What We Need Is Another War!

[Mietz]

PS2 desperately needs a public test-server.

If you look at the Roadmap, some of the items that are being worked on are quite substantial changes (stupid, but substantial).

SOE needs to have a test-server to test these things out before they go live, we can't have the live version be this constant cycle of "buff-nerf-redesign" because it wears on the player population. (especially the paying kind)

With a game like PS2 you can't just internally test things with 20 dudes and theorycraft the rest, its too complex a game with too many variables. The test server would provide the metrics to evaluate proposed changes and implement actually tested solutions instead of a "good guess" by the developers.

I mean how else are you going to test base redesigns? Class revamps?
Are you serious to release your guesswork on the live servers and then either hope for the best or let -us- sort out your errors?
I'm sorry but BETA is over and things need to be handled in a professional way. Paying customers are going to get massively miffed when you start fucking with weapon and vehicle balance or start removing or adding functionality on a weekly basis.

Seriously.

Test Servers.

Now.

[maradine]

I mean how else are you going to test base redesigns? Class revamps?


The same way everyone else does. Internal QA and a chunk of cash to Volt for a week's worth of temp bodies.

[maradine]

Actually, in SOE's case, it would probably be E4E or VMC, but point stands.

[Palerion]

I agree with the OP.

When you're dealing with real money, changes should be more thoroughly explored and tested.

I think test servers would help allow this.

[capiqu]

I'm with you.

[Mietz]

The same way everyone else does. Internal QA and a chunk of cash to Volt for a week's worth of temp bodies.

They would hire 2000 people for a week?
That seems either incredibly inefficient or downright stupid.

I mean, these things cant be tested with 200 guys on a server, if you are changing base-flow it will affect the capture priorities. Its not just localized consequences in a fight, there will be changes to general continent flow that might lead to other unforeseen consequences.
i.e The Crown Effect
Terrain plays a big role as well so you'd have to test this on multiple continents and layouts.
You need a full server, period.

Internal testing isn't going to cut it. Blizz doesn't do it, EVE doesn't do it either, the reasons for that is that with games of this scale changes have a domino effect, tweak something slightly here and you might end up with a complete shitstorm somewhere else.

Also a week is precious little time to assess any kind of larger changes that need to permeate through the population.

[SturmovikDrakon]

I'm all for it

Maybe this way we could one day actually get to test the GDC layout of Zurvan, since the game is actually lacking in any true CQC.








(I can dream can't I?)

[maradine]

Internal testing isn't going to cut it. Blizz doesn't do it, EVE doesn't do it either, the reasons for that is that with games of this scale changes have a domino effect, tweak something slightly here and you might end up with a complete shitstorm somewhere else.


You don't think internal testing on private servers occurs before builds go up on SiSi? Supplemental, you think a month on SiSi shakes out nullsec-wide effects into Tranq?

[Sirisian]

SOE needs to have a test-server to test these things out before they go live, we can't have the live version be this constant cycle of "buff-nerf-redesign" because it wears on the player population.
Kind of a naive viewpoint honestly. It assumes the cycle can be broken, which you should know by now, it can't. I wrote about this (http://www.reddit.com/r/Planetside/comments/17c2xm/ps1_vets_did_planetside_1_ever_find_balance/c84cui8) in a recent thread on reddit on a similar subject.

I think a test server is a good platform for testing bug fixes and crash type things. Testing balance, assuming it exists with any plans or fixes the developers create, really needs to be done with a large population to get data. The same is true for base redesigns. I don't see the point in testing with a possible smaller population on a test server. I personally like seeing changes to bases and redesigns even if they turn out not to work. Then again that's assuming the test server is hard to join because of special patching rules or downloading the whole client again and thus has a low population.

[Rolfski]

I can remember the devs talking about this actually.

[Rothnang]

I honestly never liked the idea of public test servers, because it puts a disproportionate amount of influence over the game in the hands of people who spend more time on the test server than actually playing the game.

[Mietz]

You don't think internal testing on private servers occurs before builds go up on SiSi? Supplemental, you think a month on SiSi shakes out nullsec-wide effects into Tranq?

Obviously I think internal testing will not go away?
Did I ever say that this should -replace- internal testing?

I said that internal testing is not enough, not that it should replace internal testing.
It is -better- than purely internal testing, thats not opinion, thats fact.
There is nothing lost by SOE to have a public test server.

Kind of a naive viewpoint honestly. It assumes the cycle can be broken, which you should know by now, it can't. I wrote about this in a recent thread on reddit on a similar subject.

I think a test server is a good platform for testing bug fixes and crash type things. Testing balance, assuming it exists with any plans or fixes the developers create, really needs to be done with a large population to get data. The same is true for base redesigns. I don't see the point in testing with a possible smaller population on a test server. I personally like seeing changes to bases and redesigns even if they turn out not to work. Then again that's assuming the test server is hard to join because of special patching rules or downloading the whole client again and thus has a low population.

I'm sorry, I don't read le reddit.
If it won't "break" the cycle, then it will at least minimize the impact of changes from what we have now, its still a win.

I honestly never liked the idea of public test servers, because it puts a disproportionate amount of influence over the game in the hands of people who spend more time on the test server than actually playing the game.

How?
Do the testers get to dictate policy?
How is this different from what we had in beta, now?

Its about giving SOE/devs the ability to gather large scale metrics before they roll out changes to live.

[ChipMHazard]

I can remember the devs talking about this actually.

Aye it was mentioned some time ago that they were going to make a test server. No idea if the idea was scrapped.

[Rothnang]

How?
Do the testers get to dictate policy?
How is this different from what we had in beta, now?

Its about giving SOE/devs the ability to gather large scale metrics before they roll out changes to live.

During Beta there were no live servers, so people didn't have to make a choice between spending their time playing on the test server and having a say in features under review or playing on the live servers where their characters get to keep their progress.

I have yet to encounter a game where test servers don't lead to a measure of test server elitism. In some games like WoW it's so bad that when a major content patch launches guilds that relocated all their efforts to the test server the month before end up beating the new raids the same day because they have already done them two dozen times by the time they are supposedly new.

[Dougnifico]

Agreed. Priority to premium players and former beta testers!

[Hamma]

I'm sure we will be seeing this coming soon.

[Mietz]

During Beta there were no live servers, so people didn't have to make a choice between spending their time playing on the test server and having a say in features under review or playing on the live servers where their characters get to keep their progress.

I have yet to encounter a game where test servers don't lead to a measure of test server elitism. In some games like WoW it's so bad that when a major content patch launches guilds that relocated all their efforts to the test server the month before end up beating the new raids the same day because they have already done them two dozen times by the time they are supposedly new.

This isn't WoW. It's not a PVE game and has no controllable content.
If a measure of "test server elitism" will exist is debatable.

And if there is, so what?
People playing on the test server should be aware of the consequences and yes, they will most likely play there more. But 2000 players isn't exactly going to drain the population to some irredeemable levels (and if it does, the game is already dead).

Yes test servers will be highly frequented and played on, everyone will want to test out new toys and designs, and thats the point of a test server:
A drawn out feature stress-test.

Agreed. Priority to premium players and former beta testers!

Absolutely against it, that would lead to exactly the problem Rothnang describes above.

Random lottery draws from all active players and an access-queue with an option to opt-in/out of test-server access/queue is the only way.

The last thing a dev team wants is a contaminated population with a disproportionate amount of vets and paying customers.

[Rothnang]

The really important part is just that when you have a test server and you take feedback from people playing on it you skew the feedback you get by default.

Let's say you put in a feature that makes the game easier for casual players, but isn't very popular with hardcore players because the lack of the feature gave them a huge advantage before.

Who do you think is going to spend time on your test server, try that feature out and spend extra time writing feedback about it? The casual players who play for 30 minutes a day, or the hardcore players that spend 6+ hours a day in the game? (And yea, it's difficult enough as it is to get any feedback from casual players)



Not saying there are no advantages to having a test server, but having one does tend to really skew how people respond to and interact with features in testing.