I think it would be far more beneficial to tackle playability snarls than the superficial stuff like graphics. Things like the insanely complicated and troublesome cooking process, which oddly enough got tweaked into the mess it's in rather than starting that way. However, part of that problem is that it got changed by people that don't actually use the mechanics for it. It might look better in code but it is so not worth it in terms of cooking anything efficiently. Or coming up with plans like preventing players from equipping weapons in a city to drive home the point that the guards exert influence whether a GM is actively playing them or not, or refining Dakkrus curse to nail repeat offenders rather than nail everybody when it could have been a /die to unstick when a GM wasn't around.
@Weltall, chances are that the SkunkWorks group won't suggest anything that can be simply fixed by bringing up the issue. The whole point is just identifying problem areas if I understand right. Very few things will be conceptual problems that a committee can come together to rethink. And on the hardware issue, PS is in no position to expect people to have hardware like is required for other games. It doesn't remotely have the content necessary to justify straining people's computers... and to be honest, PS runs worse on less than required hardware than a lot of other more complicated MMOs. The problem is that PS needs to be streamlined and it doesn't have the people to do it.
However, this attitude is the kind of stuff that made me quite trying to get a group to find solutions to problems. It's not my group wasn't working. They did great, but what's the point if the few devs that are still around are mostly coders which have the "no!" mentality (Although I will give RlyDontKNow credit for doing better about that these days). It's not that you expect a magic wand to materialize and fix stuff but at least if a problem is identified and accepted it might be one day be fixed. But turning your head and acting like there is no possible solution is part of what keep PS running in circles.
maybe because i don't swim in the sky?
Let's analyse it then.
The current engine team is composed of me and rlydontknow + 1 prospect, we both have our duties which come to priority as it's obvious and we both have computers which works well with ps. The prospect didn't give any trouble with the engine too.
So what you are asking us to do? find a ghost issue no one here can reproduce? supporting cards here no one has (provided you need also skills in writing shaders which are a different topic than generic programming)?
Do you want to do something useful provide some crash backtrace else your claim is unexistant as it's not possible to investigate on it "i've a crash somewhere in this one milion lines of code (sum cs and ps) i'm sure you can find it" - i'm sure we can't find it. (i think this well explains how it works:
http://imgur.com/jacoj ) Do you think i should go and write hate posts in the gnome bugtrackers as they don't fix multimonitor support since 2003? (yes it's still broken in many ways).
Also the stability of the client has increased a lot but the macosx client is hindering any possible release. shall we drop it? i don't think so even if i wish it daily it's not possible so what? hope someone will be able to build it before the next year.
Have you seen the server? the crashes on it are rare if not almost absent. why? we have ways to see when it crashes what happened and so we can fix it. Right now the crash we have are exotic things like a gm reloading an npc while someone is storing an item on it and clicks in the same exact instant the store button. and this got fixed too quite fast.
yeah i'm too realist... but trying to put up as a problem something which is evident to anyone like some hw crashes (and sorry but an nvidia 7600GS is not so a great piece of hw it's quite crappy but works well. you can't even play at decent framerate a game like oblivion on it while ps works quite well, here we are talking of people wanting to play with intel cards or things which are almost as old as the game itself).
Plus another thing people like to do around here is comparing apples with oranges yes they are both with a circle shape just like both directx and opengl are used to program graphic hw to make something which barely looks like a real world image but opengl wasn't supported by many vendor decently till some years ago like with ati), intel support is called crap in linux even for simple 2d graphics lately and in many cases it's at the level of opengl 1.x: they are complaining about it even to show some wobbly windows in kde, nvidia does up and downs sometimes. directx is the most well supported api but it works only on windows and anyway today it wouldn't be possible to switch to it without rewriting an huge part of the game. Likewise those stating we want to be back to crystal space 1.4 that won't happen. It took almost a year to do the transition and no one is going to convert it back, plus support for cs 1.4 is quite limited: just the other day someone was tryng to build it on a windows machine and it didn't build correctly. I don't know if he was able to figure it out but almost no one could help him as all the development effort is geared toward 1.9 which will become 2.0 and the architecture of the both is changed in a lot of ways.
There is no point in people pointing out that the client crashes and they should be fixed, i complain myself about it more than you think but if there is nothing of the above then it will stay like this because there is no workforce nor a reproduce-able scenario.
Do you want us to work like any major game or software producer?
ok -we will work on it-
glad now? PS: it works like security flaws in windows.
i was just told the bootloader of my arm embedded system is broken and i won't see anything working till december and even then i could lack functionalities if all goes well. let's see if they will respect even what they said. ah the revision of september worked flawless... and we are talking of a thing which could make a business fail and it's supposed to be supported by giants of the computer industry.