From: sunisv!sense8!tom@Sun.COM (Tom Coull) Subject: Sense8 WorldToolKit Information Date: Thu, 7 Nov 91 09:40:51 PST Dear Bob: Several of our customers have forwarded to us WTK information requests that they have seen on sci.virtual-worlds. Would you please post this information for all those interested? Many thanks. ****************************************************** WORLDTOOLKIT Version 1.01 Program Description ****************************************************** INTRODUCTION: WorldToolKit (WTK) version 1.01 is a rich set of over 230 functions written in C that enables a developer to build 3-D graphical and "virtual reality" applications. From writing custom sensor drivers to rapidly prototyping real-time simulations, WTK offers an intuitive set of functions that provide a wide range of functionality. FEATURES: High Performance -- WTK features a high-speed renderer incorporating the latest flight-simulator technology to provide superior performance on a PC. Real-time Texturing -- Apply texture bitmaps from your DVI or Targa files to the surfaces of objects. For example, you can apply an actual wood-grain image to the top of a table model, or the elevation view of an entire tree to a single polygon (using the "transparent" texture feature to be able to peer through the tree branches). "Hypertext"-Style Organization -- Use "portals" to create links between associated worlds. When a portal is "crossed", the user jumps to the named world. Use this feature to build arbitrarily complex virtual worlds. Data Import -- Use the file import facilities to create 3D graphical objects from DXF, the WTK neutral ASCII file format, and other formats. Dynamic Control of Lights -- Light your virtual worlds with an arbitrary number of directed lights. Move or change the intensity of these lights in real-time. Superior Application Development -- WTK gives superior application development power and runtime performance, yet complete applications can be built with surprisingly few lines of code. Device Drivers -- WTK includes drivers for many of the sensor devices commercially available, including: Logitech`s "Red Baron", Ascension Technology Corporation's "Bird", the CiS "Geometry Ball Jr.", the Spaceball Technologies "Spaceball", the Polhemus Navigational Sciences "Polhemus" and the Fake Space Labs "BOOM". SPECIFICATIONS: Functions are provided for: o simulation management. o user-interaction (such as polygon picking with the mouse). o universe entry and exit. o terrain generation (checkerboard, random, or from data). o viewpoint manipulation and control. o stereoscopic viewing. o sensor control and interaction. o portal creation. o texture application, manipulation, and removal. o interactive polygon color editing. o wire-frame or flat-shaded polygon display. o graphical object creation and task assignment. o hierarchical control of graphical objects. o intersection testing. o light creation and manipulation. o object and texture animation. Hardware Requirement: 80386/387 or 80486 (recommended) IBM compatible PC AT with 4MB RAM. Single ActionMedia 750 Delivery board required for monoscopic viewing. Two ActionMedia 750 Delivery boards required for stereoscopic viewing. Resolution: Adaptable, 256x240 to 512x480. Color Space: 16 bit true color for flat-shaded polygons and textures. Textures can be unshaded, shaded or transparent. Compiler Requirements: 32-bit C compiler (MetaWare High C 1.71 recommended). Memory Manager: DOS extender (PharLap 386 DOS Extender 4.0 recommended). DISTRIBUTED BY: Artificial Realities Systems S.R.L., Via Rombon, 11 - 20134, Milano, Italy, Phone (02) 264-12898, Fax (02) 264-13279 Asahi Electronics Co. Ltd., 4th Fl., KMM Bldg., 2-14-1 Asano, Kokura Kita-ku, Kitakyushu City 802, Japan, Phone (093) 511- 6471, Fax (093) 512-1780 Micron/Green, 1240 N.W. 21st Avenue, Gainseville, FL 32609, Phone (904) 376-1529, Fax (904) 376-0466 Robert McNeel and Associates, 3670 Woodland Park Avenue North, Seattle, WA 98103, Phone (206) 545-7000, Fax (206) 545-7321 Sense8 Corporation, 1001 Bridgeway, #477, Sausalito CA 94965, Phone (415) 331-6318, Fax (415) 331-9148 Virtual Presence, Ltd. 25 Corsham Street London N1 6DR U.K. Phone (071) 253 9699, Fax (071) 490-8968 FOR MORE INFORMATION: Contact your nearest distributor, or Tom Coull at Sense8 Corporation. -- Tom Coull (well!sense8!tcoull) Sense8 Corporation (415) 331-6318 (phone) (415) 331-9148 (fax) ****************************************************** WORLDTOOLKIT Version 1.01 Price List ****************************************************** COMPONENTS: ---------- WorldToolKit software (Single Machine Development License) $3,500 ActionMedia 750 Delivery Board (release 2.13.12 software) $2,495 MetaWare High C 32-bit Compiler/Debugger (Version 1.71) $895 PharLap DOS-Extender (Version 4.0) $495 OPTIONS: ------- Spaceball Technology "Spaceball" $1,595 Imagetects ImageCELs Library $495 Truevision VIDI/O Box (RGB to Composite signal converter) $995 Cable Package $225 (Includes two sets of 4-RGB to 4-RGB cables with BNC connectors and two 15-pin high-density to 4-RGB cables with BNC connectors.) PACKAGES: -------- WorldToolKit Introductory Package $6,295 (Includes WorldToolKit, one ActionMedia 750 Delivery Board (with free hardware upgrade), MetaWare High C compiler/debugger and PharLap DOS-Extender.) WorldTool Virtual Reality Presentation System Upgrade $4,995 (Includes one additional ActionMedia i750 Delivery Board for stereoscopic image generation, two Truevision VIDI/O boxes, two SyncWizard boxes and the Cable Package.) WorldTool Virtual Reality Development System $12,995 (Includes WorldToolKit, two Intel ActionMedia 750 Delivery Boards, MetaWare High C compiler/debugger, PharLap DOS-Extender, two Truevision VIDI/O boxes, two SyncWizard boxes, the Cable Package, Spaceball Technology "Spaceball" and the Imagetects ImageCELs Library.) TO ORDER: -------- Please call or fax Tom Coull at: Sense8 Corporation 1001 Bridgeway, #477 Sausalito CA 94965 415.331.6318 (phone) 415.331.9148 (fax) WorldToolKit and SENSE8 are trademarks of Sense8 Corporation. DVI and ActionMedia are trademarks of Intel Corporation. All other brand and product names are trademarks or registered trademarks of their respective holders. US prices are shown and are subject to change without notice. Tax or shipping not included. From 70353.3056@compuserve.com Sat Nov 9 23:28:20 1991 Received: from ihb.compuserve.com by milton.u.washington.edu (5.65/UW-NDC Revision: 2.1 ) id AA27727; Sat, 9 Nov 91 23:28:17 -0800 Received: by ihb.compuserve.com (5.65/5.910516) id AA14364; Sun, 10 Nov 91 02:28:23 -0500 Date: 10 Nov 91 02:14:42 EST From: Christopher Fry <70353.3056@CompuServe.COM> To: "Human Int. Technology Lab" Subject: Re: Integrated lasers for Eyephones Message-Id: <911110071441_70353.3056_CHC13-2@CompuServe.COM> Status: R We're discussing displays that have variable density pixels, high density in the center (where your fovea looks) and lower density the further out you get. >> Once we've got the display itself, a bunch of clever graphics software >>needs >> to be written. I expect each pixels to take considerably longer to draw >>than >> regular-array square pixels [and remember, we need to fill an oval, not a >> rectangle]. If our general-purpose hardware still isn't fast enough for >> manipulating those 150K pixels, then this sounds like an ideal candidate >>for >> special purpose parallel hardware. >One way to solve this problem is to spend memory on it: make the frame >buffer match the maximum resolution over the whole field of vision, use >ordinary algorithms on a fast standard processor to write the pixels, >then use special hardware on the video output side to filter output to >the display with the filter kernel based on where in the field of vision >the pixel is. Yes, you've said the obvious solution of how to drive an irregular display. If we're talking about 120 pixes per degree, then for, say 100 X 80 degrees that's 115M pixels [or at least words in your frame buffer.] Updated at 60Hz and we've got a computation and data-flow problem between the processor(s) and your big frame buffer. My hope was that it is fewer computrons to fill a 150K word frame-buffer than a 115M word fram buffer. Each one of my irregular pixels takes more computation than one of your regular ones, but is it 700 to 800 times more? If yes, then there's no computational advantage to the irregular pixels, but I suspect clever algorhithms could do it in under 100 times more, perhaps only 10 times more.