589: No mention of Id software has been removed. I thought this was cleared up last project. I am proud of CAT3D, and I would have a problem with the removal of credit to the development team. 591: The version number is on the intro screen 592: The new piracy screen is in 660: The map view is a debugging tool, not a feature. 682,694,695: The hang is fixed. Whenever a monster had no viable move it went into a loop (the trolls were all added at the last minute). Good catch. 697: The graphic screw up was fixed by adding two blocks to the map Everything else: The single lines are a result of roundoff error between two transformations. When a wall segment disapears it is a result of a back trace hitting a side that should by invisable. The objects showing up through walls is a result of the z buffer being kept in pixels (a poor design decision). Not mentioned by you: The thick bar shown when you look almost straight down a long wall is a result of the tangent table maxing out in 16 bits, causing the texture mapping to think it can optimize the wall drawing. The fish eye view when you turn right next to a wall is a result of not having enough compiled scaler routines (memory constraints) to scale thinks closer than the view plane. These problems are understood, but the corrective measures are not really feasable. I think CAT3D still stands well even with these warts. Over the past several days I have developed a ray tracing/intersecting 3-D refresh to replace the image based system from CAT3D, and it has addressed all of these problems. I would put it in CAT3D except for: it uses 286 specific code in a few places, but no big deal to replace. It is designed to work with a demand paged graphics caching system, still no big deal. I am unsure of its memory requirements, potentially a terminal problem. I haven't tested it with scaled sprites yet, this would require one or two additional days. If gamer's edge ever gets an ok to do a hard drive installable game you will probably see the new system. John Carmack