Supporting the diagnostic accuracy of 24bitColors, About logic based on science and statistics.
Diagnoses by AI or algorithms tend to be opaque black boxes where "why that result was obtained" is unclear. We value satisfaction and transparency regarding results by visualizing the logical structure of colors and optimizing the process by which only your color is derived.
The RGB color space used in general Web design actually deviates from human senses. 24bitColors adopts "OKLCH", the latest color space closest to human perceptual characteristics. This makes it possible to quantify "brightness" and "vividness" as felt by the senses and analyze even delicate nuances that could not be captured by conventional diagnosis.
We don't just ask predetermined questions in order. The system analyzes your answers in real time and instantly calculates "which colors to compare next to narrow down preferences most". Using hundreds of "observation points" placed in the 16.77 million color space, we statistically analyze your preference trends and adaptively approach the optimal color region with each selection.
This algorithm is not the final form. We adjust logic daily in response to feedback from users such as "I want to choose more achromatic colors" or "Questions are too similar". Your diagnosis results and impressions directly lead to accuracy improvement for someone else. Our exploration of color continues to evolve with you.
Yes. We calculate the 'question that narrows down candidates most' for each step, enabling efficient identification with few questions.
It may change slightly. However, drastic changes (e.g., Red to Blue) are rare; usually, it stays within similar hues or tones.
Absolutely. The result page provides color codes and harmonious schemes useful for web and graphic design.