, after consideration, they make a business decision that the game wouldn't be enough better off to make it worthwhile.

    (Incidentally, there are many other things that ANet probably could change internally that would also result in fewer issues. Some of them are as simple as having the right people in change/code management positions. But like most process-oriented changes, they are more easily said than done.)

    So by all means keep suggesting it. I think ANet should hear how unhappy people are with bugs, and especially those that seem to be preventable. Just keep in mind that, on their side of things, there are other considerations — it's a business decision whether it's the best way to deal with bug prevention. Make the suggestion and then let them decide whether its' "cheap" or easy or not.

    I honestly this shouldn't even be part of the discussion. We simply don't exactly know how much resources a PBE would cost, and yes maybe it's not worth it. I'm asking it at the players, so I'm looking for a player perspective. Saying yes doesn't mean that you think it's possible from ArenaNet's side, it means you think player testing would be a good option right?

  • @The V.8759 said:
    I honestly this shouldn't even be part of the discussion. We simply don't exactly know how much resources a PBE would cost, and yes maybe it's not worth it. I'm asking it at the players, so I'm looking for a player perspective. Saying yes doesn't mean that you think it's possible from ArenaNet's side, it means you think player testing would be a good option right?

    I'm paraphrasing the reasons the devs have given in the past. They don't think it's cost effective. My opinion is moot.

    If you want my best guess anyhow, then no, I can't see it as being an overall good thing. There will be some great feedback, some good data, but not enough. The amount of useful intel compared to distracting rants, the amount of usable data... it's a lot of trouble for little return. But, I have no personal experience with testing games. I have seen "public" testing for software products used by software developers and ... even then, the feedback & data is really weak, unless there are powerful incentives.

    "Face the facts. Then act on them. It's ...the only doctrine I have to offer you, & it's harder than you'd think, because I swear humans seem hardwired to do anything but. Face the facts. Don't pray, don't wish, ...FACE THE FACTS. THEN act." — Quellcrist Falconer

  • starlinvf.1358starlinvf.1358 Member ✭✭✭✭

    @ReaverKane.7598 said:

    @Illconceived Was Na.9781 said:

    @ReaverKane.7598 said:
    It can be publicly available, but they can still veto who accesses it. League of Legends places restrictions on who accesses it.
    GW2 can do the same. They can for example launch a PvP test build, and only open the server to players with 80+ rank on PvP and x amounts of matches, or plat+.

    That still comes with all the issues of a public server, and none of the benefits of a private one. People with Plat+ or 80+ rank don't automatically become good testers, good at giving feedback, or even able to offer criticism that applies generally rather than their specific point of view. We have many such examples already of skilled players with myopic views about the game.

    Again, player feedback is as useful as they find it. The true virtue of such a server is data (not player feedback, but server logs and statistics) that approximates closer to how it will behave on live, as opposed to how it behaves when being tested exclusively by the QA team.

    Thats not a consistent truth either. A test server is a control environment, so players are going to behave differently due to a multitude of factors. A world boss event for instance is massively different with PUGs then it is with the equivalent of a Guild Group (ie at least minimally coordinated); and on top of which, different skill levels of players can get away different types of strategies. Modern glass cannons being a glaring example of something good players can run with little trouble, but even moderately causal players would struggle with in fights that last more then 3 seconds.

    If you want useful data, you have to replicate the environment. The problem though is any environment with new elements will need weeks to reach proper equilibrium..... same as a live environment. You want a wide testing pool to capture as much berth of data as possible, but also needs to be focused enough to cut down noise inherent in the environment. This makes an ideal test pool an oxymoron. It has to contain players who universally agree on and seek a realistic average of performance, but also has a level of expertise of the system that moves them to one end of the bell curve. But the more expertise players have, the more they lean toward a status quo that lets them keep a competitive edge. Its an inherent conflict of interest, and very few players have the scruples to maintain a neutral position. Limiting access to certain demographics only biases that pool further; and is only desirable if your game (or the game mechanics) only targets an easily defined group.

    But theres also extra layer of issues with a Test server being in exclusion to the main game...... a conflict of incentive. Even during the POF Beta Weekend, I had to make a choice between trying out the new Especs in WvW, or giving up potential rewards earned for that weekend when the kill count would be insanely high. I went with the former, because it gave me a chance to try some radically wide range of builds with ZERO investment into gear. Giving me enough info to know what builds I might like and pursue post launch. My priorities were squarely self interested first, game conscious second. And a PBE would have to expect a significant chunk of the population to not be like that, AND devote a substantial amount of hours per week to generate enough data for useful metrics.

    Since none of that is possible without being disruptive or not properly reflected in behavior in a controlled environment, I'm in full support of introducing a test batch into a live environment, as it is an accurate representation of the day 1 gold rush. Out of all the data they can collect, this is by far the most crucial to curb long term damage in a persistent environment. The test batch is temporary, and they're impact ONLY disruptive on the play environment. But if they find exploits or cheese tactics that generate undeserved gains in the economy or competitive modes, they can stay contained to the test batch while the Devs clean up the updates to address them for the live release. The only down side is Salty players who cling to the status quo, and might throw a tantrum because a disruption exists..... but often times you can quell that dissent by simply offering bonus rewards to get them involved. Both sides of this are getting something out of the situation; Test players get an experience and/or knowledge to put toward the real release (which 99% is gonna be shared data by the end of the test period), and Live players get extra stuff for "tolerating" a change thats going to inevitably happen when it goes live anyway.

  • Carighan.6758Carighan.6758 Member ✭✭✭✭
    Yes

    I voted yes, but the cynist in me wants to say that the production environment works perfectly well as a beta testing one for 6 years now. Players are used to things being glitchy and broken.

  • Mea.5491Mea.5491 Member ✭✭✭✭
    No

    Weird, my comment mysteriously disappeared from the 1st page but I got no notice/reason from moderators. :o Spooky!

  • Ben K.6238Ben K.6238 Member ✭✭✭✭
    No

    Public beta tests are usually ruined by unreasonable expectations and unreasonable feedback. The only time it makes sense is when you need to test scaling with large numbers (i.e., stress testing).

  • Cynn.1659Cynn.1659 Member ✭✭✭
    edited July 25, 2018
    No

    Most devs can't accept feedback anyway.

    How dare this random nobody tell me how to do my job i did for "Insert arbitrary number of years"

  • Alga.6498Alga.6498 Member ✭✭✭✭
    edited July 25, 2018
    Yes

    If this is like a PTR, then yeah, it would be so useful.
    Just like Blizzard does before their patches goes live.

    | Separatist | Nightmare Court | Inquest | White Mantle | Sunspears | Loyalists | Ascalon | Kryta | Ebonhawke | Elona | Istan | Kourna | Vabbi | Cantha | Luxon | Kurzick | 71 characters | "Rally to me, Ascalonians!" "Keep Ascalon in your heart." "May the Gods protect you." "Be blessed by the Six."

  • Yes

    After all those bugs and miss balancings I would say we are in dire need of it.

©2010–2018 ArenaNet, LLC. All rights reserved. Guild Wars, Guild Wars 2, Heart of Thorns, Guild Wars 2: Path of Fire, ArenaNet, NCSOFT, the Interlocking NC Logo, and all associated logos and designs are trademarks or registered trademarks of NCSOFT Corporation. All other trademarks are the property of their respective owners.