matth@mars.njit.edu (Matthew Harelick) (02/12/91)
Hello
I started to read your call for a new language in sci.virtualworlds and
I have a question. When you use the term "Real-Time" do you mean graphical
systems with Hard Real-Time deadlines? The reason I ask is that I have an
interest in computer graphics and I also have a very good opportunity to
get involved in Real-Time Systems and Real-Time research. If Computer Graphics
applications are running into hard real-time limitations then that may be
a good niche for me to do my Ph.D. thesis in. I look forward to hearing from
you soon.
- Matthew Harelick
matth@mars.njit.eduhagerman@ECE.CMU.EDU (John Hagerman) (02/13/91)
matth@mars.njit.edu (Matthew Harelick) writes: > When you use the term "Real-Time" do you mean graphical systems > with Hard Real-Time deadlines? The reason I ask is that I have an > interest in computer graphics and I also have a very good > opportunity to get involved in Real-Time Systems and Real-Time > research. Unfortunately, the meaning of the term "real time" depends on who is talking. For those who work on Real Time Systems, "Real Time" (note the capitals) means "hard deadlines," as you are aware. For everyone else, "real time" (all lower case) means "fast," or, more accurately, "fast enough." You might hear someone say "I want real time graphics" when they mean they want a graphics system that will allow them to do smooth animation. The main difference is that for Real Time systems there will be a massive failure if a deadline is not met, while for real time systems there will just be frustrated users. However, for virtual reality, you might say that a frustrated user indicates a massive failure, so perhaps some formality, as you suggest, would be useful. For example, Scientific American had an article on flight simulators a while back, and there was a discussion of the desire to spend lots of computation to accurately render the scene where the pilot is looking, and to use less computation (and thus less scene accuracy) elsewhere. Therefore, a good research project would be to figure out how to join the human factors issues with soft deadline requirements to create a methodology for Real Time Virtual Reality programming. I don't know how much the Real Time community knows about soft deadlines; if that area has not been addressed, it will easily result in many research projects by itself. - John -- hagerman@ece.cmu.edu
naimark@apple.com (Michael Naimark) (02/13/91)
There's an intuitively obvious use of "real time" in cinematography which may be helpful here. Back in the early days the cameras were hand cranked: slow motion was achieved by "over-cranking," fast motion by "under-cranking." When a movie is "not too slow" and "not too fast" it is in "real time." (ie., when there exists a temporal correspondence between recording and playback).