<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: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);" class="elementToProof">
A friend and I tried getting it to create a DSP version of the classic Moog filter, and it came up with a very interesting structure to the IIR implementation. I looked all over the Internet and couldn't find anybody who had done it in exactly the same way.
Obviously it had got the concept from somewhere, but no idea where, or why it chose that over the more common implementation.<br>
</div>
<div style="font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);" class="elementToProof">
<br>
</div>
<div style="font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);" class="elementToProof">
In fact what it showed up was most people copy, or at least adapt, other people's work anyway, which is all an AI will ever do, for now at least :-)<br>
</div>
<div id="appendonsend"></div>
<div style="font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<br>
</div>
<hr tabindex="-1" style="display:inline-block; width:98%">
<div id="divRplyFwdMsg" dir="ltr"><font face="Calibri, sans-serif" style="font-size: 11pt; color: rgb(0, 0, 0);"><b>From:</b> Synth-diy <synth-diy-bounces@synth-diy.org> on behalf of Benjamin Tremblay via Synth-diy <synth-diy@synth-diy.org><br>
<b>Sent:</b> 13 October 2023 13:39<br>
<b>To:</b> Sean Ellis <tensiontype@hotmail.com><br>
<b>Cc:</b> synth-diy@synth-diy.org <synth-diy@synth-diy.org><br>
<b>Subject:</b> Re: [sdiy] Chat GPT Image analysis.</font>
<div> </div>
</div>
<div class="BodyFragment"><font size="2"><span style="font-size:11pt">
<div class="PlainText">We all know it would be possible to train an LLM to be a crackshot C++ coder.<br>
Using GPT is like talking with someone who reads all day and remembers everything they have ever read. If you ask them about something, they have a lot to say, but there’s no editorial process; it’s just a memory dump.<br>
Because generative AI is about breaking down compositions and re-composing the fragments according to the patterns it discovers, GPT can appear to know what it’s doing, but it is only good at faking the style of a certain kind of document.
<br>
It’s smart enough to extract a concept and give it a camel-case name and reference that name like it was a library class, but might leave you believing the library exists when in fact it’s computer-generated fiction.<br>
<br>
If you take something like GPT and train a new top-level layer of learning just on your domain, like DSP programming, it can become very useful, but still just a well-read analyst who makes mostly predictable suggestions through sample code.<br>
<br>
Benjamin<br>
_______________________________________________<br>
Synth-diy mailing list<br>
Synth-diy@synth-diy.org<br>
<a href="http://synth-diy.org/mailman/listinfo/synth-diy" data-auth="NotApplicable" id="OWAf403ebe4-eba1-d413-2dd4-4917277035fe" class="OWAAutoLink">http://synth-diy.org/mailman/listinfo/synth-diy</a><br>
Selling or trading? Use marketplace@synth-diy.org<br>
</div>
</span></font></div>
</body>
</html>