Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!elroy.jpl.nasa.gov!sdd.hp.com!spool.mu.edu!uunet!rosie!aozer From: aozer@next.com (Ali Ozer) Newsgroups: comp.sys.amiga.advocacy Subject: Re: How fast can the NeXT do Animation Message-ID: <684@rosie.NeXT.COM> Date: 7 May 91 20:23:20 GMT References: <1991May5.212810.28755@wam.umd.edu> <3gbG4*$*1@cs.psu.edu> <1991May6.024252.12256@leland.Stanford.EDU> Sender: news@NeXT.COM Distribution: usa Organization: Next Computer, Inc. Lines: 34 Nntp-Posting-Host: twinpeaks.next.com In article <1991May6.024252.12256@leland.Stanford.EDU> David Hopper writes: >In article <3gbG4*$*1@cs.psu.edu> Michael D Mellinger writes: >>don't know how fast the NeXT really is. The #'s quoted here are >>*quite large* and I'm not sure I will believe them until I see some >>better animation with my own eyes. >Nor I. Thanks for the skepticism, Mike. >>From: aozer@next.com (Ali Ozer) >>(from /NextDeveloper/Demos) with a screen-sized window. It runs at about >>70-75 frames/second on the NeXTstation, and at about 35-40 frames/second >>on the NeXTstation Color. With a monochrome image, the frame rate >>goes up to 80-85 frames/second on the color machine. With the default sized >>window, the frame rate is 105 on *both* the NeXTstation and the >>NeXTstation Color, and 130 with a monochrome image on the NeXTstation >>Color. (These are all highly informal timings by the way, I just did >>them while typing this message up. Not official benchmarks by any >>means!) >Not official? These are ridiculous. I wonder what he used to do the >timing. A time function at the top of the animation loop was what I used; the source is on the machine, you can try it yourself. This app uses compositing, which is fast, and it composites sections which are not the size of the whole screen. Thus these values are correct and repeatable. You can even use something like the following to time compositing --- the below code composites a 400 x 400 dark gray area, and takes 3ms doing it. With 12-bit color, it takes 16 ms/frame. With these values in mind, I hope you can see that the numbers I quoted above are perfectly reasonable. /g gstate def 0 0 400 400 0.333 setgray rectfill 500 {0 0 400 400 g 400 400 Copy composite} repeat Ali