<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Floating Point on Pranav Kumar</title><link>https://pranav083.github.io/tags/floating-point/</link><description>Recent content in Floating Point on Pranav Kumar</description><generator>Hugo -- gohugo.io</generator><language>en</language><copyright>Copyright © 2016–2030, Pranav Kumar</copyright><lastBuildDate>Mon, 20 Apr 2026 00:00:00 -0400</lastBuildDate><atom:link href="https://pranav083.github.io/tags/floating-point/index.xml" rel="self" type="application/rss+xml"/><item><title>Floating-Point Arithmetic: Performance vs. Area on an FPGA</title><link>https://pranav083.github.io/post/project/fp-arithmetic-ppa/</link><pubDate>Mon, 20 Apr 2026 00:00:00 -0400</pubDate><guid>https://pranav083.github.io/post/project/fp-arithmetic-ppa/</guid><description>
&lt;p&gt;&lt;strong&gt;ECE 4515 — Digital Design II · Virginia Tech · Spring 2026&lt;/strong&gt;&lt;/p&gt;
&lt;p&gt;A hardware design-space exploration: build the same floating-point datapath at different pipeline depths, synthesise each on a Cyclone V, and measure what the extra performance actually costs in silicon.&lt;/p&gt;
&lt;hr&gt;
&lt;h2 id="the-question"&gt;The question&lt;/h2&gt;
&lt;p&gt;Deeper pipelines let a datapath run at a higher clock, but each additional stage costs registers, and wider or faster arithmetic units cost logic and DSP blocks. On a fixed device, that budget is finite.&lt;/p&gt;</description></item></channel></rss>