<?xml version="1.0" encoding="utf-8" standalone="yes"?><?xml-stylesheet type="text/xsl" href="https://itohi.com/feed.xsl"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>FPV on ITOHI — Open-Source Hardware, Embedded Systems &amp; DIY Builds</title><link>https://itohi.com/categories/fpv/</link><description>Recent content in FPV on ITOHI — Open-Source Hardware, Embedded Systems &amp; DIY Builds</description><generator>Hugo -- gohugo.io</generator><language>en</language><copyright>2022 @ ITOHI - All rights reserved</copyright><lastBuildDate>Tue, 14 Jul 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://itohi.com/categories/fpv/index.xml" rel="self" type="application/rss+xml"/><item><title>FPV Tuning with Sintra AI — And Why I'm Not Building Another AI Platform</title><link>https://itohi.com/fpv/fpv-tuning-sintra/</link><pubDate>Tue, 14 Jul 2026 00:00:00 +0000</pubDate><guid>https://itohi.com/fpv/fpv-tuning-sintra/</guid><description>
&lt;p&gt;This post grew out of the &lt;a href="https://itohi.com/fpv/drone-detector-nn/"&gt;drone detector article&lt;/a&gt;, where I mentioned using Sintra for the FPV side of the work. That mention deserves its own treatment, because it raises a question I want to address directly: why am I using an external AI tool for this instead of building my own?&lt;/p&gt;
&lt;hr&gt;
&lt;h2 id="why-sintra-not-my-own-ai"&gt;Why Sintra, Not My Own AI&lt;/h2&gt;
&lt;p&gt;There are projects in my history I don't talk about much. MentalMentor. BabyAI. Others at various levels of completion, none of which became products. The pattern is consistent across all of them: I build the technically interesting part — the architecture works, the core functionality is there — and then the work shifts to distribution, community building, marketing. Things that require either a budget I don't have or social capital I've never managed to accumulate. I have no network. I cannot self-promote. The projects expire quietly.&lt;/p&gt;</description></item><item><title>FPV: how it all began</title><link>https://itohi.com/fpv/fpv-journey-dronefix/</link><pubDate>Mon, 13 Jul 2026 00:00:00 +0000</pubDate><guid>https://itohi.com/fpv/fpv-journey-dronefix/</guid><description>
&lt;p&gt;For years my drones travelled with me. A DJI Air first, then a Mini 3 — camera platforms that rode along on every trip, packed next to the rest of the gear. They were good at exactly one thing: hovering somewhere legal and taking a clean photo.&lt;/p&gt;
&lt;p&gt;That was the problem. Hovering at the permitted altitude, framing a mountain from a respectful distance, felt underwhelming every single time. The footage was fine. It was also lifeless — the drone observed from a polite distance and brought back postcards. Eventually I stopped packing one at all. On motorcycle trips especially, the Mini 3 stayed home; the weight and the faff weren't worth another set of hover shots.&lt;/p&gt;</description></item><item><title>Pavo20 Pro II GPS Fix Attempts: BEC Switching Noise at 1575 MHz and What Actually Helped</title><link>https://itohi.com/fpv/pavo20-gps-struggles/</link><pubDate>Mon, 13 Jul 2026 00:00:00 +0000</pubDate><guid>https://itohi.com/fpv/pavo20-gps-struggles/</guid><description>
&lt;p&gt;The Pavo20 Pro II is a capable 2.5&amp;quot; whoop — compact, powerful, with a good camera and a surprisingly solid video link even on linear whip antennas. I wanted it as a mountain trip quad: something that fits in a jacket pocket, dives canyons, and has GPS Rescue as a genuine safety net. GPS is not included; I added a module myself, soldered and configured it. That's when the problems started.&lt;/p&gt;</description></item></channel></rss>