14 replies to this topic

[TexAce]

New public test servers are incoming.
What we don't know exactly is, how often the numbers of everything are gonna be altered to test out new approaches on balance.

What I would like to propose is a way to let the community decide how the numbers are gonna be changed.

How I think this could work:

1. First we decide how many days a test phase should go. Should the values be altered every day? Or every week? I personally would be for every 7 days, since it gives us enough time to test the new balance setup and find improvements.

2. Devs provide us a datasheet of all values that we can decide upon (for example, heat, range, RoF of all weapons, torso twist, twist range, twist speed, arm movement speed ect of all mechs)

3. If we choose for example a 7 day rhythm, we decide that we provide a datasheet of all value changes to the devs at fixed time on the day (we have to stick to deadlines, if we fail, it's our fault).

4. Devs AGREE to implement the numbers every x days we decide

5. We decide in internal polls which datasheet we are going to provide to the devs. How?
Well if you have an idea for changing certian values, you make a proposition. If you get a prior defined amount of LIKES to your suggestion, you are entitled to enter the testserver-data-poll. For example if we decide that 30 likes is enough, every suggestion that recieves 30 likes is entering the final stage. In the final stage a poll is made, where everyone has to vote which of the suggestions has to be chosen. The one who has the most votes, will go to the devs.

This is my proposal how the test server can be fed with what the majority thinks are good test values and also way to take workload off the devs, because all they need to do is take the sheet and use our numbers, easy pie for them.

If you have another idea how the test servers could be run, feel free to tell and keep this topic alive.
Thanks

Edited by TexAss, 29 May 2013 - 08:14 AM.

[Valaska]

PGI do something progressive and what the majority of the community wants? I dunno about those chances, seeing as ECM was voted bad in about half a dozen polls, people stating they weren't happy with the state of the game, setting out a huge list of reasons and ways to balance it, and did **** all.

The test server won't do anything, PGI will still go about with "Working as intended"

[TexAce]

Valaska, on 29 May 2013 - 12:31 PM, said:

PGI do something progressive and what the majority of the community wants? I dunno about those chances, seeing as ECM was voted bad in about half a dozen polls, people stating they weren't happy with the state of the game, setting out a huge list of reasons and ways to balance it, and did **** all.

The test server won't do anything, PGI will still go about with "Working as intended"

I think a company wouldn't go the way and implement a test server when they
a ) don't need one
b ) don't want to test stuff

I'm the optimistic guy

Edited by TexAss, 29 May 2013 - 02:40 PM.

[Chavette]

Why would you need a week to decide how a handful of weapons work?

[TexAce]

Chavette, on 29 May 2013 - 02:37 PM, said:

Why would you need a week to decide how a handful of weapons work?

you don't need a week to decide how the numbers are, you need a week to test the given numbers throughout afterwards. And based on it develop new numbers. Also the polls need time. Discussion is something you want to have after the numbers have been tested, to see what can be improved.

You hand over the numbers. They implement them, you test the new numbers 7 days, discussion starts if they are good, what could be done to make them even better and so on.

When some numbers have been found which the community likes, they can be implemented to the live servers for everyone. Because thats the goal, find the right balance.

Edited by TexAss, 29 May 2013 - 02:55 PM.

[Esplodin]

When I first heard that PGI was adding test servers I was like "LOL, whut?"

There are no need of test servers, because what the OP is suggesting should be being done right now on the regular servers. Open BETA (actually open Alpha, since it is not anywhere near feature complete) apparently isn't any kind of BETA I'm familiar with. I can't say I'm a fan of the industry push to release half finished product while taking real money and think that's OK as long as it has beta somewhere in the title. It's cool and all we get to play earlier, but for $deity's sake changes should be coming at a much brisker pace - especially in the balancing area.

Beta yes. Pseudo Beta with separate "honest these are really the real beta test servers where we do the normal beta stuff" - no.

[Sam Slade]

Put regular people on the internet and they turn into crazed monsters; horrible combination of percieved familiarity and total annonymity... I'm guessing that's why we're getting test servers.

Thanks to much larger game design firms turning 'beta tester' into a wedge marketing tool(*preen preen* My gaming group got invited to beta test, did yours? *preen preen*) we now have an entrenched attitude to 'beta testing' that precludes any large scale change (the most recent example of this I can think of is Counterstrike: Global Offensive, I was put in the closed beta because of the number of members a specific gamer group I belong to has... no questions asked, you have numbers, you get in, buy our stuff!)

When the MWO Closed/Open Betas kicked off it was widely praised and widely cursed by the Closed Beta testers for going Beta at such an early stage; this allowed for a greater level of customer interaction as well as a greater potential for negative press. To my mind it is glaringly obvious that, due to the aforementioned attitude to testing, making heavy handed changes to the game to test and gather data will cause issues(LRMs and SRMs are exhibits A & B ).

This test server will allow major tweaks and general screwing around to find a good balance level without throwing napalm on the forums every week; I see the necessity of these new servers as an example of just how immature the wider internet gameing community has become.

Edited by Sam Slade, 30 May 2013 - 08:51 AM.

[Fooooo]

Pretty much what Sam just said. (about why they dont just use the server everyone is on now)


As people are already spending money etc and generally treating this beta as a released game, fiddling round with numbers constantly or testing out mechanics for 1 week only to have the game go back to how it was before that will only annoy those customers.

Much better to just create a new server, like they are going to, so they can avoid all of that nonsense and have the people who want to test this stuff out go there.

TBH they should of had this from the start......


Not really sure about the poll idea from the OP, I think its a bit optimistic.

I think what will end up happening is PGI will change things, we will test them and provide feedback.

They will change them again.

We will provide feedback.
(all the while catching any bugs missed by internal testers on the upcoming build.)

Then they push to live server. (with or without specific changes that were liked by testers...which may get changed again, or just pushed to the next build instead etc.)

Process repeats.

Edited by Fooooo, 30 May 2013 - 09:12 AM.

[Esplodin]

The problem is that every change is an entirely new engineering project. MG damage sucks? Crit system! LRMs OP/Suck? Flight path changes and splash dice roll changes!

The majority balance of tweaks so far have been massive changes. Changes need to be little tweaks WAY more often, and reduces the risk of Armageddon of the month. It also sets up an expectation that things change, and get fixed quickly if broken. Every implementation plan should have an associated rollback plan.

While I agree the Internet is a beautiful cathedral with a cesspool in the center, the majority of the vitriol is due to the flaws that happen again and a again that would be obvious with 5 min of testing, rather then general trolldom.

[James DeGriz]

Two things. OP? No.

Secondly, the reason for the test server coming at this stage could be, ooh I don't know, setting up a test platform now so that it's there and in place ready for the launch in a couple of months?

[Sam Slade]

Esplodin, you are correct when you say that many of the changes thus far have been big... but perhaps that's why they made them, on the open servers; big changes require a large volume of data to accuatly measure.

When it comes time to start the fine tuning mini-tweaks to damage numbers, heat numbers, RoF, etc... it would be better to do them on a small scale because the results will be obvious and the impact on the wider 'meta' reasonably estimable. The test servers will mean that we are not stuck with repeated downtime and the like just to change the damage value of an SRM by 0.02(if that was done incrementally now there would be something worse then napalm on these forums). We need these servers for them folks who aren't down with the 'testing' half of 'beta testing'.

[Kragmore]

I agree and disagree a bit here.

First of, i don't think there is a need for a seperate test server at this time. We are in beta and we are supposed to test stuff. bringing in a test server would lower the amount of data gathered.
After launch when community warfare is launched there need to be a test server so we can tst stuff before they go into the competative nature of CW.

As for these kind of tweaks, yes for the love of everything that is holy! We need continues smaller updates to be able to tell what works and what doesn't. Huge sweeping changes is something that should only be tested inhouse.
Though I'd say we need these cahnges every 14 days. That is 9 days of testing and 5 days of implementation changes.
Smaller damage tweaks will of course be done quicker then mechanics changes.... stuff take time to do! (After all, they need to Fix numbers, Compile, Do a new build, Test internaly, discuss results, Fix numbers, Compile, Do a new build, Test, and so on)

[Redshift2k5]

It's not just about 'how many days'. It's about what is coming up and what needs to be adjusted- one day, they could have to adjust a stat value ten times, while for a different topic changing a bunch of values once would be sufficient.

We are also not in a position to dictate what should be tested, some things will require system overhauls while other are just mere numbers tweaks.

Forcing a number of days is not the right approach, you need flexibility.

[Ihasa]

If PGI does their test server like Cryptic Studios does (and they resemble Cryptic more and more every day) their test server thought process will go like this:
Is the game playable despite bugs and balance or incomplete features?
If yes push to live on schedule. Discuss fixing someday.
If no, schedule a fix and push to live on schedule anyway.

I have no faith that a test server will help anything at this point.

[C E Dwyer]

Other test servers I've been involved with, worked two ways..

Everquests test servers were open all the time, you could import your characters into them and give feed back (it worked like an extra server with the odd wipe now and then).

WoT internally test then open the test servers for a 3 day period, three times, before placing on the game servers people get oppotunies to suggest tweaks and highlight anything internal testing failed to pick up, for MWO I think this or something similar is best