Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!sdd.hp.com!news.cs.indiana.edu!noose.ecn.purdue.edu!dirac!gibbs.physics.purdue.edu!murphy From: murphy@gibbs.physics.purdue.edu (William J. Murphy) Newsgroups: comp.sys.amiga.advocacy Subject: Re: Why buy a NeXT? (was: a pointless flamefest about the blitter) Message-ID: <5139@dirac.physics.purdue.edu> Date: 21 May 91 06:12:25 GMT References: <0sHyo86@cs.psu.edu> <91137.021356MBS110@psuvm.psu.edu> Sender: news@dirac.physics.purdue.edu Organization: Purdue Univ. Physics Dept, W.Lafayette, IN Lines: 39 In article kls30@DUTS.ccc.amdahl.com writes: >In article <91137.021356MBS110@psuvm.psu.edu> MBS110@psuvm.psu.edu (That neat guy, Mark Sachs) writes: >> >>The only possible use I can see is in education, more specifically in >>computer science labs and any gullible undergraduates you can con >>into buying the things. You can't even use a NeXT in a "real" lab (what are >>you going to plug the instruments into?) >~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ >Into the DSP port or use a SCSI interface converter. The DSP port can run >a serial connection at a couple of megabits/sec (16x88kx2). So getting >the data is NOT a problem. I agree with Kent. You can use the SCSI to IEEE488 convertor fromo IOTech. There are four different A/D systems that will do the job of 16-bit sampling. As for how it will do in a real lab, I will let you know. Our research group is gearing up to use the NeXT and a PC in a Hearing Science lab. I am not happy about the PC, but we already have invested about $3000 in a signal generation system for it, might as well use it. One problem among the many we are investigating is whether or not evoked otoacoustic emissions are really like/similar to spontaneous otoacoustic emissions, but below the noise floor of our spectrum analyzer. Although the problem could have been solved equally well with an Amiga, we did it with the NeXT (just to start another flame fest. 8-) ). We calculated an FFT of 2 million points with the NeXT from 160 seconds of 20 kHz sampled ear canal signal. I saw peaks I've never seen before using the spectrum analyzer with its 1024 pt. FFTs. So, I would have to take issue with Mark Sachs about the usefulness of the NeXT in the lab. It has certainly been more usefull in performing calculations. I would say that Amigas could equally handle the calculations, but when is C= going to implement Virtual Memory AND Protected Memory in AmigaDOS? I seem to recall Valentin Papelea (Spelling?) talking about how to implement VM about a year or two back before he was hired by C=. So where is it? Bill Murphy murphy@physics.purdue.edu West Lafayette, IN: Where progress takes a back seat to tradition.