<?xml version="1.0" encoding="utf-8"?>
<feed xmlns="http://www.w3.org/2005/Atom">
  <title>The Relling Blog</title>
  <subtitle>Sourced writing on robot deployment, machine vision and automation economics.</subtitle>
  <link href="https://rellingsystems.com/resources/blog"/>
  <link rel="self" href="https://rellingsystems.com/feed.xml"/>
  <id>https://rellingsystems.com/resources/blog</id>
  <updated>2026-10-24T00:00:00Z</updated>
  <entry>
    <title>How to start a machine vision project, from the first pick to a station your team owns</title>
    <link href="https://rellingsystems.com/resources/blog/starting-your-first-machine-vision-project"/>
    <id>https://rellingsystems.com/resources/blog/starting-your-first-machine-vision-project</id>
    <published>2026-10-24T00:00:00Z</published>
    <updated>2026-10-24T00:00:00Z</updated>
    <author><name>Relling Engineering</name></author>
    <summary>How to start a machine vision project: pick the right first application, write a one-page spec (template inside), run feasibility, then budget and accept it.</summary>
  </entry>
  <entry>
    <title>Manual vs automated visual inspection: what the research says about people, and when cameras win</title>
    <link href="https://rellingsystems.com/resources/blog/manual-vs-automated-visual-inspection"/>
    <id>https://rellingsystems.com/resources/blog/manual-vs-automated-visual-inspection</id>
    <published>2026-10-24T00:00:00Z</published>
    <updated>2026-10-24T00:00:00Z</updated>
    <author><name>Relling Engineering</name></author>
    <summary>Manual vs automated visual inspection, from the research: why inspectors catch about 80 to 85% of defects, where people beat cameras, plus a decision table.</summary>
  </entry>
  <entry>
    <title>Machine vision software compared: pick the platform your team can still edit in year two</title>
    <link href="https://rellingsystems.com/resources/blog/machine-vision-software-compared"/>
    <id>https://rellingsystems.com/resources/blog/machine-vision-software-compared</id>
    <published>2026-10-24T00:00:00Z</published>
    <updated>2026-10-24T00:00:00Z</updated>
    <author><name>Relling Engineering</name></author>
    <summary>Machine vision software compared: HALCON vs VisionPro, Zebra Aurora, Sherlock, pylon vTools, Keyence, OpenCV and AI tools, with licensing and support.</summary>
  </entry>
  <entry>
    <title>Machine vision maintenance: a daily-to-annual checklist that catches drift before your customer does</title>
    <link href="https://rellingsystems.com/resources/blog/machine-vision-maintenance-checklist"/>
    <id>https://rellingsystems.com/resources/blog/machine-vision-maintenance-checklist</id>
    <published>2026-10-24T00:00:00Z</published>
    <updated>2026-10-24T00:00:00Z</updated>
    <author><name>Relling Engineering</name></author>
    <summary>Machine vision maintenance checklist by interval: golden-sample checks, lens cleaning, LED aging, backups, change control and spares, plus what to log.</summary>
  </entry>
  <entry>
    <title>Machine vision camera resolution: how many megapixels you need, and when more hurts</title>
    <link href="https://rellingsystems.com/resources/blog/machine-vision-camera-resolution"/>
    <id>https://rellingsystems.com/resources/blog/machine-vision-camera-resolution</id>
    <published>2026-10-24T00:00:00Z</published>
    <updated>2026-10-24T00:00:00Z</updated>
    <author><name>Relling Engineering</name></author>
    <summary>How many megapixels do you need for machine vision? Rules of thumb with the reasoning, the myths that waste money, and three worked inspection examples.</summary>
  </entry>
  <entry>
    <title>How to choose a machine vision lens without wasting the camera you paid for</title>
    <link href="https://rellingsystems.com/resources/blog/how-to-choose-a-machine-vision-lens"/>
    <id>https://rellingsystems.com/resources/blog/how-to-choose-a-machine-vision-lens</id>
    <published>2026-10-24T00:00:00Z</published>
    <updated>2026-10-24T00:00:00Z</updated>
    <author><name>Relling Engineering</name></author>
    <summary>How to choose a machine vision lens: mounts, focal length, image circle, resolving power, telecentric vs fixed focal lenses, filters and a 10-step checklist.</summary>
  </entry>
  <entry>
    <title>Cognex DataMan vs Keyence SR: which barcode reader for which job</title>
    <link href="https://rellingsystems.com/resources/blog/dataman-vs-keyence-sr-barcode-readers"/>
    <id>https://rellingsystems.com/resources/blog/dataman-vs-keyence-sr-barcode-readers</id>
    <published>2026-10-24T00:00:00Z</published>
    <updated>2026-10-24T00:00:00Z</updated>
    <author><name>Relling Engineering</name></author>
    <summary>Cognex DataMan vs Keyence SR barcode readers compared on DPM reading, liquid lens autofocus, ISO verification, scan tunnels and setup, with a test plan.</summary>
  </entry>
  <entry>
    <title>Cognex In-Sight vs Keyence: the smart cameras, model by model</title>
    <link href="https://rellingsystems.com/resources/blog/cognex-in-sight-vs-keyence-vs"/>
    <id>https://rellingsystems.com/resources/blog/cognex-in-sight-vs-keyence-vs</id>
    <published>2026-10-24T00:00:00Z</published>
    <updated>2026-10-24T00:00:00Z</updated>
    <author><name>Relling Engineering</name></author>
    <summary>Cognex In-Sight vs Keyence VS, model by model: In-Sight 2800, 3800, 3900 and 6900 vs VS, VS-G and IV4 on AI training, optics, I/O and what drives real cost.</summary>
  </entry>
  <entry>
    <title>Vision system validation: how to qualify an inspection station with IQ, OQ, PQ and gage R&amp;R</title>
    <link href="https://rellingsystems.com/resources/blog/vision-system-validation"/>
    <id>https://rellingsystems.com/resources/blog/vision-system-validation</id>
    <published>2026-10-23T00:00:00Z</published>
    <updated>2026-10-23T00:00:00Z</updated>
    <author><name>Relling Engineering</name></author>
    <summary>Vision system validation for inspection stations: what goes in IQ, OQ and PQ, how big the test set must be, gage R&amp;R vs attribute agreement, and a checklist.</summary>
  </entry>
  <entry>
    <title>Vision system false rejects: why your camera fails good parts, and how to stop it</title>
    <link href="https://rellingsystems.com/resources/blog/vision-system-false-rejects"/>
    <id>https://rellingsystems.com/resources/blog/vision-system-false-rejects</id>
    <published>2026-10-23T00:00:00Z</published>
    <updated>2026-10-23T00:00:00Z</updated>
    <author><name>Relling Engineering</name></author>
    <summary>Vision system false rejects trace to lighting drift, dirty optics or a threshold set on 30 parts. Find all 8 causes with our troubleshooting flowchart.</summary>
  </entry>
  <entry>
    <title>Machine vision vs computer vision: same math, different job</title>
    <link href="https://rellingsystems.com/resources/blog/machine-vision-vs-computer-vision"/>
    <id>https://rellingsystems.com/resources/blog/machine-vision-vs-computer-vision</id>
    <published>2026-10-23T00:00:00Z</published>
    <updated>2026-10-23T00:00:00Z</updated>
    <author><name>Relling Engineering</name></author>
    <summary>Machine vision vs computer vision, from integrators who build both into cells: lighting, cycle time, PLC I/O, validation, where AI fits and how to choose.</summary>
  </entry>
  <entry>
    <title>Machine vision system cost in 2026, and why the camera is the cheap part</title>
    <link href="https://rellingsystems.com/resources/blog/machine-vision-system-cost"/>
    <id>https://rellingsystems.com/resources/blog/machine-vision-system-cost</id>
    <published>2026-10-23T00:00:00Z</published>
    <updated>2026-10-23T00:00:00Z</updated>
    <author><name>Relling Engineering</name></author>
    <summary>Machine vision system cost in 2026, from a $761 sensor to $150k+ 3D stations. Cited list prices, the full station cost stack, and how to compare quotes.</summary>
  </entry>
  <entry>
    <title>How to calculate machine vision ROI, with every number shown</title>
    <link href="https://rellingsystems.com/resources/blog/machine-vision-roi"/>
    <id>https://rellingsystems.com/resources/blog/machine-vision-roi</id>
    <published>2026-10-23T00:00:00Z</published>
    <updated>2026-10-23T00:00:00Z</updated>
    <author><name>Relling Engineering</name></author>
    <summary>Machine vision ROI made concrete: what to count, a payback example with every number shown, and how false rejects can quietly add five months to payback.</summary>
  </entry>
  <entry>
    <title>Cognex and Keyence alternatives: nine options, and when each one actually wins</title>
    <link href="https://rellingsystems.com/resources/blog/cognex-and-keyence-alternatives"/>
    <id>https://rellingsystems.com/resources/blog/cognex-and-keyence-alternatives</id>
    <published>2026-10-23T00:00:00Z</published>
    <updated>2026-10-23T00:00:00Z</updated>
    <author><name>Relling Engineering</name></author>
    <summary>Cognex and Keyence alternatives for 2026, compared by an integrator: Omron, Zebra, SICK, Basler, MVTec, LMI and AI-first vendors, with a decision tree.</summary>
  </entry>
  <entry>
    <title>AI visual inspection in 2026: what works on a real line, and what still doesn&#x27;t</title>
    <link href="https://rellingsystems.com/resources/blog/ai-visual-inspection"/>
    <id>https://rellingsystems.com/resources/blog/ai-visual-inspection</id>
    <published>2026-10-23T00:00:00Z</published>
    <updated>2026-10-23T00:00:00Z</updated>
    <author><name>Relling Engineering</name></author>
    <summary>AI visual inspection in 2026, from people who install it: what deep learning catches, where it fails, how many images you need, and a pilot plan that works.</summary>
  </entry>
  <entry>
    <title>2D vs 3D machine vision: when you actually need depth</title>
    <link href="https://rellingsystems.com/resources/blog/2d-vs-3d-machine-vision"/>
    <id>https://rellingsystems.com/resources/blog/2d-vs-3d-machine-vision</id>
    <published>2026-10-23T00:00:00Z</published>
    <updated>2026-10-23T00:00:00Z</updated>
    <author><name>Relling Engineering</name></author>
    <summary>2D vs 3D machine vision: what a 2D camera can&#x27;t measure, how laser, structured light, stereo and ToF compare on real specs, and a decision tree for your line.</summary>
  </entry>
  <entry>
    <title>The factory floor is forty years behind the office. That is the opportunity</title>
    <link href="https://rellingsystems.com/resources/blog/the-floor-is-forty-years-behind"/>
    <id>https://rellingsystems.com/resources/blog/the-floor-is-forty-years-behind</id>
    <published>2026-09-02T00:00:00Z</published>
    <updated>2026-09-04T00:00:00Z</updated>
    <author><name>David Lam</name></author>
    <summary>The office got the 21st century and the factory floor mostly did not. US productivity growth fell from 2.1 to 0.8 percent a year, durable-goods manufacturing accounts for about half of that, and AI use is concentrated in offices and large firms. Why the floor is behind, what changed, whether AI should be allowed to run things, and what the bridge looks like now.</summary>
  </entry>
  <entry>
    <title>ISO 10218 and R15.06 were rewritten. What changed for your cell</title>
    <link href="https://rellingsystems.com/resources/blog/iso-10218-r1506-2025-what-changed"/>
    <id>https://rellingsystems.com/resources/blog/iso-10218-r1506-2025-what-changed</id>
    <published>2026-08-27T00:00:00Z</published>
    <updated>2026-09-03T00:00:00Z</updated>
    <author><name>Jai Relan</name></author>
    <summary>ISO 10218-1 and 10218-2 were republished in January 2025 and the US adoption ANSI/A3 R15.06-2025 followed in September. ISO/TS 15066 is folded in, robots are now classified, cybersecurity is in scope, and the 2012 US standard is withdrawn. What a buyer should actually do about it.</summary>
  </entry>
  <entry>
    <title>What the 2026 robot order numbers actually say</title>
    <link href="https://rellingsystems.com/resources/blog/robot-orders-2026-what-the-numbers-say"/>
    <id>https://rellingsystems.com/resources/blog/robot-orders-2026-what-the-numbers-say</id>
    <published>2026-08-18T00:00:00Z</published>
    <updated>2026-09-03T00:00:00Z</updated>
    <author><name>Vidyut Baradwaj</name></author>
    <summary>North American robot orders grew 4.3 percent in units in Q2 2026 but 21.3 percent in value. That gap, plus a non-automotive majority and a cobot share that is bigger in units than in dollars, is the real story. Sourced analysis of the A3 and IFR data.</summary>
  </entry>
  <entry>
    <title>Why most robot pilots never reach production</title>
    <link href="https://rellingsystems.com/resources/blog/why-robot-pilots-stall"/>
    <id>https://rellingsystems.com/resources/blog/why-robot-pilots-stall</id>
    <published>2026-08-06T00:00:00Z</published>
    <updated>2026-09-01T00:00:00Z</updated>
    <author><name>Anya Singh</name></author>
    <summary>Robot pilots rarely fail on the robot. They fail on part presentation, an easy pilot part, an undefined finish line, no owner after handover, and a pilot scoped to prove feasibility rather than to run. Five failure modes and how to write them out of a contract.</summary>
  </entry>
  <entry>
    <title>The best robot integrators for high-mix manufacturing in 2026</title>
    <link href="https://rellingsystems.com/resources/blog/best-robot-integrators-high-mix-manufacturing"/>
    <id>https://rellingsystems.com/resources/blog/best-robot-integrators-high-mix-manufacturing</id>
    <published>2026-07-16T00:00:00Z</published>
    <updated>2026-09-02T00:00:00Z</updated>
    <author><name>Cabrel Happi</name></author>
    <summary>Five ways to get a robotic cell onto a high-mix, low-volume floor in 2026, ranked on fit rather than size: Relling, Formic, Acieta, JR Automation, and ATS. With a stated method, sourced figures, and a disclosure.</summary>
  </entry>
</feed>
