<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<p>Okay, that was silly. I would need mondo processor power to do
that. Looks like its FPGA time! Yay! I like Xilinx so the
questiob becomes how much FPGA do I need? Spartan-6?</p>
<p>--timmers<br>
</p>
<br>
<div class="moz-cite-prefix">On 8/4/2018 9:14 AM, Tim Ressel wrote:<br>
</div>
<blockquote type="cite"
cite="mid:57aa3b6d-4212-1cd6-fef4-d6fcb713dc46@circuitabbey.com">
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<p>Thanks everyone! It looks like the FPGA route is doable,
although those CIC filters make me think I can do this on a
processor after all. Maybe 2 procs: one for handling MIDI and
ADC tasks, and one as a oscillator core. Use SPI to write values
to the osc core. NCOs are cheap so doing 9 instances is no
problemo. I need VCAs on each NCO but that is just a multiply.
That might just work.</p>
<p>--timbo<br>
</p>
<br>
<div class="moz-cite-prefix">On 8/4/2018 6:06 AM, Scott
Gravenhorst wrote:<br>
</div>
<blockquote type="cite"
cite="mid:20180804130658.EDFE2981EBE@pi3c.lan">
<pre wrap="">Michael Zacherl <a class="moz-txt-link-rfc2396E" href="mailto:sdiy-mz01@blauwurf.info" moz-do-not-send="true"><sdiy-mz01@blauwurf.info></a> wrote:
</pre>
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">On 4 Aug 2018, at 06:38, Scott Gravenhorst <a class="moz-txt-link-rfc2396E" href="mailto:music.maker@gte.net" moz-do-not-send="true"><music.maker@gte.net></a> wrote:
but I did it the "wrong" way by making an FIR filter with a
huge kernel. The filter did the job to remove most of the alias artifacts, but it
did some weird frequency dependent amplitude things especially at the high end of the
audio spectrum. I later learned that it is better to use cascaded smaller kernel
filters,
</pre>
</blockquote>
<pre wrap="">Hi,
“kernel†== order of the filter?
m.
</pre>
</blockquote>
<pre wrap="">Sorry, I neglected to state that the filter is a FIR type.
FIR filters have a kernel of coefficients which define the impulse response. The larger the
kernel, the more definition one can have in the output spectral profile. Larger kernels also take
more time to compute which is a drawback of the FIR filter. FIR filters can make good brick wall
filters, that is, very flat response with steep roll off into the stop band.
-- ScottG
________________________________________________________________________
-- Scott Gravenhorst
-- <a class="moz-txt-link-freetext" href="http://scott.joviansynth.com/" moz-do-not-send="true">http://scott.joviansynth.com/</a>
-- When the going gets tough, the tough use the command line.
-- Matt 21:22
</pre>
<br>
<fieldset class="mimeAttachmentHeader"></fieldset>
<br>
<pre wrap="">_______________________________________________
Synth-diy mailing list
<a class="moz-txt-link-abbreviated" href="mailto:Synth-diy@synth-diy.org" moz-do-not-send="true">Synth-diy@synth-diy.org</a>
<a class="moz-txt-link-freetext" href="http://synth-diy.org/mailman/listinfo/synth-diy" moz-do-not-send="true">http://synth-diy.org/mailman/listinfo/synth-diy</a>
</pre>
</blockquote>
<br>
<pre class="moz-signature" cols="72">--
--Tim Ressel
Circuit Abbey
<a class="moz-txt-link-abbreviated" href="mailto:timr@circuitabbey.com" moz-do-not-send="true">timr@circuitabbey.com</a></pre>
<br>
<fieldset class="mimeAttachmentHeader"></fieldset>
<br>
<pre wrap="">_______________________________________________
Synth-diy mailing list
<a class="moz-txt-link-abbreviated" href="mailto:Synth-diy@synth-diy.org">Synth-diy@synth-diy.org</a>
<a class="moz-txt-link-freetext" href="http://synth-diy.org/mailman/listinfo/synth-diy">http://synth-diy.org/mailman/listinfo/synth-diy</a>
</pre>
</blockquote>
<br>
<pre class="moz-signature" cols="72">--
--Tim Ressel
Circuit Abbey
<a class="moz-txt-link-abbreviated" href="mailto:timr@circuitabbey.com">timr@circuitabbey.com</a></pre>
</body>
</html>