9 replies to this topic

[Monkey Lover]

I don't understand why we don't test the final builds to find bugs. Every major patch there are tons of bugs. Even simple ones like jumper jets sound effects not working.

Its almost like we're only getting a preview and its less about testing.

[Kaptain]

lol

While I would have agreed with this in the past I am quite happy that the latest round of PTS changes took a great deal from the communities feedback.

[Monkey Lover]

Kaptain, on 05 July 2017 - 03:32 PM, said:

While I would have agreed with this in the past I am quite happy that the latest round of PTS changes took a great deal from the communities feedback.

They're doing the same thing with the skill tree. They took some feed back then 2 week before they do a big change without testing.

Testing is almost worthless if we dont test with final values. Only thing we're testing now is visuals of the weapons.

Edited by Monkey Lover, 05 July 2017 - 03:37 PM.

[50 50]

Because it would never get released?

[Reno Blade]

release schedule is tight and "wasting" time for PTS deployment that can be used to fine-tune and build the production patch is one part.

If we would run for 2 more months now, we would be able to test a lot more, but we don't and nobody would be willing to wait for this big of a patch for "fine tuning".

I would love to see more drastic changes (e.g. 1/4th values for pulse lasers, or doubling all values of LRMs) but this is not a sandbox for players, but a testbed for PGI to make sure nothing is broken from the mechanics and we have a rough balance.

Balance is an ongoing process anyways, so there will be endless changes on production in the next x months for balancing anyway.

[Ruar]

I assume it's a funding thing. They don't seem to have many assets for patches and coding improvements so any test has to be limited because they can't afford to spend a few rounds fine tuning. Otherwise we'd get planned change notes released weeks, maybe a month, out and then the first test for a few days, second test for a few more days, and then final test for a few. They don't tell us plans because they are kind of stuck with whatever they first create and have to live with it. The skill tree is a prime example.

[Tyroki]

Ruar, on 05 July 2017 - 10:09 PM, said:

I assume it's a funding thing. They don't seem to have many assets for patches and coding improvements so any test has to be limited because they can't afford to spend a few rounds fine tuning. Otherwise we'd get planned change notes released weeks, maybe a month, out and then the first test for a few days, second test for a few more days, and then final test for a few. They don't tell us plans because they are kind of stuck with whatever they first create and have to live with it. The skill tree is a prime example.

They might have had more money overall, had they not both screwed up so much or tried to screw everyone over on numerous occasions.

[Ruar]

Tyroki, on 05 July 2017 - 10:24 PM, said:

They might have had more money overall, had they not both screwed up so much or tried to screw everyone over on numerous occasions.

Agreed. I stopped spending with the way the skill tree design was handled and I won't spend again unless they change the ST design to something easier to use. I used to keep spending and hope the devs would eventually change but now I wait till the change happens. I just can't support a company who refuses to use a better player design because of pride.

[tokumboh]

It does seem a false economy:

From my view they could have started the PTS earlier by essentially doing a PTS for each weapoon type and at least iron bugs out of them separately. how they capture the data is also interestingly vague there is no questionnaire no real systematic approach as to what worked, what didn't and why. So basically you are left to the whim of feelings and emotion rather than data.

it is what got them into the mess due to the skill tree. They said they looked at the data but when people started giving them their own circumstances it appeared that what they had concluded from the data was wildly different from everyone else. We are obviously luck that there is less than 20-30K active players in any one month and the number that play more than a match a day is less than 15K in some months so I would have thought gather better metrics would be easier.

As to the software development itself. it seems like we are the alpha beta and gold testers. I am sure that there will as many changes after the this goes live as there have been before. The reality is what is a good mech chages basically every 6 months due to quirks etc and so there is no real incentive to get it right as we have a patch every month.

What concerns me is that new content seems to take a back seat to basically tinkering with quirks, where is the new maps, game modes, solaris, what about FW, most of these thing have taken a back seat. For example I would have thought you could have take the grim plexus approach and created maps from resources you already have as an example as a quick fix

Since the development is opaque we will never know or understand and I fear that much of their rational will not stand up to scrutiny

[Doctor Dinosaur]

1. Test in production
2. Roll in some patches
3. ???
4. Profit!

Jokes aside, the PTE is not the right place to test balances, not enough testers resulting in too small matchups (ye olde complain about 4vs.4). I also guess that the PTE population will most probably consist of more seasoned players who have different views of what is a "good" weapon/system than some casual. But the game has to be balanced for both, so "testing" in live environment it is...

PGI could (and should imho) reward PTE participance (is that a word?), but this could in return harm the live server population, so there's that...