Adobe Lightroom Classic 12.3

The most significant addition in version 12.3 is the feature, which utilizes machine learning to remove digital noise while preserving intricate details that traditional manual sliders often smear.

Users can now use the Point Curve tool within local masking adjustments for precise control over contrast and color in specific areas.

Manipulate individual Red, Green, and Blue channels locally to correct color casts or apply creative color grading to specific parts of an image. 4. Workflow and Interface Tweaks

At launch, the technology works specifically on Bayer and X-Trans mosaic raw files, which covers the majority of modern mirrorless and DSLR cameras from brands like Sony, Canon, Nikon, and Fujifilm. Adobe Lightroom Classic 12.3

Between 2021 and early 2023, Lightroom Classic users suffered from a frustrating phenomenon known informally as "rubber banding." When moving sliders (especially Texture, Clarity, and Dehaze), the image would lag, overshoot, and snap back. It was infuriating.

Released in April 2023, marked a significant turning point in Adobe’s flagship desktop editing software. While many updates focus on minor performance tweaks, version 12.3 brought a game-changing feature that users had requested for years: AI-powered noise reduction.

To run 12.3 smoothly, especially the AI features: The most significant addition in version 12

Click the new button (located under the Manual Noise Reduction section).

For any active Lightroom Classic subscriber, updating to version 12.3 is a no-brainer.

: This flagship feature uses artificial intelligence to remove noise from high-ISO RAW images while preserving detail. It creates a new, separate DNG file (e.g., filename-Enhanced-NR.dng ) next to the original. It was infuriating

Despite the excitement, the release was not without its quirks.

Ensure your GPU drivers are updated to the latest studio or stable release.

Click . Lightroom will generate a new, clean DNG file alongside your original image.

Adobe addressed significant backend performance: