Color Accuracy & SSI Testing
Manufacturer CRI sheets rarely predict how a COB renders saturated packaging reds or cosmetic skin tones on a 1/250 s shutter. This pillar documents our Spectral Similarity Index workflow (aligned with AMPAS SSI thinking), IEEE 1789 flicker checks, and the dimming curves that cause green–magenta drift below 20% output — the failure mode jewelry and cosmetics shooters report most often on Reddit.
Articles in this pillar
- High CRI LED Color Shift When Dimming: What Product Sets Should Measure
Why Ra 96+ COB LEDs still drift green–magenta when dimmed—and how to test spectral shift at 15–50% output before a packshot day goes to retouch.
- Flicker-Free LED Lights for Photography: Shutter Bands That Hold
Test LED flicker against real shutter speeds and dim levels—IEEE 1789 mindset, 50/60 Hz mains, and a cart-ready safe band for product stills.
Related shootouts
- Aputure 120d II vs Godox SL150II: product photography 2026
- Nanlite Forza 150 vs Godox ML60II: tabletop 2026
- Amaran 200d S vs Godox LA200D: color accuracy 2026
Other pillars
FAQ
No. CRI averages eight pastel swatches; SSI compares a light’s spectrum to a reference illuminant for cinema-style matching. Our SSI-to-CRI converter tool explains the math and when each metric misleads.
For mixed-vendor sets, we want ≥ 85 SSI to a 5600 K reference for packshots, and ≥ 82 when dimming below 15% if you shoot glossies — lower scores predict visible duv shifts in neutral grays.
Select shoots include downloadable DNG references (noted in the article). The weekly cron expands this library.
CIE colorimetry, IEEE 1789 for temporal light modulation, and NPL guidance on spectral radiometry — linked in every long guide.