Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!uflorida!simulation From: simulation@uflorida.cis.ufl.edu (Moderator: Paul Fishwick) Newsgroups: comp.simulation Subject: SIMULATION DIGEST V17 N4 Message-ID: <24172@uflorida.cis.ufl.EDU> Date: 22 Aug 90 14:19:30 GMT Sender: seeger@uflorida.cis.ufl.EDU Reply-To: simulation@uflorida.cis.ufl.edu Lines: 251 Approved: fishwick@uflorida.cis.ufl.edu Volume: 17, Issue: 4, Wed Aug 22 10:16:59 EDT 1990 +----------------+ | TODAY'S TOPICS | +----------------+ (1) RE: Simula 67 (2) Simulation Tools in Physical Education (3) Numerical Integration and Stability (4) Review of Simulation Tools (5) RE: MAC Simulation Software (msg 1) (6) RE: MAC Simulation Software (msg 2) * Moderator: Paul Fishwick, Univ. of Florida * Send topical mail to: simulation@bikini.cis.ufl.edu OR post to comp.simulation via USENET * Archives available via FTP to bikini.cis.ufl.edu (128.227.224.1). Login as 'ftp', use your last name as the password, change directory to pub/simdigest. Do 'type binary' before any file xfers. * Simulation Tools available by doing above and changing the directory to pub/simdigest/tools. ----------------------------------------------------------------------------- Date: Thu, 9 Aug 90 07:12:41 EST From: Richard E. Nance - SRC To: cho@cs.ucla.edu Subject: Simulation of Distributed Systems Cc: simulation@bikini.cis.ufl.edu => Organization: UCLA Computer Science Department => => => Hi, Netter, => => I plan to simulate resource or load management methods in Time Warp, and => I am curious whether anyone have already built such simulators. => I know that Orna Berry builted one Time Warp simulator, but it was written => in SIMULA 67 whose compiler is not available to me. Simula 67 for Macintosh/MPW is in the public domain and can be obtained from Lund Software House AB Box 7056 S-220 07 LUND Sweden or copied from an existing disk which they distribute. => cho@cs.ucla.edu Dick Nance nance@vtopus.cs.vt.edu ------------------------------ Return-Path: To: comp-simulation@beaver.cs.washington.edu Path: uw-june!cypark From: cypark@cs.washington.edu (Chang Yun Park) Newsgroups: comp.simulation Subject: REQUEST: high-level tools for physical education Date: 13 Aug 90 23:06:39 GMT Organization: U of Washington, Computer Science, Seattle This inquiry is for my friend in Korea, looking for simulation packages or languages. (Originally, he asked the following question to me, but I couldn't help him. I think this group is the place to have some ideas.) Please tell me any hints or guides. I will forward all of them to him. Thank you. ------------- As a professor in physical education, he has been thinking about computer-aided education; for example, he wants to show his students how a change of running style has effect on speed. Visualization would be better if possible. All information in his field was already collected, and some basic plans of what and how to do are made. However, he has no idea on how to use computers. Although he is willing to learn, he doesn't know where he can start. Any experiences for similar applications ? Are there any (commercial) programs or languages easy to learn and use, and with some graphic display ? ------------- ChangYun Park (cypark@june.cs.washington.edu) ------------------------------ To: comp-simulation@sdcc6.UCSD.EDU Path: sdcc6!beowulf!heirich From: aheirich@UCSD.EDU (Alan Heirich) Newsgroups: comp.simulation Subject: Continuous System Simulation Summary: Help: integration of autonomous systems with mixed variables Keywords: numerical integration, dynamical systems Date: 17 Aug 90 00:52:41 GMT Sender: news%sdcc6@ucsd.edu Distribution: all Organization: California Institute of Technology Nntp-Posting-Host: beowulf.ucsd.edu I am writing a package for simulation of continuous dynamical systems. The core of simulator is a numerical integration loop. I have what may be a rather naive question about numerical integration, and my guess as to an answer to my question. I would be grateful if any numerical integration experts out there can tell me whether my head is screwed on straight. In return for helpful information I would be glad to share my package when it is ready. (First version is for IBM-AT's). Most continuous system simulation packages require you to write first order rate equations. In other words, all differential equations have the form d x/d t = f (x, t, ...) It would be nice to model autonomous systems (where the O.D.E.'s are not all written with respect to t), e.g. d x/d t = f (x, t, ...) d y/d x = g (y, x, ...) For the sake of argument, assume that you are using 2nd order Runge-Kutta numerical integration to simulate the trajectories of your state variable(s) x (and y). Then the second example would update according to the following rule: xk1 = f (x, t, ...) * delta_t xk2 = f (x + xk1, t + delta_t) * delta_t x_next = x + (xk1 + xk2) / 2 delta_x = x_next - x yk1 = f (y, x, ...) * delta_x yk2 = f (y + yk1, x + delta_x) * delta_x y_next = y + (yk1 + yk2) / 2 My question is this: are common integration methods (like 2nd order Runge-Kutta) numerically stable in this sort of situation? My guess is that they are not, because the size of delta_x is not fixed and is hard to predict. Thus the method might, for example, step over a singularity in the y trajectory and miss it. Since I think this answer is probably correct, my *real* question is: are there any methods which don't have this problem? Or more generally, how do people manage to simulate autonomous continuous systems defined by differential equations? Alan Heirich heirich@aurel.cns.caltech.edu ============================================ Program in Computation and Neural Systems California Institute of Technology Pasadena, CA 91125 ------------------------------ To: comp-simulation@ucsd.edu Path: cod!burkley From: burkley@cod.nosc.mil (V. J. Burkley) Newsgroups: comp.simulation Subject: Review of simulation tools/environments Date: 17 Aug 90 16:25:52 GMT Distribution: usa Organization: Naval Ocean Systems Center, San Diego Hi, I am somewhat new to creating simulations. I am currently trying to find out people's experiences with two products. The first is Simscript II.5 from CACI and the other is SES Workbench (I don't know who makes it). I would also appreciate any information on other products that run on Sun workstations that are suitable for large scale simulations. Please reply to my e-mail. I will post a summary if there is interest. Thanks. Joe Burkley ------------------------------ To: uunet!comp-simulation@uunet.UU.NET Path: inmet!ma From: ma@inmet.inmet.com Newsgroups: comp.simulation Subject: Simulation tools advice needed Date: 16 Aug 90 20:06:00 GMT Nf-Id: #N:inmet:27600001:000:238 Nf-From: inmet.inmet.com!ma Aug 16 16:06:00 1990 I am planning to do some performance modeling of a large distributed software system. Are there any good simulation tools available on the Macintosh for doing this? I will appreciate your advice. malgosia askanas (ma@inmet.inmet.com) ------------------------------ To: comp-simulation%munnari@munnari.OZ.AU Path: munnari.oz.au!stats.mu.oz.au!koula From: koula@stats.mu.oz.au (Koula Courtot 6605) Newsgroups: comp.simulation,sci.math.stat Subject: Request for simulation packages Keywords: simulation Date: 20 Aug 90 22:53:34 GMT Sender: news@cs.mu.oz.au Could anyone please let me know what packages are available on simulation of random processes for a Macintosh system. Could you please e-mail your responses to me. My email address is: koula@mugga.stats.mu.oz.au Thanks. ------------------------------ END OF SIMULATION DIGEST ************************