From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me X-Spam-Level: *** X-Spam-Status: No, score=3.1 required=5.0 tests=BAYES_40,FROM_NUMERIC_TLD, INVALID_MSGID,REPLYTO_WITHOUT_TO_CC autolearn=no autolearn_force=no version=3.4.4 X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 103376,bdaec5f760ac4ee1 X-Google-Attributes: gid103376,public From: steved@pacifier.com@199.2.117.163 (Steve Doiel) Subject: Re: The rate you do the things you do... Date: 1996/08/01 Message-ID: <4tpb8p$7n5@news.pacifier.com>#1/1 X-Deja-AN: 171808689 references: <4to49s$595@dfw.dfw.net> organization: Pacifier BBS, Vancouver, Wa. ((360) 693-0325) reply-to: steved@pacifier.com (Steve Doiel) newsgroups: comp.lang.ada Date: 1996-08-01T00:00:00+00:00 List-Id: David Weller writes: >I have a feeling this will raise more questions than answers, but here > goes... > > >I'm trying to cut through the hype and understand whether Windows NT >can support "simulation software" rates in the 30-60Hz range. We have >a, um, enthusiastic MS supporter that sez, "Sure, no problem!". >Before I expend labor hours attempting to prove them wrong (or right, >for that matter), I'd be interested in feedback from anybody else in >this community that has gone through such a venture yet. We're not >looking for hairy details, just a general range for now (tops out at >10Hz? 5Hz?). I personally am VERY leery of a "business-based" OS >being able to support real-time scheduling rates, but we have a lot of >management pressure to examine PC-based COTS products for future >approaches. This ain't my idea, folks, but I do have a professional >responsibility to shoot it down^H^H^H^H^H^H^H^H^H^H^H^H^H examine its >potential :-) > I am looking forward to hearing other responses. We are using NT for what we call our "GUI Router". An embedded computer sends data to the GUI Router which relays the message to a Workstation using TCP/IP sockets. I have seen good results... except when doing disk accesses on the system running the GUI Router with other applications, at which time there are significant delays. We are running our GUI Router at "high" priority, and will soon try running it "real time" to see if we can get rid of delays during disk access. With our current arrangement 1K makes it round trip in approx 30 msec. Steve Doiel