fox@bek-mc.caltech.edu (Jim Fox) (03/02/89)
I work in a biological laboratory investigating visual function. We use a Masscomp mainframe to generate full field textured displays for testing high level visual processes. We have an optic bench for creating simple luminance bars that can be rotated, scrolled and expanded or contracted. This and the data acquisition are controlled by a PDP 11/34. We need to replace our data acquisition hardware, and necessarily, our software. We are looking into a variety of microcomputer based systems. The majority of good data acquisition systems that we have found are based on an IBM pc bus. I am the pro-Amiga advocate, I think a 2000HD, or a 2500 system with a bridgecard running the data acquisition system on the pc side, and the amiga side doing low-level control of some lab peripherals, and some graphics displays (eventually replacing the optic bench) would be an extremely attractive approach. Particularly because we would like to be able to maintain some standard communications between the mainframe and the acquistion unit. The mainframe speaks Unix and X, so what better than an Amiga? Can the net help me out? I need to know if anybody out there has tried a similar system, or know of someone who has and can give me some info. The list of hardware and software we are looking for is included below. It is both incomplete (we are still adding to it) and over blown (we are asking for more than we need but will take what we can get). Thanks for any help, and the usual, please respond by e-mail, I will summarize to the net if there is enough interest, etc, etc, etc. Data Acquisition Hardware: Hardware clocks (2) with 100 microsec. accuracy. 16 analog input var. freq. 16 bit. OR A/D- minimum of 8 channels at 20 KHz/channel -> 160 KHz min sample rate. expandable to 16 or 24 channels with DMA to disk or large RAM cache. Gain range for A/D 10v-10milliv. 8 analog output var. freq. 16 bit. 16 digital input var. freq. 16 bit. with asynchronous interrupt and "time-stamping" ( ie. storage of input word and clock time together). 8 digital output var. freq. 16 bit. positive only, or positive/negative voltages. Board to System interrupt capability. System to Board Interrupt capability. 4 programmable pulse generators. 16 TTL I/O channels. 1 Meg onboard memory. direct to disk output (DMA). streaming tape backup. onboard 20MHz. cpu (?). onboard 20MHz. math coprocessor (?). 4 Schmitt triggers. ethernet port. parallel port. SCSI port. Stepper Motor Drivers- minimum of 4 channels. * A/D and Digital input must be able to operate simultaneouly and independantly. Software: High level language interface, or Macro calls for I/O. Signal averaging & reconstruction. Field potential analysis. EEG analysis. Spike train analysis. Standard windowing system. Source code available. Export data to Macintosh and/or IBM format for typesetting /publication Spreadsheet for analysis - fast, handles data arrays larger than RAM, data format exportable to graphing and desktop publishing. Advanced graphing and analysis package - publication quality - graphs with log scales - high quality fonts - graphs with inset figures Desktop publishing system - can take above as input. -Postscript capability. ____________________________________________________________________ IBM, Masscomp, Macintosh, Unix, and X are trademarks of their respective owners, and are not to be used for signifying anything else ____________________________________________________________________ / fox@bek-mc.caltech.edu / or J. Fox / Division of Biology 216-76 / Caltech / Pasadena, CA. 91125 / (818) 356-3642/6824 hrs 8-5 M-F