From kong.gsfc.nasa.gov!arensb Mon Jul 13 07:39:58 1992 Return-Path: Received: from 128.183.115.33 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Mon, 13 Jul 92 07:38 PDT Received: from kong.gsfc.nasa.gov by lego.gsfc.nasa.gov (5.61/1.35) id AA14827; Mon, 13 Jul 92 10:36:45 -0400 Received: by kong.gsfc.nasa.gov (4.1/SMI-4.1) id AA22530; Mon, 13 Jul 92 10:36:44 EDT Date: Mon, 13 Jul 92 10:36:44 EDT From: arensb@kong.gsfc.nasa.gov (Andrew Arensburger - RMS) Message-Id: <9207131436.AA22530@kong.gsfc.nasa.gov> To: eps@reed.edu Subject: Re: Samples and Macs > From tom@dacnet.com Wed Jul 8 10:34:53 1992 > Anyway, from reading > past digests, I've seen references to transferring samples with MIDI. How is > this done? Is it like a sysex dump or something? Yes, it is possible, at least in one direction (computer->EPS). Although there is no "SysEx bulk dump" command for the EPS as there is for synths, there are various other SysEx commands that allow you to do the same thing ("Get Instrument Parameters", "Get Layer Parameters", "Get Wavesample Parameters", "Get Wavesample Data" etc). The EPS SysEx library allows you to perform all of these quite easily [*]. Unfortunately, there is one big deficiency in the EPS SysEx call set: there doesn't seem to be any way of finding out which wavesamples are in which layers. So you can create an instrument entirely via MIDI, in which case you can specify which wavesamples go in which layers. However, if you want to dissect a sound on the EPS via MIDI, you can get a list of existing layer numbers and a list of existing wavesample numbers, but you cannot (legally[%]) find out which wavesamples are in which layers. [*]: I'll release version 2.0 shortly. Really, folks! I'm really working on it! I just have to double-check for leaks and write a few examples. [%]: I have a theory that you can do some nasty things to trick the EPS into sending you this sort of "privileged information", but I haven't tried it yet. -- Andrew Arensburger | "Anyway, it never gets cold on Earth. Cold arensb@kong.gsfc.nasa.gov | enough to freeze water, yes, but not carbon ...!uunet!dftsrv!kong!arensb | dioxide." -- "Fallen Angels" From nwnexus.wa.com!sounds!brianw Mon Jul 13 10:44:46 1992 Return-Path: Received: from 192.135.191.1 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Mon, 13 Jul 92 10:44 PDT Received: by nwnexus.wa.com id AA23839 (5.65c/IDA-1.4.4 for reed.edu!eps); Mon, 13 Jul 1992 10:47:29 -0700 Received: by sounds.wa.com (NeXT-1.0 (From Sendmail 5.52)/NeXT-2.0) id AA02608; Mon, 13 Jul 92 10:30:53 PDT Date: Mon, 13 Jul 92 10:30:53 PDT From: sounds!brianw@nwnexus.wa.com (Brian Willoughby) Message-Id: <9207131730.AA02608@ sounds.wa.com > Received: by NeXT Mailer (1.63) To: eps@reed.edu (EPS Mailing List) Subject: Re: Samples and Macs | From: arensb@kong.gsfc.nasa.gov (Andrew Arensburger - RMS) | Yes, it is possible, at least in one direction (computer->EPS). I have also successfully transferred EPS samples to my Apple ][ computer, so EPS->computer is possible as well. | Unfortunately, there is one big deficiency in the EPS SysEx call | set: there doesn't seem to be any way of finding out which wavesamples are | in which layers. So you can create an instrument entirely via MIDI, in which | case you can specify which wavesamples go in which layers. However, if | you want to dissect a sound on the EPS via MIDI, you can get a list of | existing layer numbers and a list of existing wavesample numbers, but you | cannot (legally[%]) find out which wavesamples are in which layers. Take a look at the description of the "Layer Parameter Block" in the SysEx documentation. There is an array of 88 wavesample numbers called the "Layer Map" which I believe lists which wavesample is on each key for that Layer. (It also shows that an EPS Layer is limited to 88 notes instead of the full MIDI range of 128). I used the word *believe* because I remember having some trouble interpreting the values while I was writing an EPS Editor for Multimedia Windows. There was a definite correlation between wavesamples and the numbers reported by the EPS, but I can't remember if I ever figured it out. | [*]: I'll release version 2.0 shortly. Really, folks! I'm really working | on it! I just have to double-check for leaks and write a few examples. My name from dropped from the EPS mailing list years ago, so could you fill me in on the details of your EPS SysEx Library? Personal email is OK. | [%]: I have a theory that you can do some nasty things to trick the EPS | into sending you this sort of "privileged information", but I haven't tried | it yet. Just exactly what nasty things are you thinking of? Brian Willoughby BrianW@SoundS.wa.com SoundSoftware Bellevue, WA From psy.uwa.edu.au!scott Mon Jul 13 23:27:25 1992 Return-Path: Received: from 128.250.1.21 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Mon, 13 Jul 92 23:26 PDT Received: from wapsy.psy.uwa.oz.au by munnari.oz.au with SMTP (5.83--+1.3.1+0.50) id AA29111; Tue, 14 Jul 1992 16:26:05 +1000 (from scott@psy.uwa.edu.au) Received: by psy.uwa.edu.au (4.1/SMI-4.1) id AA14563; Tue, 14 Jul 92 14:24:34 WST Date: Tue, 14 Jul 92 14:24:34 WST From: scott@psy.uwa.edu.au (Scott Fisher) Message-Id: <9207140624.AA14563@psy.uwa.edu.au> To: eps@reed.edu Subject: Mac2fx, SCSI and EPS 16 Plus Hi SCSI EPS MAC'rs I had a guy just call me who is having problems getting his Mac 2 FX to communicate with his EPS 16 Plus via SCSI (SP-1) He wants to use Alchemy with his 16 Plus. He has a 44 meg removable SCSI drive and the following configurations work... Mac<--->44 drive<--->termination EPS<--->44 drive These don't work... Mac<--->44 Drive<--->EPS Mac<--->EPS So it appears to be some kind of termination problem. Any ideas? Help! :-) >From what I know (little) it looks scary to fix . Regards Scott. _______________________________________________________________________________ Scott Fisher [scott@psy.uwa.oz.au] PH: Aus [61] Perth (09) Local (380 3272). _--_|\ N Department of Psychology / \ W + E University of Western Australia. Perth --> *_.--._/ S Nedlands, 6009. PERTH, W.A. v *** ERROR 144 - REBOOT? is a registered trademark of ENSONIQ Corp *** ------------------------------------------------------------------------------- From nwnexus.wa.com!sounds!brianw Mon Jul 13 23:44:03 1992 Return-Path: Received: from 192.135.191.1 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Mon, 13 Jul 92 23:42 PDT Received: by nwnexus.wa.com id AA02712 (5.65c/IDA-1.4.4 for reed.edu!EPS); Mon, 13 Jul 1992 23:45:55 -0700 Received: by sounds.wa.com (NeXT-1.0 (From Sendmail 5.52)/NeXT-2.0) id AA04196; Mon, 13 Jul 92 23:34:06 PDT Date: Mon, 13 Jul 92 23:34:06 PDT From: sounds!brianw@nwnexus.wa.com (Brian Willoughby) Message-Id: <9207140634.AA04196@ sounds.wa.com > Received: by NeXT Mailer (1.63) To: EPS@reed.edu (EPS Mailing List) Subject: Any NeXT owners out there doing uploading/downloading of EPS samples or disks? Hello all, Since the archives have several messages from someone using an HP Unix box to read and write EPS floppies, I was wondering if it would also be possible to do this on a NeXT. I know that there is at least one other reader out there with a NeXT (not to mention altosax which IS a NeXT), so have you tried to find a way to do this? Also, where are the *sources* to epsread and epswrite? I would love to attempt porting them to the NeXT. Thanks, Brian Willoughby BrianW@SoundS.WA.com SoundSoftware Bellevue, WA USA From dacnet.com!tom Tue Jul 14 06:54:29 1992 Return-Path: Received: from 192.77.177.1 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Tue, 14 Jul 92 06:54 PDT Message-Id: Received: by stubby.dacnet.com (16.6/16.2) id AA22538; Tue, 14 Jul 92 09:53:58 -0400 From: Tom Roehl Subject: Re: Any NeXT owners out there doing... To: eps@reed.edu Date: Tue, 14 Jul 92 9:53:53 EDT X-Mailer: ELM [version 2.3 PL0] > Since the archives have several messages from someone using an HP Unix box to > read and write EPS floppies, I was wondering if it would also be possible to > do this on a NeXT. I know that there is at least one other reader out there > with a NeXT (not to mention altosax which IS a NeXT), so have you tried to > find a way to do this? > Also, where are the *sources* to epsread and epswrite? I would love to > attempt porting them to the NeXT. I would also be very interested in a UNIX version of epsread/epswrite. I work everyday on a Sun clone workstation with a 3 1/2" disk just begging to transfer samples.... :^) As far as I'm concerned, a UNIX version would be much more convenient than any kind of PC version, because then I wouldn't have to go through all kinds of contortions to get the disk images to a PC. Thanks, Tom From wri.com!andre Tue Jul 14 08:26:35 1992 Return-Path: Received: from 140.177.10.12 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Tue, 14 Jul 92 08:24 PDT Received: from rurutu.wri.com by dragonfly.wri.com with SMTP id AA20491 (5.65c/IDA-1.4.4 for ); Tue, 14 Jul 1992 10:24:50 -0500 Return-Path: Date: Tue, 14 Jul 92 10:24:49 -0500 From: andre@wri.com Message-Id: <9207141524.AA01859@rurutu.wri.com> Received: by NeXT Mailer (1.63) To: eps@reed.edu Subject: Re: Mac2fx, SCSI and EPS 16 Plus Termination seems to operate strangely on the EPS-16. I'm running an 80-meg Quantum based drive, and the only way I got it to work was to run it *without* termination. I'm using a short cable, and nothing else connected. I've heard from the techies around here that it's weird, but not unheard of. And, heck, if it works, one doesn't question it. AK From wri.com!andre Tue Jul 14 08:32:54 1992 Return-Path: Received: from 140.177.10.12 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Tue, 14 Jul 92 08:32 PDT Received: from rurutu.wri.com by dragonfly.wri.com with SMTP id AA20861 (5.65c/IDA-1.4.4 for ); Tue, 14 Jul 1992 10:32:38 -0500 Return-Path: Date: Tue, 14 Jul 92 10:32:37 -0500 From: andre@wri.com Message-Id: <9207141532.AA01891@rurutu.wri.com> Received: by NeXT Mailer (1.63) To: EPS@reed.edu Subject: Re: Any NeXT owners out there doing uploading/downloading of EPS samples or disks? Yo, I'm a NeXT user, but not much of a hacker, so I can't be of much help to re-engineer anything. I am surrounding by technical folks, however, and if I also had access to the source code, maybe I could get some help on porting the thing. If there is a Unix version out there, it could make things a lot easier. AK From cps.msu.edu!boland Tue Jul 14 09:17:04 1992 Return-Path: Received: from 35.8.56.86 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Tue, 14 Jul 92 09:13 PDT Received: from indian.cps.msu.edu by atlantic.cps.msu.edu (4.1/rpj-5.0); id AA05917; Tue, 14 Jul 92 12:13:46 EDT Received: by indian.cps.msu.edu (4.1/4.1) id AA09042; Tue, 14 Jul 92 12:13:45 EDT Date: Tue, 14 Jul 92 12:13:45 EDT From: boland@cps.msu.edu Message-Id: <9207141613.AA09042@indian.cps.msu.edu> To: eps@reed.edu Subject: Wavesample Layering & Keyboard Slits? Could some intellectual person explain the following to me? situation: I have two wavesamples (instruments) of "verbal samples". Let's say I set one voice to C4 and the other one to C6 on the other instrument. The instruments play the "verbal samples" on all keys at different pitches. What I would like to do is create a new instrument, bring only the useful parts of those instruments to the newly created instrument and have the new instrument contain the two wavesamples with only the useful keys being active. What I've tried: I created a new instrument, created two layers, copied each wavesample to a different layer. And before I copied the wavesample, I went into "Instrument Key Range" and assigned them to C4 and C6 respectively. What I ended up getting was a very large and shitty new instrument. ;-) The creator of the "George" samples was very resourceful. Somehow he has George saying various things on various keys. This is what I'm trying to do and am getting highly stressed out over. Thanks for any direction or advice!! B-) boland@cps.msu.edu P.S. I've successfully used a 15' RS-232C as an extension to the 2' SCSI cable that came with my removable drive. No Problems. From ulnar!brinkley Tue Jul 14 10:11:49 1992 Return-Path: Received: from 128.95.167.1 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Tue, 14 Jul 92 10:11 PDT Received: from ulnar.biostr.washington.edu by bashful.u.washington.edu (5.65/UW-NDC Revision: 2.22 ) id AA21300; Tue, 14 Jul 92 10:11:36 -0700 Received: by ulnar.biostr.washington.edu (NeXT-1.0 (From Sendmail 5.52)/NeXT-2.0 (w/wrb mods of 4/2/92)) id AA00349; Tue, 14 Jul 92 10:05:57 PDT Date: Tue, 14 Jul 92 10:05:57 PDT From: brinkley@ulnar.biostr.washington.edu (Jim Brinkley) Message-Id: <9207141705.AA00349@ ulnar.biostr.washington.edu > Received: by NeXT Mailer (1.63) To: sounds!brianw@nwnexus.wa.com (Brian Willoughby) Subject: Re: Any NeXT owners out there doing uploading/downloading of EPS samples or disks? Cc: brinkley@ulnar, eps@reed.edu Brian, I'm probably the other reader you mention who has a Next. I have SoftPc (PC emulator) so I was hoping to just use this with the floppy on a Nextstation, but the NeXT workspace manager needs to see a NeXT or PC formatted disk. I think the only way around this would be to write a separate device driver for the NeXT floppy but I'm not sure. I also heard something about a package called mtools that might be another way to create device drivers. Thats as far as I got with the NeXT since I don't have much time (or knowledge of device drivers). I ended up using epsread and epswrite on the PCs in the medical school library, and that worked fine, although I've only gotten around to downloading one sample. I think the source for epsread and epswrite was distributed but I'm not sure - maybe someone else knows. If I ever find time I'd really like to develop and/or use some software connecting the Next and the EPS - the market ought to be huge (2 users so far!), but it should be fun. Jim Brinkley University of Washington From fl08-g.comm.mot.com!schickda Tue Jul 14 10:21:10 1992 Return-Path: Received: from 129.188.136.100 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Tue, 14 Jul 92 10:20 PDT Received: from comm.mot.com ([145.1.3.2]) by pobox.mot.com (4.1/SMI-4.0) id AA02633; Tue, 14 Jul 92 12:31:00 CDT Received: from fl08-g.comm.mot.com (node_27d3b.comm.mot.com) by comm.mot.com (4.1/SMI-4.0) id AA13073; Tue, 14 Jul 92 12:30:26 CDT Received: by fl08-g.comm.mot.com ( 5.52 (84)/5.17) id AA18857; Tue, 14 Jul 92 13:07:59 EDT Message-Id: <9207141707.AA18857@fl08-g.comm.mot.com> From: schickda@fl08-g.comm.mot.com (David Schick) Date: Tue, 14 Jul 92 13:07:50 EDT Subject: Re: Wavesample Layering & Keyboard Slits? To: eps@reed.edu Hello:\> boland@cps.msu.edu spake un2 the masses: ~ situation: I have two wavesamples (instruments) of "verbal samples". Let's ~ say I set one voice to C4 and the other one to C6 on the other instrument. ~ The instruments play the "verbal samples" on all keys at different pitches. ~ What I would like to do is create a new instrument, bring only the useful ~ parts of those instruments to the newly created instrument and have the ~ new instrument contain the two wavesamples with only the useful keys being ~ active. Try this: Create a new instrument, then copy the wavesample from each of your two instruments 2the new instrument. Once u have both wavesamples in the new instrument, you can select each individually and edit their pitch parameters on the EDIT PITCH page. There are three parameters you'll be concerned with (although I don't remember their names). 2of them r on the same page and determine the range of keys on the keyboard which will trigger the wavesample; you can set each of these by actually using the keyboard. The other1 is on the screen 2the immediate right. This parameter effectively determines the pitch range of the wavesample. (It really sets the key at which the sample is played back at its original pitch). When I combine speech samples in this manner, I usually sample them in2 the same instrument so that I don't have 2move them around later, but there have been times when I needed to move a wavesample from one instrument to another. It seems to me that when I do this, the pitch parameters are not accurately ported 2the new instrument. Any1 else notice this? ********************************************************************** * Zoltan Android schickda@fl08-g.comm.mot.com * * * * Above the apathetic conscious memory display exists a descending * * ladder into that choking, dark experience which impales the mind * * in having yet to experience again that blow of infinity, relating * * the obtrusive brilliance of biased consequences due to all of past * * and future infinity to the true maxim of eternal compounding sense * * experience. - Farlac Gornst * ********************************************************************** From uunet.UU.NET!grebyn!brian Wed Jul 15 07:01:32 1992 Return-Path: Received: from 137.39.1.5 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Wed, 15 Jul 92 06:59 PDT Received: from uunet.uu.net (via LOCALHOST.UU.NET) by relay1.UU.NET with SMTP (5.61/UUNET-internet-primary) id AA15093; Wed, 15 Jul 92 09:59:48 -0400 Received: from grebyn.UUCP by uunet.uu.net with UUCP/RMAIL (queueing-rmail) id 095819.12052; Wed, 15 Jul 1992 09:58:19 EDT Received: by grebyn.com (5.57/smail2.3/07-01-87) id AA13015; Wed, 15 Jul 92 09:04:07 -0400 Received: by daily.grebyn.com (5.57/UUCP-Project/02.16.86-kan-10.20.91) id AA10470; Wed, 15 Jul 92 09:08:38 -0400 From: brian@grebyn.com (Brian Bishop) Message-Id: <9207151308.AA10470@daily.grebyn.com> Subject: Re: Wavesample Layering & Keyboard Slits? To: eps@reed.edu Date: Wed, 15 Jul 92 9:08:35 EDT X-Mailer: ELM [version 2.3 PL0] > From schickda@fl08-g.comm.mot.com Tue Jul 14 14:39:45 1992 > Message-Id: <9207141707.AA18857@fl08-g.comm.mot.com> > From: schickda@fl08-g.comm.mot.com (David Schick) > Date: Tue, 14 Jul 92 13:07:50 EDT > Subject: Re: Wavesample Layering & Keyboard Slits? > To: eps@reed.edu > > > Try this: > > Create a new instrument, then copy the wavesample from each of > your two instruments 2the new instrument. Once u have both > wavesamples in the new instrument, you can select each > individually and edit their pitch parameters on the > EDIT PITCH page. There are three parameters you'll be concerned > with (although I don't remember their names). I think I do! > 2of them r on > the same page and determine the range of keys on the keyboard > which will trigger the wavesample; you can set each of these by > actually using the keyboard. I think these are RANGE LO and RANGE HI. > The other1 is on the screen 2the > immediate right. This parameter effectively determines the pitch > range of the wavesample. (It really sets the key at which the > sample is played back at its original pitch). That's the ROOT key. P.S. Anybody know of an EPS-16+ command page chart available somewhere? It would really help... From psy.uwa.edu.au!scott Wed Jul 15 18:24:51 1992 Return-Path: Received: from 128.250.1.21 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Wed, 15 Jul 92 18:24 PDT Received: from wapsy.psy.uwa.oz.au by munnari.oz.au with SMTP (5.83--+1.3.1+0.50) id AA08064; Thu, 16 Jul 1992 11:24:03 +1000 (from scott@psy.uwa.edu.au) Received: by psy.uwa.edu.au (4.1/SMI-4.1) id AA16657; Thu, 16 Jul 92 09:22:33 WST Date: Thu, 16 Jul 92 09:22:33 WST From: scott@psy.uwa.edu.au (Scott Fisher) Message-Id: <9207160122.AA16657@psy.uwa.edu.au> To: eps@reed.edu Subject: EPS/EPS16+ Command chart. >From grebyn!brian@uunet.UU.NET Wed Jul 15 22:33:59 1992 >Subject: Re: Wavesample Layering & Keyboard Slits? > > P.S. Anybody know of an EPS-16+ command page chart available somewhere? > It would really help... I saw an add in the "Classifieds" of the hacker... "EPS/EPS16+ Navigator: Want all EPS Direct-Dial commands at your fingertips? How about command sequences (mini-tutorials) dealing with sampling, editing effects, MIDI/Multitimbral use, sequencing, song construction and more? The Ultimate EPS Cheatsheet! Super-condensed, logically organized information on a double-sided, plastic covered quick reference sheet. Send $6.95 plus $1 S/H ($2 outside the US) to: Keith Perterson, 711 Park Ave, Dunkirk, NY 14048." Warning: I am just copying this from the June 1992 (#84) Transoniq Hacker and as such deal at your own risk. Regards Scott. _______________________________________________________________________________ Scott Fisher [scott@psy.uwa.oz.au] PH: Aus [61] Perth (09) Local (380 3272). _--_|\ N Department of Psychology / \ W + E University of Western Australia. Perth --> *_.--._/ S Nedlands, 6009. PERTH, W.A. v *** ERROR 144 - REBOOT? is a registered trademark of ENSONIQ Corp *** ------------------------------------------------------------------------------- From csrd.uiuc.edu!bordner Wed Jul 15 20:20:09 1992 Return-Path: Received: from 128.174.162.46 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Wed, 15 Jul 92 20:19 PDT Received: from sp1.csrd.uiuc.edu by s46.csrd.uiuc.edu with SMTP id AA19305 (5.65c/IDA-1.4.4 for ); Wed, 15 Jul 1992 22:19:51 -0500 Received: by sp1.csrd.uiuc.edu (4.1/SMI-4.1) id AA07245; Wed, 15 Jul 92 22:19:51 CDT Date: Wed, 15 Jul 92 22:19:51 CDT From: bordner@csrd.uiuc.edu (James Bordner) Message-Id: <9207160319.AA07245@sp1.csrd.uiuc.edu> To: eps@reed.edu Subject: Re: Wavesample Layering & Keyboard Slits? brian@grebyn.com (Brian Bishop) writes: > P.S. Anybody know of an EPS-16+ command page chart available somewhere? > It would really help... Well, I made one awhile back for my EPS-- maybe someone can use it as a template for making one for the EPS-16+ as well and send it to the list. P.S. Hope using more than 80 columns doesn't cause too many people too many problems. It's 105 columns wide, so adjust your printer accordingly ;-) P.P.S. I made this using EPS O.S. version 2.40. Anyone know if the commands are any different from the newest O.S. version? If so, maybe someone can use this as a template and send it to the list... :-) P.P.P.S If anyone finds any mistakes, please let me know. --james --- snip --- snip --- snip --- snip --- snip --- snip --- snip --- snip --- ------------ EPS Commands ------------ Instrument Env 1 Env 2 Env 2 +-------------------------+-------------------------+-------------------------+-------------------------+ | 0 create new instrument | 0 no commands on page | 0 no commands on page | 0 no commands on page | | 1 copy instrument | 1 calibrate keyboard | | | | 2 delete instrument | 2 software information | | | | 3 save instrument | 3 examine dos status | | | | 4 save bank | 4 examine analog inputs | | | | 5 create preset | 5 msb adjustment | | | | | 6 dc offset adjustment | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | +-------------------------+-------------------------+-------------------------+-------------------------+ Seq.Song Pitch Filter Amp +-------------------------+-------------------------+-------------------------+-------------------------+ | 0 create new sequence | 0 edit pitch table | 0 no commands on page | 0 normalize gain | | 1 copy sequence | 1 copy pitch table | | 1 volume smoothing | | 2 delete sequence | 2 delete pitch table | | 2 mix wavesamples | | 3 save current sequence | 3 extrapolate pitch tabl| | 3 merge wavesamples | | 4 save song + all seqs | | | 4 splice wavesamples | | 5 rename song/sequence | | | 5 fade in | | 6 sequencer information | | | 6 fade out | | 7 erase song + all seqs | | | | | 8 append sequence | | | | | 9 change sequence length| | | | |10 edit song steps | | | | | | | | | | | | | | | | | | | +-------------------------+-------------------------+-------------------------+-------------------------+ Midi LFO Wave Layer +-------------------------+-------------------------+-------------------------+-------------------------+ | 0 midi sys-ex recorder | 0 clear data | 0 Create new wavesample | 0 create new layer | | | 1 copy data | 1 copy wavesample | 1 copy layer | | | 2 replicate data | 2 delete wavesample | 2 delete layer | | | 3 reverse data | 3 wavesample information| | | | 4 invert data | 4 truncate wavesample | | | | 5 add data | 5 cross fade loop | | | | 6 scale data | 6 reverse cross fade | | | | | 7 ensemble cross fade | | | | | 8 bowtie cross fade loop| | | | | 9 bidirectional x-fade | | | | |10 make loop longer | | | | |11 synthesized loop | | | | |12 convert sample rate | | | | |13 copy wave parameters | | +-------------------------+-------------------------+-------------------------+-------------------------+ System Track +-------------------------+ +-------------------------+ | 0 load mirage-dsk sound | | 0 quantize track | | 1 set number of voices | | 1 copy track | | 2 format floppy drive | | 2 erase/undefine track | | 3 copy OS to disk | | 3 erase controller | | 4 save global parameters| | 4 erase key pressure | | 5 load global parameters| | 5 erase key range | | 6 create directory | | 6 erase program changes | | 7 change storage device | | 7 merge two track | | 8 save macro file | | 8 event edit track | | 9 format SCSI drive | | 9 transpose track | |10 copy/backup/restore | |10 scale controller | | | |11 shift track by clocks | | | | | | | | | +-------------------------+ +-------------------------+ From mail.uunet.ca!web.apc.org!becker!bdb Wed Jul 15 20:27:00 1992 Return-Path: <@mail.uunet.ca:becker!bdb@web.apc.org> Received: from 137.39.1.5 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Wed, 15 Jul 92 20:26 PDT Received: from mail.uunet.ca (via uunet.ca) by relay1.UU.NET with SMTP (5.61/UUNET-internet-primary) id AA05590; Wed, 15 Jul 92 23:26:35 -0400 Received: from web.apc.org ([142.77.253.8]) by mail.uunet.ca with SMTP id <10347>; Wed, 15 Jul 1992 23:26:09 -0400 Received: by web.apc.org (/\==/\ Smail3.1.25.1 #25.5) id ; Thu, 16 Jul 92 03:24 GMT Received: by becker.GTS.ORG (smail2.5/bdb-15Aug90) id AA21431; Wed, 15 Jul 92 21:35:17 EDT (-0400) Newsgroups: eps Path: bdb From: bdb@becker.GTS.ORG (Bruce Becker) Subject: Re: Samples and Macs Message-Id: <1992Jul16.013330.21367@becker.GTS.ORG> Organization: G. T. S., Toronto, Ontario References: <9207131436.AA22530@kong.gsfc.nasa.gov> Date: Wed, 15 Jul 1992 21:33:30 -0400 To: eps@reed.edu In article <9207131436.AA22530@kong.gsfc.nasa.gov> arensb@kong.gsfc.nasa.gov (Andrew Arensburger - RMS) writes: || From tom@dacnet.com Wed Jul 8 10:34:53 1992 || Anyway, from reading || past digests, I've seen references to transferring samples with MIDI. How is || this done? Is it like a sysex dump or something? | | Yes, it is possible, at least in one direction (computer->EPS). |Although there is no "SysEx bulk dump" command for the EPS as there is for |synths, there are various other SysEx commands that allow you to do the |same thing ("Get Instrument Parameters", "Get Layer Parameters", "Get |Wavesample Parameters", "Get Wavesample Data" etc). The EPS SysEx library |allows you to perform all of these quite easily [*]. How much of this will work on th EPS-16?? -- ,u, Bruce Becker Toronto, Ontario a /i/ Internet: bdb@becker.gts.org Uucp: ...!web!becker!bdb `\o\-e \*/*\*/*\*/*\*/*\*/*\*/*\*/*\*/*\*/*\*/*\*/*\*/*\*/*\*/ _< /_ "Awaken the Geek Within" - Anthony Robot From psy.uwa.edu.au!scott Wed Jul 15 22:08:42 1992 Return-Path: Received: from 128.250.1.21 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Wed, 15 Jul 92 22:07 PDT Received: from wapsy.psy.uwa.oz.au by munnari.oz.au with SMTP (5.83--+1.3.1+0.50) id AA14992; Thu, 16 Jul 1992 15:07:13 +1000 (from scott@psy.uwa.edu.au) Received: by psy.uwa.edu.au (4.1/SMI-4.1) id AA17682; Thu, 16 Jul 92 13:05:36 WST Date: Thu, 16 Jul 92 13:05:36 WST From: scott@psy.uwa.edu.au (Scott Fisher) Message-Id: <9207160505.AA17682@psy.uwa.edu.au> To: eps@reed.edu Subject: RUMOUR 8 MEG OK Rumour-mongers :-).... Here's the latest as quoted from the Chicken Chronicle, Dated Vol 2 issue 1, 1992 "So the hot issue is expanded memory for BOTH the EPS and 16 Plus. We heard from a reliable undisclosed source (in Canada) that someone has solved the two major barriers to producing an expansion board that exceeds the present 2 meg limit; He has cracked the OS and altered the chip addresses and has found a way to power the SIMM board as well. Well the story is noww that he has produced a couple but is now not going to produce them for the public because of another rumor: ENSONIQ will introduce some 8 Meg proprietary SIMM's borards at the end of summer. Sound interesting?" Stay tuned for more rumours... Regards Scott. _______________________________________________________________________________ Scott Fisher [scott@psy.uwa.oz.au] PH: Aus [61] Perth (09) Local (380 3272). _--_|\ N Department of Psychology / \ W + E University of Western Australia. Perth --> *_.--._/ S Nedlands, 6009. PERTH, W.A. v *** ERROR 144 - REBOOT? is a registered trademark of ENSONIQ Corp *** ------------------------------------------------------------------------------- From ads.com!pdel Thu Jul 16 00:23:11 1992 Return-Path: Received: from 128.229.30.16 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Thu, 16 Jul 92 00:22 PDT Received: from bert.ads.com by ads.com (5.65+/1.34v1.3) id AA05862; Thu, 16 Jul 92 00:24:35 -0700 From: pdel@ads.com (Peter Delevoryas) Received: by bert.ads.com (5.65+/4.7) id AA04275; Thu, 16 Jul 92 00:24:29 -0700 Message-Id: <9207160724.AA04275@bert.ads.com> Subject: Re: RUMOUR 8 MEG To: scott@psy.uwa.edu.au (Scott Fisher) Date: Thu, 16 Jul 92 0:24:28 PDT Cc: eps@reed.edu In-Reply-To: <9207160505.AA17682@psy.uwa.edu.au>; from "Scott Fisher" at Jul 16, 92 1:05 pm X-Mailer: ELM [version 2.3 PL11] > "So the hot issue is expanded memory for BOTH the EPS and 16 Plus. We > ENSONIQ will introduce some 8 Meg proprietary SIMM's borards at the end of > summer. Drool DroolDroolDroolDroolDroolDroolDroolDroolDroolDroolDroolDroolDroolDroolDr oolDroolDroolDroolDroolDroolDroolDroolDroolDroolDroolDroolDroolDroolDroolDrool > Sound interesting?" Drool From nwnexus.wa.com!sounds!brianw Thu Jul 16 01:47:35 1992 Return-Path: Received: from 192.135.191.1 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Thu, 16 Jul 92 01:47 PDT Received: by nwnexus.wa.com id AA15993 (5.65c/IDA-1.4.4 for reed.edu!eps); Thu, 16 Jul 1992 01:49:54 -0700 Received: by sounds.wa.com (NeXT-1.0 (From Sendmail 5.52)/NeXT-2.0) id AA04087; Thu, 16 Jul 92 00:56:49 PDT Date: Thu, 16 Jul 92 00:56:49 PDT From: sounds!brianw@nwnexus.wa.com (Brian Willoughby) Message-Id: <9207160756.AA04087@ sounds.wa.com > Received: by NeXT Mailer (1.63) To: eps@reed.edu (EPS Mailing List) Subject: Re: RUMOUR 8 MEG So how can it be possible that you can actually play multiple instruments or wavesamples from an address space larger than 2 Mbytes when I've always understood that the Ensoniq DOC II (Digital Oscillator Chip) only generates 21 address bits? I'm sure that someone could rewrite the Motorola 68000 based OS to support more than 2 Meg, since it has a 16 Mbyte address space - and I'm equally sure that someone could also power the additional RAM - but this rumor sounds like pure wishful thinking. The heart of the EPS - which makes it so unique compared to all of the other samplers and computers out there - is the Ensoniq DOC II which handles transferring 20 digital wavesamples at up to 50 Ksamples per second as well as 40 digital filters. This chip and the socket it fits into are most likely limited to a 2 Mbytes addressing space, and the RAM expansion card most certainly is limited to 10 address bits (10 Row bits plus 10 Column bits equals 1 Msample of memory). Unless someone had the resources to engineer another custom chip of this complexity, I doubt that any of us (EPS and EPS 16+ owners alike) will see anything that allows access to more wavesample memory. HOWEVER: there is always the possibility of a RAM-based floppy or hard disk replacement for loading sounds - much like the Ensoniq FLASH cards. But don't be fooled - you'll never load a 3 Mbyte killer piano multisample into your EPS* I've never really understood all the fuss about the Flash cards, since they are nothing more than a really fast drive. My Quantum P105S is fast enough for me and does not suffer from multiple writes (other than fragmentation). Later folks, and I wouldn't mind if my comments start some discussion on the internals of the EPS series. Its been far too long since Dick Lord discussed what goes on "Inside the EPS" way back in issue #39 of the Transoniq Hacker from September 1988 ! Brian Willoughby SoundSoftware Bellevue, WA USA | OK Rumour-mongers :-).... Here's the latest as quoted from the | Chicken Chronicle, | | Dated Vol 2 issue 1, 1992 | | "So the hot issue is expanded memory for BOTH the EPS and 16 Plus. We | heard from a reliable undisclosed source (in Canada) that someone has | solved the two major barriers to producing an expansion board that | exceeds the present 2 meg limit; He has cracked the OS and altered the | chip addresses and has found a way to power the SIMM board as well. | Well the story is noww that he has produced a couple but is now not | going to produce them for the public because of another rumor: ENSONIQ | will introduce some 8 Meg proprietary SIMM's borards at the end of summer. | Sound interesting?" | | Stay tuned for more rumours... From sndcrft.dialix.oz.au!steveq Thu Jul 16 03:49:58 1992 Return-Path: Received: from 128.250.1.21 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Thu, 16 Jul 92 03:47 PDT Received: from uniwa.uwa.edu.au by munnari.oz.au with SMTP (5.83--+1.3.1+0.50) id AA24081; Thu, 16 Jul 1992 20:44:23 +1000 (from steveq@sndcrft.dialix.oz.au) Received: by uniwa.uwa.edu.au (5.65c) id AA25985; Thu, 16 Jul 1992 18:44:05 +0800 Received: from sndcrft.DIALix.oz.au by DIALix.oz.au id aa12139; 16 Jul 92 18:27 WST Received: by sndcrft.DIALix.oz.au (HERMES RMAIL 1.00 Rev. Jan 16 1992) id <0brgcra@sndcrft.DIALix.oz.au>; 15 Jul 92 23:49 MET From: steveq@sndcrft.dialix.oz.au (Steve Quartly) Message-Id: Organization: Sound Craft Creative Music Subject: Off for a while To: eps@reed.edu Reply-To: steveq@sndcrft.dialix.oz.au X-Software: HERMES GUS 1.00 Rev. Jan 16 1992 Date: Wed, 15 Jul 1992 23:44:50 MET Hi Gang, Just a quick note, Im off to Barcelona for the Olympics as of 1700hrs tomorrow night, so this is the last youll see of me for the next 3 weeks. Any queries regarding EPS Disk Wizard v2.1 will be answered in about 3 weeks. Have fun, see you soon.......... OLE! -- <><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> S t e v e Q u a r t l y, P e r t h W e s t e r n A u s t r a l i a, _--_|\ N PH: Aus [61] Perth (09) Local (309 4445). / \ W + E Perth --> *_.--._/ S 43rd Law of Computing: Anything tha can go wr v error: Segmentation violation -- Core dumped. steveq@sndcrft.DIALix.oz.au <><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> From kong.gsfc.nasa.gov!arensb Thu Jul 16 06:24:10 1992 Return-Path: Received: from 128.183.115.33 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Thu, 16 Jul 92 06:23 PDT Received: from kong.gsfc.nasa.gov by lego.gsfc.nasa.gov (5.61/1.35) id AA02512; Thu, 16 Jul 92 09:23:10 -0400 Received: by kong.gsfc.nasa.gov (4.1/SMI-4.1) id AA04491; Thu, 16 Jul 92 09:23:12 EDT Date: Thu, 16 Jul 92 09:23:12 EDT From: arensb@kong.gsfc.nasa.gov (Andrew Arensburger - RMS) Message-Id: <9207161323.AA04491@kong.gsfc.nasa.gov> To: eps@reed.edu Subject: Re: Samples and Macs > From @mail.uunet.ca:becker!bdb@web.apc.org Wed Jul 15 23:44:55 1992 > In article <9207131436.AA22530@kong.gsfc.nasa.gov> arensb@kong.gsfc.nasa.gov (Andrew Arensburger - RMS) writes: > | Yes, it is possible, at least in one direction (computer->EPS). > |Although there is no "SysEx bulk dump" command for the EPS as there is for > |synths, there are various other SysEx commands that allow you to do the > |same thing ("Get Instrument Parameters", "Get Layer Parameters", "Get > |Wavesample Parameters", "Get Wavesample Data" etc). The EPS SysEx library > |allows you to perform all of these quite easily [*]. > > How much of this will work on th EPS-16?? All of the above, allowing for the fact that I don't have a 16+ to test it on. The EPS 16+ uses the same external command set as the Classic, with some modifications (e.g., different parameter numbers) and additional commands (e.g., "Save Instrument To Disk"). The data structures are of the same size for both instruments; the 16+ has one additional data structure type (effect). In those cases where the 16+ has more fields than the Classic (e.g., "Get Layer Parameters"), it uses holes in the data structure: officially-unused words, low words (which are normally 0), and so forth. Version 2.0 of the EPS SysEx library knows about all this stuff, and will do the Right Thing, especially if it is told that it is talking to a 16+ rather than a Classic. As far as a Bulk Dump utility is concerned, the one thing that I haven't managed to do via MIDI is get the list of all wavesamples inside a layer. /AA/ From wri.com!andre Thu Jul 16 10:02:05 1992 Return-Path: Received: from 140.177.10.12 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Thu, 16 Jul 92 10:00 PDT Received: from rurutu.wri.com by dragonfly.wri.com with SMTP id AA22887 (5.65c/IDA-1.4.4 for ); Thu, 16 Jul 1992 12:00:26 -0500 Return-Path: Date: Thu, 16 Jul 92 12:00:25 -0500 From: andre@wri.com Message-Id: <9207161700.AA01592@rurutu.wri.com> Received: by NeXT Mailer (1.63) To: eps@reed.edu (EPS Mailing List) Subject: Re: RUMOUR 8 MEG > Date: Thu, 16 Jul 92 00:56:49 PDT > From: sounds!brianw@nwnexus.wa.com (Brian Willoughby) > To: eps@reed.edu (EPS Mailing List) > Subject: Re: RUMOUR 8 MEG > I've never really understood all the fuss about the Flash cards, > since they are nothing more than a really fast drive. My Quantum > P105S is fast enough for me and does not suffer from multiple > writes (other than fragmentation). Gee, I thought the FlashBank was supposed to be like extra memory, i.e., expands upon the 2 megs to add another 512 or more to the avaiable RAM. Is this not true? Does the Flash really only store stuff, and then load it into available RAM? AK From t-bone.cs.unlv.edu!dflint Thu Jul 16 10:49:57 1992 Return-Path: Received: from 131.216.16.2 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Thu, 16 Jul 92 10:48 PDT Message-Id: Received: from t-bone.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id aa13750; 16 Jul 92 10:45 PDT To: eps@reed.edu Subject: Re: RUMOUR 8 MEG In-reply-to: Your message of "Thu, 16 Jul 92 00:56:49 PDT." <9207160756.AA04087@ sounds.wa.com > Date: Thu, 16 Jul 92 10:45:59 -0700 From: dflint@t-bone.cs.unlv.edu >So how can it be possible that you can actually play multiple instruments or >wavesamples from an address space larger than 2 Mbytes when I've always understood >that the Ensoniq DOC II (Digital Oscillator Chip) only generates 21 address bits? Perhaps so, but (at least on the original EPS) if you add the 4X memory expander (2M of memory) you still have the original 512K for sequencer memory-- obviously, the 68000 does have a hand in memory management. >I'm sure that someone could rewrite the Motorola 68000 based OS to support more >than 2 Meg, since it has a 16 Mbyte address space - and I'm equally sure that >someone could also power the additional RAM - but this rumor sounds like pure >wishful thinking. >The heart of the EPS - which makes it so unique compared to all of the other >samplers and computers out there - is the Ensoniq DOC II which handles transferring >20 digital wavesamples at up to 50 Ksamples per second as well as 40 digital >filters. This chip and the socket it fits into are most likely limited to a 2 >Mbytes addressing space, and the RAM expansion card most certainly is limited to 10 >address bits (10 Row bits plus 10 Column bits equals 1 Msample of memory). Unless >someone had the resources to engineer another custom chip of this complexity, I >doubt that any of us (EPS and EPS 16+ owners alike) will see anything that allows >access to more wavesample memory. Perhaps so. Unfortunately, I don't know enough about the EPS internals to say one way or the other if this is a limiting factor. BUT, a few years ago, at least one company who made 4X expanders for the EPS promoted thier memory board as being expandable to an 8X when Ensoniq re-wrote the OS to accomadate it. Since the DOC II is limited to 21 address lines, it might be possible for the 68000 to bank switch the memory for the DOC, a la the old apple II. This would effectively give the DOC an address space as large as the memory space of the 68000, assuming all 24 address lines for the 68000 are in place. (Jeez, here I thought that the apple II was dead, and segmented memory in IBM PC's was on its way out-- will us poor computer programmers ever be saved from the horror of segmented memory? :-) >HOWEVER: there is always the possibility of a RAM-based floppy or hard disk >replacement for loading sounds - much like the Ensoniq FLASH cards. But don't be >fooled - you'll never load a 3 Mbyte killer piano multisample into your EPS* I'm not so sure about this-- perhaps individual wave samples would be limited to 2M, but why not a >2M instrument? >I've never really understood all the fuss about the Flash cards, since they are >nothing more than a really fast drive. My Quantum P105S is fast enough for me and >does not suffer from multiple writes (other than fragmentation). >Later folks, and I wouldn't mind if my comments start some discussion on the >internals of the EPS series. Its been far too long since Dick Lord discussed what >goes on "Inside the EPS" way back in issue #39 of the Transoniq Hacker from >September 1988 ! >Brian Willoughby >SoundSoftware >Bellevue, WA >USA Mike Evans dflint@unlv.edu From nwnexus.wa.com!sounds!brianw Thu Jul 16 12:44:22 1992 Return-Path: Received: from 192.135.191.1 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Thu, 16 Jul 92 12:43 PDT Received: by nwnexus.wa.com id AA21595 (5.65c/IDA-1.4.4 for reed.edu!eps); Thu, 16 Jul 1992 12:46:00 -0700 Received: by sounds.wa.com (NeXT-1.0 (From Sendmail 5.52)/NeXT-2.0) id AA05468; Thu, 16 Jul 92 12:17:32 PDT Date: Thu, 16 Jul 92 12:17:32 PDT From: sounds!brianw@nwnexus.wa.com (Brian Willoughby) Message-Id: <9207161917.AA05468@ sounds.wa.com > Received: by NeXT Mailer (1.63) To: eps@reed.edu (EPS Mailing List) Subject: Re: RUMOUR 8 MEG Thanks for the ideas, Mike: | >So how can it be possible that you can actually play multiple instruments or | >wavesamples from an address space larger than 2 Mbytes when I've always | >understood that the Ensoniq DOC II (Digital Oscillator Chip) only generates | >21 address bits? | | Perhaps so, but (at least on the original EPS) if you add the | 4X memory expander (2M of memory) you still have the original 512K | for sequencer memory-- obviously, the 68000 does have a hand in memory | management. You're exactly right, Mike, but that is precisely my point. You can add memory for other functions like sequencing or off-line storage, but never more than 1 Msample for playback of waves. The 68000 does not need to use the DOC II for sequencing, it might as well be running in a separate machine. See my original comment below about the 68000. | >I'm sure that someone could rewrite the Motorola 68000 based OS to support | >more than 2 Meg, since it has a 16 Mbyte address space - and I'm equally sure | >that someone could also power the additional RAM - but this rumor sounds like | >pure wishful thinking. | | >The heart of the EPS - which makes it so unique compared to all of the other | >samplers and computers out there - is the Ensoniq DOC II which handles | >transferring 20 digital wavesamples at up to 50 Ksamples per second as well | >as 40 digital filters. This chip and the socket it fits into are most likely | >limited to a 2 Mbyte addressing space, and the RAM expansion card most | >certainly is limited to 10 address bits (10 Row bits plus 10 Column bits | >equals 1 Msample of memory). Unless someone had the resources to engineer | >another custom chip of this complexity, I doubt that any of us (EPS and EPS | >16+ owners alike) will see anything that allows access to more wavesample | >memory. | | Perhaps so. Unfortunately, I don't know enough about the EPS | internals to say one way or the other if this is a limiting factor. | BUT, a few years ago, at least one company who made 4X expanders for | the EPS promoted thier memory board as being expandable to an 8X when | Ensoniq re-wrote the OS to accomadate it. Perhaps they were going to organize the extra memory like a drive. Perhaps they did not know very much about the EPS and thought they could do more than they actually could. They never did deliver on their promise. Perhap it was a total marketing lie in an attempt to make their product sound more attractive than the other 4X expanders. | Since the DOC II is limited to 21 address lines, it might be | possible for the 68000 to bank switch the memory for the DOC, a la the | old apple II. This would effectively give the DOC an address space as | large as the memory space of the 68000, assuming all 24 address lines | for the 68000 are in place. (Jeez, here I thought that the apple II | was dead, and segmented memory in IBM PC's was on its way out-- will | us poor computer programmers ever be saved from the horror of | segmented memory? :-) Mike, this is a very reasonable idea for some other single processor systems, but it wouldn't work for the DOC II and 68000 combination. The key factor is that there is no way to get *inside* the DOC II to make these enhancements. With the Apple ][, there was only one processor and the memory. Apple had complete control over both the OS that the Apple ][ was running and the memory hardware. By keeping the OS in non-bank-switched memory, it could execute commands to temporarily substitute additional memory for existing memory. However, some of the original limitations were still there. For example, a program larger than 64K could not be run unless its internals were rewritten to access the bank-switch hardware. How would you "rewrite" the internals of the DOC II to accomodate bank-switched memory? The segmented PC is based on Intel processors, which are an example of the chip manufacturer changing the internal signals to accomodate extra memory. You would need to first be able to build a DOC II before you could add segment registers to its architecture. The 68000 is completely separate from the DOC II, and quite powerless to do 20 voice sampling without the DOC II. While the DOC II is mixing 20 wavesamples at 50 kHz and computing second order interpolation between adjacent samples in memory, there is absolutely no way that the 68000 could predict the exact moment to switch banks for an individual wavesample access. The result is that you could never hear multiple wavesamples stored in different banks. Flipping the bank switch would cause any wavesamples which were currently playing to lose access to their memory. The sounds resulting from accessing the alternate banks would be quite unmusical unless you waited for the EPS to be completely silent before flipping the bank switch. I just had an idea. You could add another DOC II on a daughter board, and rewrite the EPS OS to control two separate DOCs, each with its own 2 Msample memory. You would then need to mix the analog outputs of each DOC II. The disadvantage would be that the current EPS uses digital mixing for all of its 20 waves before conversion to analog, while the dual DOC II system would have to use the noisier method of analog mixing. Any thoughts about this idea? | >HOWEVER: there is always the possibility of a RAM-based floppy or hard disk | >replacement for loading sounds - much like the Ensoniq FLASH cards. But | >don't be fooled - you'll never load a 3 Mbyte killer piano multisample into | >your EPS* | | I'm not so sure about this-- perhaps individual wave samples would be | limited to 2M, but why not a >2M instrument? See my comments above. You could not hear wavesamples from different banks at the same time. If you want a polyphonic EPS, you can't have more than 2 Msample of memory without buying a different machine. | Mike Evans | dflint@unlv.edu Thanks for showing interest, and I hope I didn't sound critical. There are some good ideas floating around which might as well be discussed since this is th EPS group! Brian Willoughby SoundSoftware Bellevue, WA USA From RCNVMS.RCN.MASS.EDU!NLEONARD Thu Jul 16 13:39:36 1992 Return-Path: Received: from 134.241.10.5 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Thu, 16 Jul 92 13:39 PDT Received: from RCNVMS.RCN.MASS.EDU by RCNVMS.RCN.MASS.EDU (PMDF #12408) id <01GMGBQMCMTSDU7DFY@RCNVMS.RCN.MASS.EDU>; Thu, 16 Jul 1992 16:41 EST Date: Thu, 16 Jul 1992 16:41 EST From: NLEONARD@RCNVMS.RCN.MASS.EDU Subject: Hacker #39 To: eps@reed.edu Message-id: <01GMGBQMCMTSDU7DFY@RCNVMS.RCN.MASS.EDU> X-VMS-To: IN%"eps@reed.edu" What is the jist of the article on the EPS architecture. Can anyone post the contents of this article of send me a photo copy? I'll send a self address stamped envelope. Neil Leonard nleonard@rcnvms.rcn.mass.edu From t-bone.cs.unlv.edu!dflint Thu Jul 16 14:22:03 1992 Return-Path: Received: from 131.216.16.2 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Thu, 16 Jul 92 14:20 PDT Message-Id: Received: from t-bone.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id aa20397; 16 Jul 92 14:12 PDT To: eps@reed.edu Subject: Re: RUMOUR 8 MEG In-reply-to: Your message of "Thu, 16 Jul 92 12:17:32 PDT." <9207161917.AA05468@ sounds.wa.com > Date: Thu, 16 Jul 92 14:12:28 -0700 From: dflint@t-bone.cs.unlv.edu >| Perhaps so. Unfortunately, I don't know enough about the EPS >| internals to say one way or the other if this is a limiting factor. >| BUT, a few years ago, at least one company who made 4X expanders for >| the EPS promoted thier memory board as being expandable to an 8X when >| Ensoniq re-wrote the OS to accomadate it. >Perhaps they were going to organize the extra memory like a drive. Perhaps they >did not know very much about the EPS and thought they could do more than they >actually could. They never did deliver on their promise. Perhap it was a total >marketing lie in an attempt to make their product sound more attractive than the >other 4X expanders. Actually, they made it sound like everything depended on Ensoniq simply rewriting the OS to accomodate the extra memory. Personally, I don't like the idea of a RAM drive, UNLESS the memory was backed up by a battery, or was EEPROM (or at least wouldn't be lost when the power was shut off). Even then, Considering how long it takes to load a sound from a hard disk, the advantages would be debatable. (Even my slow 45M Syquest can load a huge piano sample in about 10 sec.) >| Since the DOC II is limited to 21 address lines, it might be >| possible for the 68000 to bank switch the memory for the DOC, a la the >| old apple II. This would effectively give the DOC an address space as >| large as the memory space of the 68000, assuming all 24 address lines >| for the 68000 are in place. (Jeez, here I thought that the apple II >| was dead, and segmented memory in IBM PC's was on its way out-- will >| us poor computer programmers ever be saved from the horror of >| segmented memory? :-) >Mike, this is a very reasonable idea for some other single processor systems, >but it wouldn't work for the DOC II and 68000 combination. The key factor is >that there is no way to get *inside* the DOC II to make these enhancements. >With the Apple ][, there was only one processor and the memory. Apple had >complete control over both the OS that the Apple ][ was running and the memory >hardware. By keeping the OS in non-bank-switched memory, it could execute >commands to temporarily substitute additional memory for existing memory. >However, some of the original limitations were still there. For example, a >program larger than 64K could not be run unless its internals were rewritten to >access the bank-switch hardware. How would you "rewrite" the internals of the >DOC II to accomodate bank-switched memory? Hmmmm...... It seems that I made an incorrect assumption about the internal layout of the EPS. I assumed that the 68000 did all the memory management for the DOC. (which, now that I think about it, doesn't make a whole lot of sense, since the DOC must be able to access large chunks of memory, not just accept data from the 68000). [more explanations about segmented memory deleted] >I just had an idea. You could add another DOC II on a daughter board, and >rewrite the EPS OS to control two separate DOCs, each with its own 2 Msample >memory. You would then need to mix the analog outputs of each DOC II. The >disadvantage would be that the current EPS uses digital mixing for all of its 20 >waves before conversion to analog, while the dual DOC II system would have to >use the noisier method of analog mixing. Any thoughts about this idea? Interesting idea-- but I think it would make the machine much to complicated (perhaps not to an end user, but electrically). Besides, it would probably be easier (and cleaner) to just buy a second EPS and get a MIDI-controllable automated mixer. :-) [stuff deleted about >2M instruments] >See my comments above. You could not hear wavesamples from different banks at >the same time. If you want a polyphonic EPS, you can't have more than 2 Msample >of memory without buying a different machine. Speaking of a different machine, does the 16+ have the same memory limitation? If so, is it for the same reason, or did Ensoniq design it to have a larger address space, and just doesn't support it yet in the OS? >Brian Willoughby >SoundSoftware >Bellevue, WA >USA Mike Evans dflint@unlv.edu From nwnexus.wa.com!sounds!brianw Thu Jul 16 16:44:39 1992 Return-Path: Received: from 192.135.191.1 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Thu, 16 Jul 92 16:42 PDT Received: by nwnexus.wa.com id AA24703 (5.65c/IDA-1.4.4 for reed.edu!EPS); Thu, 16 Jul 1992 16:45:42 -0700 Received: by sounds.wa.com (NeXT-1.0 (From Sendmail 5.52)/NeXT-2.0) id AA07793; Thu, 16 Jul 92 16:27:27 PDT Date: Thu, 16 Jul 92 16:27:27 PDT From: sounds!brianw@nwnexus.wa.com (Brian Willoughby) Message-Id: <9207162327.AA07793@ sounds.wa.com > Received: by NeXT Mailer (1.63) To: EPS@reed.edu (EPS Mailing List) Subject: Re: RUMOUR 8 MEG The continuing discussion... | >| BUT, a few years ago, at least one company who made 4X expanders for | >| the EPS promoted thier memory board as being expandable to an 8X when | >| Ensoniq re-wrote the OS to accomadate it. | | >Perhaps they were going to organize the extra memory like a drive. Perhaps | >they did not know very much about the EPS and thought they could do more than | >they actually could. They never did deliver on their promise. Perhaps it | >was a total marketing lie in an attempt to make their product sound more | >attractive than the other 4X expanders. | | Actually, they made it sound like everything depended on Ensoniq | simply rewriting the OS to accomodate the extra memory. Personally, I | don't like the idea of a RAM drive, UNLESS the memory was backed up by | a battery, or was EEPROM (or at least wouldn't be lost when the power | was shut off). Even then, Considering how long it takes to | load a sound from a hard disk, the advantages would be debatable. | (Even my slow 45M Syquest can load a huge piano sample in about 10 | sec.) You're right. I wasn't thinking too clearly when I inferred that they may have intended the extra memory to work like a drive. At the time. 8 Meg was probably only feasible using DRAM, which is not easily backed up using a battery. What do you think of Ensoniq's FLASHBANK (TM) (Ensoniq's capitalization)? | >I just had an idea. You could add another DOC II on a daughter board, and | >rewrite the EPS OS to control two separate DOCs, each with its own 2 Msample | >memory. You would then need to mix the analog outputs of each DOC II. The | >disadvantage would be that the current EPS uses digital mixing for all of its | >20 waves before conversion to analog, while the dual DOC II system would have | >to use the noisier method of analog mixing. Any thoughts about this idea? | | Interesting idea-- but I think it would make the machine much to | complicated (perhaps not to an end user, but electrically). Besides, | it would probably be easier (and cleaner) to just buy a second EPS and | get a MIDI-controllable automated mixer. :-) Or, better yet, a rack mount EPS like the EPS-M ! I have no idea if you could synchronize the two sequencers, though. | >of memory without buying a different machine. | | Speaking of a different machine, does the 16+ have the same memory | limitation? If so, is it for the same reason, or did Ensoniq design it | to have a larger address space, and just doesn't support it yet in the OS? I am completely guessing here, but I assume that the only new chips in the 16+ are those that deal with the DSP features. I assume that the other new 16+ features are due to advanced OS programming which utilizes existing DOC II capabilities. I vaguely remember a reference to a DOC III. Does anyone have any more information about the existence or features of the "DOC III" ? | Mike Evans | dflint@unlv.edu Brian Willoughby SoundSoftware Bellevue, WA USA From compuserve.com!70661.2661 Thu Jul 16 23:28:22 1992 Return-Path: <70661.2661@compuserve.com> Received: from 149.174.128.3 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Thu, 16 Jul 92 23:27 PDT Received: by ihc.compuserve.com (5.65/5.910516) id AA01323; Fri, 17 Jul 92 02:27:09 -0400 Date: 17 Jul 92 02:21:33 EDT From: Simon Insect-o-cutor <70661.2661@CompuServe.COM> To: Subject: EPS and K2000 Message-Id: <920717062133_70661.2661_DHJ33-1@CompuServe.COM> Hi, All this talk about memory has made me want at least another 2 megs pretty severely :). Maybe the next generation EPS... The thing I'm really wondering about is the Kurzweil K2000. I've read that these will read EPS sample disks. Does anyone know what this actually involves? I mean, does it read the disk and build an instrument out of the Ensoniq data, or does it just read the wavesamples? Just curious, not that I could afford a K2000 any time soon. I really enjoy my 16+, but the more I work with it the more the 2 Mb RAM bothers me, and a K2000 with say, 16 Mb of RAM sounds great. The fact that I'd get to keep all my Ensoniq sounds is even nicer. Anxiously hoping Ensoniq will outdo Kurzweil... Jay From wri.com!andre Fri Jul 17 07:10:39 1992 Return-Path: Received: from 140.177.10.12 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Fri, 17 Jul 92 07:10 PDT Received: from rurutu.wri.com by dragonfly.wri.com with SMTP id AA23390 (5.65c/IDA-1.4.4 for ); Fri, 17 Jul 1992 09:10:16 -0500 Return-Path: Date: Fri, 17 Jul 92 09:10:15 -0500 From: andre@wri.com Message-Id: <9207171410.AA00374@rurutu.wri.com> Received: by NeXT Mailer (1.63) To: eps@reed.edu (EPS Mailing List) Subject: FlashBank Question Could someone on this list answer this for me? I've thought about getting the FlashBank upgrade, but from what I've been reading in the Rumour thread, it's no better than a hard disk. I thought all this time that it was like a somewhat hardwired addition to memory. Is that not the case at all? Does one really have to load sounds out of the FlashBank into RAM? Thanks, AK From ads.com!pdel Fri Jul 17 11:28:03 1992 Return-Path: Received: from 128.229.30.16 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Fri, 17 Jul 92 11:26 PDT Received: from bert.ads.com by ads.com (5.65+/1.34v1.3) id AA11786; Fri, 17 Jul 92 11:28:09 -0700 From: pdel@ads.com (Peter Delevoryas) Received: by bert.ads.com (5.65+/4.7) id AA05722; Fri, 17 Jul 92 11:28:07 -0700 Message-Id: <9207171828.AA05722@bert.ads.com> Subject: Re: FlashBank Question To: andre@wri.com Date: Fri, 17 Jul 92 11:28:06 PDT Cc: eps@reed.edu In-Reply-To: <9207171410.AA00374@rurutu.wri.com>; from "andre@wri.com" at Jul 17, 92 9:10 am X-Mailer: ELM [version 2.3 PL11] > > Could someone on this list answer this for me? > > I've thought about getting the FlashBank upgrade, but from what I've > been reading in the Rumour thread, it's no better than a hard disk. I > thought all this time that it was like a somewhat hardwired addition > to memory. Is that not the case at all? Does one really have to load > sounds out of the FlashBank into RAM? I don't think so. The Flashbank adds ROM memory to the 16+. That would seem to me that with a maxed out 16+, you could have 2Mb Ram, plus 1mb flashbank ROM, which would make a total of 3mb of available sounds/sysex or sequences. BUT, whether you can use all 3mb at the same time is another question. I think the case might be that you can either use the 2mb RAM data,or the 1mb Flashbank. Followups, corrections? PD From fscott.sco.COM!sco.sco.COM!jondr Fri Jul 17 11:52:41 1992 Return-Path: <@fscott.sco.COM:jondr@sco.sco.COM> Received: from 137.39.1.5 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Fri, 17 Jul 92 11:52 PDT Received: from sco.sco.COM by relay1.UU.NET with SMTP (5.61/UUNET-internet-primary) id AA20524; Fri, 17 Jul 92 14:52:17 -0400 Received: from fscott.sco.COM by sco.sco.COM id aa00706; Fri, 17 Jul 92 11:52:32 PDT To: eps@reed.edu In-Reply-To: andre@wri.com's message of Fri, 17 Jul 92 09:10:15 -0500 <9207171410.AA00374@rurutu.wri.com> Subject: FlashBank Question From: Desi The Three-Armed Wonder Comic Sender: jondr@sco.COM Reply-To: jondr@sco.COM Date: Fri, 17 Jul 92 11:56:26 PDT Message-Id: <9207171156.aa17575@fscott.sco.COM> >I've thought about getting the FlashBank upgrade, but from what I've >been reading in the Rumour thread, it's no better than a hard disk. I >thought all this time that it was like a somewhat hardwired addition >to memory. Is that not the case at all? Does one really have to load >sounds out of the FlashBank into RAM? the sounds on the flashbank DO NOT use main system RAM. if you have 2000 free system blocks, then load a sound off the flashbank, you still have 2000 free system blocks. weird but true. the thing i absolutely hate about the flashbank is that it is very inflexible. you cannot delete sounds on the flashbank individually, you can only reformat the entire FB. also, any sound that lives in the flashbank cannot be edited. not one single parameter. we have the oex-6 and we're always shuttling sounds around from one output to another. if we load a sound off the flashbank, we are stuck with its output configuration. if you only use the two main outs, this might not be such a problem. if you always use certain sounds in exactly the same way with no tweaking (a large piano sound for example), the flashbank might be useful. having the OS on the bank is a good thing too - the machine boots instantly and there's none of this "INSERT OS DISK" stuff when you sample... make sure you have saved your global parameters to the OS floppy that you use to generate the FB's OS image first! you can't SAVE GLOBAL PARAMETERS to the FB - you get a "WRONG DISK TYPE" error. i had to reformat the FB two or three times before I got the OS with the correct parameters saved. one more thing about the FB - writing to it is very slow! loading is near instantaneous - faster than a HD even, but writing and reformatting crawls. any more questions, feel free to ask... Jon Drukman (God's personal DJ) uunet!sco!jondr jondr@sco.com ------------------------------------------------------------------------------- The vacuum of the mundane holds few secrets. From noc.vitalink.com!ejm Fri Jul 17 13:02:20 1992 Return-Path: Received: from 132.240.17.2 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Fri, 17 Jul 92 13:01 PDT Received: from yamaha.NOC.Vitalink.COM by mescal.NOC.Vitalink.COM with SMTP id AA02740 (5.65c/IDA-1.4.4 for ); Fri, 17 Jul 1992 12:54:14 -0700 Received: from localhost by yamaha.NOC.Vitalink.COM (5.65+V1.2/V1.3) id AA03636; Fri, 17 Jul 92 13:00:46 -0700 Message-Id: <9207172000.AA03636@yamaha.NOC.Vitalink.COM> To: jondr@sco.com Cc: eps@reed.edu Subject: Re: FlashBank Question In-Reply-To: Your message of "Fri, 17 Jul 92 11:56:26 PDT." <9207171156.aa17575@fscott.sco.COM> Date: Fri, 17 Jul 92 13:00:45 -0700 From: ejm@noc.vitalink.com >the sounds on the flashbank DO NOT use main system RAM. if you have >2000 free system blocks, then load a sound off the flashbank, you >still have 2000 free system blocks. weird but true. This tells me that the DOC + related chips *can* access > 2mb of space. Perhaps 4mb, but probably much more. ... Erik From nwnexus.wa.com!sounds!brianw Fri Jul 17 13:42:34 1992 Return-Path: Received: from 192.135.191.1 by reed.edu (/\==/\ Smail3.1.25.1 #25.21) id ; Fri, 17 Jul 92 13:42 PDT Received: by nwnexus.wa.com id AA10067 (5.65c/IDA-1.4.4 for reed.edu!EPS); Fri, 17 Jul 1992 13:45:21 -0700 Received: by sounds.wa.com (NeXT-1.0 (From Sendmail 5.52)/NeXT-2.0) id AA10187; Fri, 17 Jul 92 13:37:44 PDT Date: Fri, 17 Jul 92 13:37:44 PDT From: sounds!brianw@nwnexus.wa.com (Brian Willoughby) Message-Id: <9207172037.AA10187@ sounds.wa.com > Received: by NeXT Mailer (1.63) To: EPS@reed.edu (EPS Mailing List) Subject: Re: FlashBank Question | From: Desi The Three-Armed Wonder Comic | >I've thought about getting the FlashBank upgrade, but from what I've | >been reading in the Rumour thread, it's no better than a hard disk. I | >thought all this time that it was like a somewhat hardwired addition | >to memory. Is that not the case at all? Does one really have to load | >sounds out of the FlashBank into RAM? | | the sounds on the flashbank DO NOT use main system RAM. if you have 2000 | free system blocks, then load a sound off the flashbank, you still | have 2000 free system blocks. weird but true. My EPS with 4X memory (and no FLASHBANK) shows 4091 free system blocks when booted. Have you ever tried loading flashbank sounds on a machine which already has the maximum RAM? I have a hunch that a machine which is already full of RAM would need to copy the Instruments from flashbank to RAM in order to play them. Therefore, I would assume that an EPS starting with 4000 free system blocks would show that some were used after loading a sound off the flashbank. Considering the ingenious ways that Ensoniq has provided for multiple memory configurations, I wouldn't be surprised if a machine which is NOT full of RAM would be able to squeeze the flashbank instruments in without using any RAM. If any of you recall the original EPS, there are three memory configurations. The unexpanded EPS has 256 Kwords of memory, with 1024 blocks shared between wavesamples and sequences. The 2X memory expander increases this to 512 Kwords (256K internal + 256K external), but the total 2048 blocks are still shared. The 4X memory increases this again to 1280 Kwords (256K internal + 1024K external), but the memory can no longer be shared between wavesamples and sequences. On the 4X EPS configuration, there are 4096 system blocks and 1024 blocks for the sequencer. Go to the Command+Seq.Song+6.Amp page (Sequencer Information) and hit enter a few times. On my 4X EPS, I see "All Seq Data=1024 Blks". This results from the limitation that the DOC II cannot access more than 1024 Kwords of wavesample memory, so the remainder of the 1280 Kwords is used by the EPS OS for sequencer information (which the DOC II never accesses). Based upon the system architecture described above, I would surmise that Jon's EPS has 512 Kwords of RAM and 512 Kwords of FLASHBANK which are all accessable by the DOC II. By the way 512 Kwords is actually 1 Mbyte, since the EPS 16+ uses 16-bit words. Besides, I don't think that you can plug in the 1 Mbyte EPS 16+ RAM card at the same time as the 1 Mbyte FLASHBANK, so there is no way to max out the RAM on a 16+. The EPS Classic uses 13-bit memory, so the Mbyte sizes advertized were always nearly completely confusing and unnatural. There is one advantage of the EPS Classic over the EPS 16+ which I am still happy about: you can never have 1280 Mwords of memory in a 16+, the absolute maximum is 1024 Mwords. Consequently, song data always competes with wavesample data for memory space. The EPS Classic is still the most expandable unit. Jon, I assume that you have a flashbank, considering your detailed experience. I don't have one, but I am very curious to find out more about them. Do you have an EPS 16+? Do flashbank cards work with the standard EPS? Does the EPS 16+ report block sizes any differently than the EPS? If the block sizes are not the same, then my comments above may be incorrect, but I don't think so considering the detailed specification sheet I have on the EPS 16+. Brian Willoughby SoundSoftware Bellevue, WA USA