<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=us-ascii">
<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">
What I did was put a 4*4 multiplier in a 256x9 ROM with a carry bit and addressed it through some 4/8/12 bit shifters multiple times. Saves a lot of single bit shifting.</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">
Whatever happened to x9 memories - used to be commonplace back then.<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> Donald Tillman <don@till.com><br>
<b>Sent:</b> 25 November 2022 21:36<br>
<b>To:</b> brianw <brianw@audiobanshee.com><br>
<b>Cc:</b> Mike Bryant <mbryant@futurehorizons.com>; 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="" style="word-wrap:break-word; line-break:after-white-space">The AMD2901 was a general purpose ALU slice with a set of simple operations. It couldn't do multiplication by itself. You had to use a lot of shift and add microcoded cycles for a multiply
operation; which ends up taking a very long time.
<div class=""><br class="">
</div>
<div class="">What you need is a circuit that can do a 16x16 multiply and accumulate in a single fast cycle.</div>
<div class=""><br class="">
</div>
<div class="">The TRW MPY-16 chip came out around 1978 and could do that in about 200nS for $300.00 and a significant amount of support circuitry. (From memory, don't quote me.)</div>
<div class=""><br class="">
</div>
<div class="">Before that you would have had to use an awful lot of gates and MSI chips.</div>
<div class=""><br class="">
</div>
<div class=""> -- Don</div>
<div class="">
<div class="">
<div dir="auto" class="" style="word-wrap:break-word; line-break:after-white-space">
<div style="color:rgb(0,0,0); font-family:Menlo; font-size:12px; font-style:normal; font-variant-caps:normal; font-weight:normal; letter-spacing:normal; text-align:start; text-indent:0px; text-transform:none; white-space:normal; word-spacing:0px">
--<br class="">
Donald Tillman, Palo Alto, California<br class="">
<a href="https://www.till.com" class="">https://www.till.com</a></div>
</div>
</div>
<div><br class="">
<blockquote type="cite" class="">
<div class="">On Nov 25, 2022, at 11:53 AM, brianw <<a href="mailto:brianw@audiobanshee.com" class="">brianw@audiobanshee.com</a>> wrote:</div>
<br class="x_Apple-interchange-newline">
<div class="">
<div class="">Wow! A quick search for AMD2901A mentions floating point processing. I'm impressed for '78.<br class="">
<br class="">
I couldn't find a data sheet. Most of the search hits seemed to land on some other part.<br class="">
<br class="">
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 class="">
<br class="">
</div>
</div>
</blockquote>
</div>
<br class="">
</div>
</div>
</body>
</html>