This just makes NO sense.

[Ghoest9]

Im fine with SOE focusing on performance over content at a conceptual level.
But the story they are telling us makes NO sense at all.
Its so absurd that Im becoming curious as to why the more prominet fan voices seem to be intentionally ignoring the obvious.

Only a minority of developers are code monkeys with the skills to optimize a game.
We have all heard the extremely reasonable explanation many times before most of the people on staff simple can not help with stability and perfornces issues.

So what changed? Did they send all the devs to night school for computer science?



the story coming from Smedely is false - everyone should acknowledge the obvious here.

[Timithos]

I'm sticking with Smed's verbiage of "shifting priorities" as a realistic statement. I agree with you that it's silly to think most of the dev team could be working on optimization. They'd be shooting themselves in the head - and the foot - if they focused so much resources on optimization and dropped everything else. Although if delaying content is beneficial in that the content delivered continues to degrade stability and optimization, then I'm all for delaying content until they get optimization under control. (Kind of like when they shifted more resources to bug squashing a few months ago, and now they adequately - I just say adequately - take care of bugs to my bare minimum satisfaction.

[Wahooo]

I don't think it was said as 'our whole team is working on optimization' rather the number one priority will be optimization.
The reason this would slow down other aspects is, that yeah graphics and other departments can still do what they do, but getting that stuff into the game and tested is probably some of the same resources that are used in the optimization.

So a lot of aspects can still be worked on, but only to a point and the finishing steps are going to have to wait until the optimization is complete.

I don't think there is any reason to see some conspiracy that isn't there about who maybe doing what on this game.

Besides there must have been some kind of breaking point as performance has seemed to be about the #1 complaint for most of the player base. If there has been some kind of critical mass type number in people saying they are quitting specifically due to performance? Time to really jump and do something.

[Obstruction]

it's pretty simple, they probably got to live test on the PS4 networks and realized how much optimization work they really had to do before launch this fall. seeing as they are committed to being available at launch (read: stable before launch) and that they are going to have to serve millions of PS4s with downloads on day 1...

then yeah it is probably crunch time for every department. even if you think your department is done, like textures and models, you probably need to be double checking everything or at least standing by in case some BS crops up like "oh hey we have a particle effect somewhere in the thousands of particle effects that seems to be crashing clients. find it. now."

[NewSith]

Originally Posted by Ghoest9

Im becoming curious as to why the more prominet fan voices seem to be intentionally ignoring the obvious.

So far it's nothing but a statement.


Besides, I would go as far as to say that this is probably one of the cases of "boss speaketh the wordeth", where it's just an optimistic/enthusiastic statement, based on the lack of knowledge of his resources.

[Crator]

It's odd though because Smed stated players will just have to understand they are shifting focus on optimization. As to insinuate that they are not doing anything else but this. But perhaps he just meant that updates and such are going to slow down? Or does the new stuff created by the art team have to pass through the same people doing optimizations so they can't take stuff that is made fto code into the game. In which case the art team would be creating/working on several things which would create a back log. IDK.

Originally Posted by Obstruction

it's pretty simple, they probably got to live test on the PS4 networks and realized how much optimization work they really had to do before launch this fall. seeing as they are committed to being available at launch (read: stable before launch) and that they are going to have to serve millions of PS4s with downloads on day 1...

According to Smed tweets this announcement has nothing to do with PS4. They are simply not happy with how the game is performing on the PC. With what type of data they are using the determine this I am not sure. I'm guessing several sources (internal benchmarks/community feedback/etc.).

[Kail]

You can think of developing new content like a pipeline - even if only the actual programmers were working on optimization, then that part of the pipe is blocked.

However it's important to remember that optimization happens at a global scale. Programmers enhancing some code to run faster may then make asset size the bottleneck, which the artists fix and then thread contention on the network reader/writer threads pop up, etc etc.

[NapalmEnima]

There ARE things artists and level designers can do to improve performance. Reduce poly counts, reduce object counts, that sort of thing.

The statement could Actually Be Right.

Or maybe not. Dunno.

[Ghoest9]

Originally Posted by NapalmEnima

. Reduce poly counts, reduce object counts, that sort of thing.

.

This is absolutely not happening with the PC game.

[typhaon]

I'm not sure we can do much more than judge the results.

Like others have suggested, I think this probably has more to do with the state of the game's readiness with PS4 launch looming.

[wasdie]

Originally Posted by Ghoest9

This is absolutely not happening with the PC game.

What makes you so certain? They could be adjusting LoD models to have less detail, removing polys from stuff you cannot see, adjusting models so that less polys are wasted on small details you can't see.

[Ghoest9]

Originally Posted by wasdie

What makes you so certain? They could be adjusting LoD models to have less detail, removing polys from stuff you cannot see, adjusting models so that less polys are wasted on small details you can't see.

What you are suggesting is silly but i cant actually prove it wrong.

But it doesnt really matter as artists are only a portion of the development crew that is helping to optimize the gamer performance.

[kubacheski]

Originally Posted by Ghoest9

What you are suggesting is silly but i cant actually prove it wrong.

But it doesnt really matter as artists are only a portion of the development crew that is helping to optimize the gamer performance.

I'm confident that you can't prove it wrong because it's a true statement. I mentioned in another post, drop a pebble in pool of water and see what happens. Yes you think that the artists is only small part as the engine is rendering the images on your screen, but the definitions contained in every model may need to change to interface with the changes made to the engine. Hell, due to the complexity of the graphics, it would be the absolute first place I'd start.

And each and every dev will be doing 3 things - testing, testing, testing. Then you have to have analysis of the data collected from these tests.

Is it performing better?
Did some areas actually get worse?
Is the extra check for content_xxx in the logic worth having right now as content_xxx is not implemented yet?
Yea it's cool to track how many shots fired, but do we need to also track the vector of the bullet/ray so we can save a slight amount of traffic?

There's so much in a game this size that there's a million little things to slice off of it. Someone just has to decide which 10,000 things of the million have got to go in the name of performance. (Well actually in performance improvement so that we can have better player retention, so that we can make more money, but you get the point.) And that's Smed and team.

So let them get to work. No amount of us complaining and asking questions will change their path. You can say you'll withhold your sub money and not buy that giraffe camo, but they're a path they hope will have you buying again and in the end you'll probably be thankful that they did the optimization and throw even more money at them to "reward" them for a slightly bettter gameplay.

P.S. If they're optimizing the engine, then it's complete bullshit that it wont affect PS2 on PS4 and EQ:Next. Hell it may be that they've got to upgrade the ForgeLight that currently runs the PC version due to finding some massive bug in implementing one of the other 2 games running ForgeLight.

[ChipMHazard]

Originally Posted by wasdie

What makes you so certain? They could be adjusting LoD models to have less detail, removing polys from stuff you cannot see, adjusting models so that less polys are wasted on small details you can't see.

As far as I remember they have already done some changes like this in the past.

[AThreatToYou]

Originally Posted by wasdie

What makes you so certain? They could be adjusting LoD models to have less detail, removing polys from stuff you cannot see, adjusting models so that less polys are wasted on small details you can't see.

wasdie has a very good point.

Considering how rapid content is developed in PlanetSide 2, I would not be surprised if there is something everyone who is hired onto PS2 permanently can do toward optimizing the game. My (limited) experience with video game dev tells me that cutting needless polys, shaving extra pixels off of textures, adding more depth to the LOD system, improving the occlusion boxes on the maps, and making sure animations terminate exactly when they are no longer needed are things that the art-oriented devs can do (and should be expected to be able to do).

In addition, the more user-related code monkeys like Flash/java/HTML developers really should be going back and looking over the GUI that has been thus far modified. With how I hear SOE does things (there are some developers that are not permanent, and some devs don't have to track the changes they make so long as it wasn't planned in the patch [or something like that]), looking over things that might have literally invisible functions would be important.

And let us not forget PS2's multi-core support.

With all that said, I doubt literally everyone is dropping their current priorities and working solely on performance improvements. I am expecting that it is a shift in dev priority and thus there are much more people working on improving the performance of the game than before.