> Quick and easy way out, compressed textures for the clouds as in .dds > textures > :D > (sorry about that, couldn't help myself :P)
I doubt it's a general solution since I get 30% framerate reduction for dds textures. But you can do it for yourself and see if it helps your case - all models and texture sheets are in Models/Weather/. > I've played around a bit with the config settings, the greatest effect > was > noticed with the visible range slider, and max clouds in loop. Max. clouds in loop is only effective when you have dynamical weather on - which you shouldn't have on in the first place if you are low on framerate. There it determines how many clouds per frame are moved - the rest remains unmoving. Visible range is by far the most effective as there's an area law involved - the whole range of the slider corresponds to a factor 13 in performance consumption. But the slider is runtime - so you can use it according to current cloud density. > The most efficient way seems to have everything else maxed out, and > selected. Not always. Asymmetric buffering is good if you fly straight, but very bad if you circle - for soaring it's completely unsuitable as you'd constantly shuffle things in and out of the buffer. With CACHE_ALL it's probably more efficient than it used to be... > Visible range hits really hard >25 km here. (this is the biggest hitter, > I > guess it's to be expected since this increases the area drawn) > Max clouds in loop > 0.3-0.4 of the slider (I'd guess about 200 km ?) > (too bad > the sliders don't display the actual value somewhere :( ) The slider really represents an number (from 100-400) - the range is derived from that number dependent on the density of clouds. > Also massive effects where noticed when disabling any of the buffering > options, > or lowering their value. (I guess system ram helps here) Basically you unload up to half of the clouds in your rear hemisphere - for sure that frees RAM. The idea of the option is that you can have more visibility in forward direction where you need it at the expense of your rear hemisphere at which you usually don't look. > Back to the efficiency thingy, clouds further out (>20 km ?) should > really be > big boxes, as it's not worth it do draw them one by one, don't know how > to > solve this but I guess we could come up with an idea that's both visually > pleasing and efficient to draw. I've been thinking about the problem for a year now (my rational was that layered clouds from 30.000 ft should be visible for much more than 55 km, and the only way to do that is using sheets. I haven't been able to come up with a viable solution :-( So if you have one, let me know... it's a much harder problem than one would assume naively. Cheers, * Thorsten ------------------------------------------------------------------------------ Enable your software for Intel(R) Active Management Technology to meet the growing manageability and security demands of your customers. Businesses are taking advantage of Intel(R) vPro (TM) technology - will your software be a part of the solution? Download the Intel(R) Manageability Checker today! http://p.sf.net/sfu/intel-dev2devmar _______________________________________________ Flightgear-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/flightgear-devel
