Forum: Poser - OFFICIAL


Subject: Thoughts on a new, improved Poser

Paloth opened this issue on Nov 21, 2007 ยท 30 posts


Dale B posted Sun, 25 November 2007 at 2:31 PM

You forgot the kitchen sink in there, bopper...... :P P7Pro is slated to have up to a 10 node network renderer and a 64 bit render engine, so that's coming. Normal mapping is an advancement of bumpmapping; where a Poser current bumpmap is a grayscale image, a normal map uses RGB and sometimes the alpha channel. The color channels would provide XYZ 'bump' data, increasing the depth and detail a map can generate, and the alpha channel usually handles albedo, or reflectivity. It would be very nice, but it -might- break the current bump mapping scheme. On the other hand, the increase in non polygonal detail would be worth losing the old bumps, just like losing the bums from P4 was a godsend.... Weightmapping is a kind of rigging system. Instead of the falloff zones and spherical rotation zones Poser currently has, a weightmap system would use the same boning, but have 'maps' of where the mass of a body part is, and how it changes due to the effect of the bones involved being moved. IIRC, A Poser mesh would no longer have to be cut, as the weightmaps would control the deformation and folding along the joints. It would probably be a nightmare to retool all the existing figures for weightmapping, but if Poser could support both the current system for legacy figures, and a weightmap system for more modern and new figures, it would be an animator's wish come true. SM needs to add an 'Animation' tab to the workspace, and move all the animator's tools there, and make a few new ones. A 3 pane graph editor, with default XYZ translation/rotation would be very helpful. A dopesheet has translation controls available at a keyframe of a bodypart would make the animation pallete a true dopesheet. That way those who just pose with it could stay on the front page and not have any animation stuff to confuse them, and the animators could have -all- their tools ready and onscreen. Realtime or much faster dynamic cloth is likely doable, but would require the core app to be multi-thread capable. All cloth sims run a simulation cycle; Poser is at a big disadvantage due to the high number of polygons in the meshes. If weight mapping were implemented, it should be possible to build a low poly bounding box slightly larger than the actual figure and have the cloth collide with that...which would cut down sim time dramatically. It would be nice if someone were coding a softbody dynamics system for Poser....but you would run into the same polygon count calculation issue. As for the modeler/texture painter, you might as well take a look at Modo and be done with it. And getting half of that for $129-$200 would be a steal.... But its a nice dream.