SONDRICK@HASARA11.BITNET (Rick Jansen) (06/21/91)
In our visualization lab we plan to use a Sony LVR6000 video disc for recording computer animations and still pictures. This appears to be not quite so trivial as we thought it would be at first. Especially the synchronization of video equipment and the program delivering the pictures itself is not quite clear (to me). For example, Wavefront Data Visualizer has an animation facility, where you can specify the number of frames to generate. It can take a while (seconds) before the next frame is ready. As far as I can see (correct me if I'm wrong please), there is no way to trigger the video equipment to record a frame as it is finished, from Data Visualizer. It IS possible though, to capture the frames in RLB files, and process them later, recording the animation frame by frame onto video disc. This has to be done with a self-written application. Silicon Graphics has a product called Video Creator to record onto video tape. Video Creator (VC) is connected to a Silicon Graphics workstation. It is connected to the RGB signals, and is controlled via the SCSI interface from an Iris. Via a coax cable VC controls a little box called V-Lan box, which controls the actual video recorder. In a window on the Iris screen you can control the operation of Video Creator, like start and stop the VCR, record single frames etc etc. Schematically this setup looks like this: +----------+ RGB +---------------+ PAL/NTSC +--------+ | SGI |-->--------| Video Creator |--------->| VCR | | Iris | +---------------+ +--------+ | | | | +-------+ | | |-->---------+ +-->| V-Lan |-->---+ +----------+ SCSI +-------+ control We intend not to use a VCR, but the mentioned Sony disc. We are still investigating if a Sony disc can be controlled by a V-Lan box. (These boxes are configurable.) Using a disc that can hold 36250 frames per side requires some administration, which is not covered in the above schematic. We could imagine a setup like the following scheme, where the laserdisc is controlled via a Mac or PC, while also an administration of recorded sequences and stills is kept. +----------+ RGB +---------------+ PAL/NTSC +-----------+ | SGI |-->--------| Video Creator |--------->| Sony 6000 | | Iris | +---------------+ +-----------+ | | | | +-------+ | | |-->---------+ +-->| PC |-->---+ +----------+ SCSI | + | control | disk | +-------+ Well, so much for ideas and plans. Does anyone on the net already have experience with a similar setup including a Sony disc and Silicon Graphics equipment? Any information would be very welcome! Rick Jansen SARA, Amsterdam, The Netherlands
johnc@boulder.ColoState.EDU (John Cooley) (06/26/91)
In article <91172.145508SONDRICK@HASARA11.BITNET> SONDRICK@HASARA11.BITNET (Rick Jansen) writes: >In our visualization lab we plan to use a Sony LVR6000 video disc >for recording computer animations and still pictures. This appears >to be not quite so trivial as we thought it would be at first. > >For example, Wavefront Data Visualizer has an animation facility, >where you can specify the number of frames to generate. It can take >a while (seconds) before the next frame is ready. As far as I can >see (correct me if I'm wrong please), there is no way to trigger the >video equipment to record a frame as it is finished, from Data >Visualizer. It IS possible though, to capture the frames in RLB >files, and process them later, recording the animation frame by >frame onto video disc. This has to be done with a self-written >application. > >Silicon Graphics has a product called Video Creator to record onto >video tape. Video Creator (VC) is connected to a Silicon Graphics >workstation. It is connected to the RGB signals, and is controlled >via the SCSI interface from an Iris. Via a coax cable VC controls a >little box called V-Lan box, which controls the actual video >recorder. In a window on the Iris screen you can control the >operation of Video Creator, like start and stop the VCR, record >single frames etc etc. Schematically this setup looks like this: > (Schematic and other stuff deleted) > >We intend not to use a VCR, but the mentioned Sony disc. We are >still investigating if a Sony disc can be controlled by a V-Lan box. >(These boxes are configurable.) > >Well, so much for ideas and plans. Does anyone on the net already >have experience with a similar setup including a Sony disc and >Silicon Graphics equipment? Any information would be very welcome! > >Rick Jansen >SARA, Amsterdam, The Netherlands We use a Sony LVR5000 and a Silicon Graphics Personal Iris. We also have a Panasonic AG6750 Time-Lapse VCR. We will soon be adding a souped-up Amiga A2000 with a Video Toaster and an editing VCR. It currently looks like we will control the editing deck with a manual jog-and-shuttle rather than with V-Lan or other computer control. The LVR5000 and the AG6750 are both controlled via RS232 ports from the SGI. We do not have Video Creator. The Sony technical people tell me that the LVR5000 can be controlled by a V-Lan device if it is first converted to RS422. This is not hard to do, but it costs about US$500 to do it. A small hardware hack would make the port switchable. We use both the Panasonic and the LVR5000 for raw video. The Panasonic is used for student experiments; the LVR is the last step in preparation of presentation material. Video on the LVR is edited onto a final tape with an editing deck, or for something really fancy, is sent to a professional post-production shop. We always lay RGB onto the video disk. It will play it back in NTSC if required for tape. The controlling software for both the LVR5000 and the Panasonic was written by G. W. Hannaway and Associates, Boulder, CO. However, we have the programming manual for the Sony (and could get one for the Panasonic) so writing our own drivers probably would not be too difficult. The SGI is on a small network which has over 3 GB of temporary disk space. We use this space to stage frames for recording. We maintain a log of recorded frames on the video disk. We use the Scientific Visualization Suite from the National Center for Supercomputing Applications in Champagne, Illinois (the suite contains applications for the Apple Macintosh as well as the SGI), apE from the Ohio Supercomputing Center, the Wavefront Visualizer Series, and software from the Pittsburgh Supercomputing Center and the San Diego Super- computing Center to generate frames. The ideal solution is a complete V-Lan setup, with RS422 ports on the SGI (check out VideoFramer - Video Creator is usually overkill) and the videodisc, along with a pair of V-Lan controlled editing decks to make master tapes with titles, scene changes and sound. The cheapest such deck would be a Panasonic AG7750 (US$6500) S-VHS deck. Sony 3/4" decks are better; if you have a lot of money BetaCam is broadcast grade. Along with all this, an Abekas frame buffer helps. The Video Toaster is one relatively cheap solution for titles, effects and video switching in NTSC. There are others. Anyway, we don't have enough money for all this stuff. We'll be content with relatively easy insert editing with some titles, etc. We can already do the raw video reasonably well. If someone wants to ask in a few months, I'll probably be able to say something about how the Toaster and the editing are working out. Buy the way, I do know about the PC-VCR. It has some advantages, but we can't control it with the SGI without writing our own drivers (can be done, but we like to avoid that sort of thing). Also, the time code on it is proprietary, and, while it works for locating a particular frame, it's pretty clumsy. It would be easier if it had computer control, time code, *and* jog-and-shuttle type control. John Cooley Advanced Technical Confusing Colorado State University