4 May 1998 23:12:00 -0400

Related articles |
---|

Interpreters and (math) speed stefan.wils@zorro.ruca.ua.ac.be (1998-05-04) |

Re: Interpreters and (math) speed marlet@irisa.fr (1998-05-07) |

Re: Interpreters and (math) speed di6adag@cse.hks.se (Adam Granicz) (1998-05-07) |

Re: Interpreters and (math) speed bernecky@acm.org (Robert Bernecky) (1998-05-12) |

From: | stefan.wils@zorro.ruca.ua.ac.be (Stefan Wils) |

Newsgroups: | comp.compilers |

Date: | 4 May 1998 23:12:00 -0400 |

Organization: | UUNET Benelux (post does not reflect views of UUNET Benelux) |

Keywords: | performance |

Hi

I'm kind of new to this area, but I am in the process of designing a

'graphics language' : a language which has a very extended set of

graphics primitives and a huge number of mathematical routines and

other routines. Algorithms like raytracing, or lens flare

implementations and z-buffering are then to be programmed in this

language, kind of like RenderMan but much extremer and flexible.

So my question is : what do people, who are thinking about

compiler/interpreter technology all the time, think of using an

interpreter for such a task ? Could it be made fast enough, especially

for calculations (note that support for real-time graphics is not the

purpose ! We're talking about rendering realistic graphics here,

e.g. over a network. But even then speed is an issue.)

I don't know if Im gonna get response, but all tips and hints about

whether or not it can be done and what techniques to use are more than

welcome. A tip on where to start would be nice too.

Thanks in advance,

Stefan Wils (stefan.wils@zorro.ruca.ua.ac.be)

[Depends on the granularity of the operations. If each op is big enough,

the interpreter overhead vanishes into the noise. That's why interpretive

matrix math systems are popular. -John]

--

Post a followup to this message

Return to the
comp.compilers page.

Search the
comp.compilers archives again.