ynsaen opened this issue on Jun 15, 2004 ยท 44 posts
ynsaen posted Fri, 18 June 2004 at 6:34 PM
uh oh. lol ;) My intention in this thread is to examine why it is so many folks are calling for the inclusion of OpenGL in Poser, when I can see no noticeable benefit given that Poser needs to remain independent of hardware. I'm against doing so -- I don't find that OpenGL is any adequate asolution to the problem than the current one is. What I've gleaned so far is that the major rationale for it is to speed up being able to turn the camera in large, complex scenes. Which seems a bit odd to me given that the tradeoffs are going to be much greater. OpenGL, per spec as published, does not have the feature set that the current display renderer does. It strikes me as cutting off the nose to spite the face. So while the cool stuff is neat to talk about, it's sorta off base, lol. That said: "Another new method I doubt will ever be used by poser." (re: displacemet mapping) Check out the stats on Firefly, hon. It's primary strength is in displacement. This feature of the engine is about 80% exposed, as well (possibly fully -- would have to check with Stewer to be sure). It's in use now. Witht he new displacement mapping capabilities of zBrush, we'll start seeing some impressive stuff coming out using it as both the renderer and the program are explored further. Firefly also supports HDRI and IBL -- and IBL is partially exposed to the end user at this time. It does posess sub surface scattering -- the skin shader node within P5 it is a partial implementation of it, as well. It fakes it better. The original Sree3D engine that drives the preview system is still present, and I won't argue that it's so heavily built into the core code that it would take an untoward effort. It's what drives the Preview system in Poser and Bryce, and used to drive in Carrara as well. But I will stand by my statements that Poser 5's current rendering engine can, in the right hands, produce finer renders than and hardware based game system in use at present. The link to the article on Desbrun's algorithm is a nice one, and while it could be immediately applicable, what's the licensing structure on it, and did you not notice it's primary focus is in polygon reduction for 3D scanners, with an aim towards real world applicaitons? While game systems will tke advantage of it and eventually modelers will enjoy it, it's applicability towards rendering capability outside of speed is marginal at best. Nor am I saying that poser can cut it with the "big boys" -- which are light years of game rendering systems. And the top of the line ones are tied to specific cards because they combine functions and are designed specifically with the card in mind. I am decidedly not a CL fanboy. Absolutely not. Although my escape there comes from being a girl. Also, I've been here easily as long as you, though the name has been changed from my more innocent days. (three or four times, I think) Software renderers with hardware assitance in general are out of the price range of the casual hobby user, or even most starving artists. The gist of this thread, though, isn't the features of the final rendering agent, but rather the benefit of OpenGL in the realtime preview screen instead of the current one. and good responses, btw. Thanks!
thou and I, my friend, can, in the most flunkey world, make, each of us, one non-flunkey, one hero, if we like: that will be two heroes to begin with. (Carlyle)