<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=Windows-1252">
<style type="text/css" style="display:none;"> P {margin-top:0;margin-bottom:0;} </style>
</head>
<body dir="ltr">
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0); background-color: rgb(255, 255, 255);" class="elementToProof">
Almost every minicomputer of the time used 2901s. You could do floating point but it was slow (and also non-standard as IEEE-754 was only in its early stages so each machine had different characteristics).</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0); background-color: rgb(255, 255, 255);" class="elementToProof">
<br>
</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0); background-color: rgb(255, 255, 255);" class="elementToProof">
I actually got the first TMS320 in the UK (helped working for HP back then :-) and as you say the 16 bit format was a bit restricting, and indeed the product we intended to use it for was scrapped and a version using analogue filters was developed which turned
out to be cheaper.<br>
</div>
<div id="appendonsend"></div>
<hr style="display:inline-block;width:98%" tabindex="-1">
<div id="divRplyFwdMsg" dir="ltr"><font face="Calibri, sans-serif" style="font-size:11pt" color="#000000"><b>From:</b> brianw <brianw@audiobanshee.com><br>
<b>Sent:</b> 25 November 2022 19:53<br>
<b>To:</b> Mike Bryant <mbryant@futurehorizons.com><br>
<b>Cc:</b> synth-diy@synth-diy.org <synth-diy@synth-diy.org><br>
<b>Subject:</b> Re: [sdiy] Buchla 295 10-band comb filter topology</font>
<div> </div>
</div>
<div class="BodyFragment"><font size="2"><span style="font-size:11pt;">
<div class="PlainText">Wow! A quick search for AMD2901A mentions floating point processing. I'm impressed for '78.<br>
<br>
I couldn't find a data sheet. Most of the search hits seemed to land on some other part.<br>
<br>
The TMS320 DSP came out in 1983. Fixed-point DSP can work quite well, if you don't mind coefficient quantization moving your filter knee at bit away from ideal. Some older "DSP" gear just has discrete math chips in TTL or equivalent, with an A/D and D/A around
it and a slow CPU to change controls signals without actually processing the audio itself.<br>
<br>
<br>
On Nov 22, 2022, at 3:55 AM, Mike Bryant <mbryant@futurehorizons.com> wrote:<br>
> I hadn't realised it was that old, but was definitely possible though rather expensive. I used banks of AMD2901A for digital signal processing in several HP instruments around that time. I built a 'digital Fairlight' using them a bit later on but then
Yamaha launched the DX7 which killed any hope of selling mine at a viable price.<br>
>> From: Guy McCusker <guy.mccusker@gmail.com><br>
>> Sent: 22 November 2022 11:46<br>
>> <br>
>> Would a workable digital implementation have been possible in 1978? I<br>
>> always assume the answer is no but one can be surprised.<br>
>> <br>
>> On Tue, Nov 22, 2022 at 11:25 AM Mike Bryant <mbryant@futurehorizons.com> wrote:<br>
>> ><br>
>> > > Buchla’s later filterbank, 296 Spectral Processor does a much better job of leaving the signal alone when all the bands are zeroed.<br>
>> ><br>
>> > I read this far and immediately assumed he'd done it digitally<br>
>> ><br>
>> ><br>
>> > > But, those filters are 3 stages deep and the module has almost a million trim pots.<br>
>> ><br>
>> > But then I discovered not :-)<br>
>> ><br>
>> > ________________________________<br>
>> > From: mark verbos <mark@verboselectronics.com><br>
>> > Sent: 22 November 2022 11:20<br>
>> ><br>
>> > I have one of these. Definitely has “dramatic changes in sound”. In fact, everything that goes through it sounds like it went down the tube of a vacuum cleaner. The pots on the filters are trims, not on the panel and just regular single pots. The panel
controls are just level sliders. It is pretty similar to the Serge Resonant EQ, which I’m sure is no accident. It does a good job of taking normal synth sounds and making them more weird, acoustic and less placeable as normal synth sounds. Buchla’s later filterbank,
296 Spectral Processor does a much better job of leaving the signal alone when all the bands are zeroed. But, those filters are 3 stages deep and the module has almost a million trim pots.<br>
>> ><br>
>> > Mark<br>
>> ><br>
</div>
</span></font></div>
</body>
</html>