Blue vs White Noise

Flat against rising: 0.00 dB per octave and +3.08. There is exactly one situation where the brighter of the two is the right answer, and most of this page is about how narrow that situation is.

Past Flat, and What Is Left Behind

White noise measured 0.00 dB per octave here, flat to ±0.08 dB. Blue noise measured +3.08 dB per octave, with ±0.43 dB of deviation from a straight line — it is generated by the same filter topology as pink noise with the poles and zeros exchanged, which makes blue and pink exact mirrors of each other about the flat line that white occupies.

Anchor the two spectra together at 100 Hz and blue sits about 10 dB above white at 1 kHz, about 16 dB above at 4 kHz and about 20 dB above at 10 kHz. But that framing flatters blue noise, because both signals are normalised to the same perceived loudness — −20 LUFS under ITU-R BS.1770-4 K-weighting — so there is no extra energy available from anywhere. Everything blue noise gains at the top it has taken from the bottom. The accurate description is not that blue noise is brighter than white; it is that blue noise has abandoned the lower half of the spectrum in order to be brighter.

That is the whole trade, and it is why this comparison resolves so quickly. White noise has full strength everywhere. Blue noise has more than full strength in a narrow region and almost nothing anywhere else.

It Is Piling Energy Where You Are Most Sensitive

The ear canal resonates, and the result is that human hearing is at its most acute somewhere around 2–5 kHz. This is not a subtle effect: two signals with identical raw levels, one weighted low and one weighted high, are not remotely equally loud, which is why every colour here is matched by K-weighted perceived loudness rather than by signal level. Without that step, switching to blue noise would be an unpleasant surprise every time.

White noise already places a great deal of energy in that sensitive region — it is one of the reasons white noise is described as sharp rather than as neutral. Blue noise doubles down on it. At the same measured loudness, blue delivers more of its content into the 3–6 kHz band than any other colour on this site.

Two consequences follow. The first is that blue noise reaches the fatiguing part of a sound faster than white does: if white noise wears on you after two hours, blue noise will do it sooner. The second is that blue noise is the colour where volume discipline is worth taking most seriously, since 3–6 kHz is the region most associated with noise-induced hearing loss. The risk from any sound is a product of level and duration rather than of colour — but at a given perceived loudness, blue is the one delivering the most into the sensitive band.

The One Job Blue Noise Does Better — and How Narrow It Is

Masking requires spectral overlap: a masker raises your detection threshold for sounds near it in frequency and does very little for sounds far from it. So the case for blue noise is exactly and only this — an intrusion that is high-pitched enough that even white noise is running out of relative strength up there.

The candidates are specific: coil whine from a laptop or a power supply, a squealing fan bearing, an ultrasonic-adjacent hiss from a display or a charger, a distant electronic chirp. If that thin, high, needling quality is what is bothering you, none of the darker colours will touch it, and blue noise is the correct tool.

Now the deflation, because it is more useful than the recommendation. White noise is at full strength across the entire consonant and whine region — 2 kHz, 4 kHz, 6 kHz — and for most high-pitched intrusions that is already enough. Blue noise's practical advantage does not really open up until the target sits above roughly 6 to 8 kHz, and above that the intrusion is often better solved at source than masked at all. In other words, the honest recommendation for a high-pitched problem is: try white first, and reach for blue only if white leaves the whine standing out on top of it.

One genuine non-masking use is worth mentioning. Listening to blue noise immediately after brown noise is the fastest way to understand what spectral tilt actually means, and it makes the other four colours much easier to tell apart afterwards. As a calibration exercise for your own ears it takes about thirty seconds and it is worth doing once.

Why the Term Exists at All, and Why It Is Not About Listening

White noise and blue noise are both genuine terms of art, but they earn their keep in different fields, and knowing that explains why one of them turns up constantly in sleep writing and the other almost never does.

Blue noise's home is signal and image processing, where blue-noise dither patterns became the standard for digital halftoning after Robert Ulichney's 1988 work in Proceedings of the IEEE. The reasoning is elegant: when error has to be introduced somewhere — quantising a photograph down to black and white dots, distributing samples in a renderer — you want it concentrated at high spatial frequencies, because that is where the visual system is least able to resolve structure. The same principle drives blue-weighted dither at the quantisation stage in audio.

Notice what the argument is in every one of those cases: put the noise where the recipient will not notice it. That is close to the opposite of the proposition involved in deliberately playing a sound into your own ears at a level where you certainly will notice it. White noise, meanwhile, is both the raw material every other colour here is filtered from and the sound that has actually been studied as a sleep and attention intervention. The full account of what blue noise is for is worth reading if this is the part that interests you.

The Research Position: One Weak Base and One Empty One

White noise is the most-studied colour, and that still does not amount to much. Riedy and colleagues (Sleep Medicine Reviews, 2021) reviewed 38 articles on continuous white noise and similar broadband noise as a sleep aid and rated the quality of the evidence that continuous noise improves sleep as very low under GRADE, noting that this contradicts how widely it is used and raising the possibility that continuous noise may negatively affect sleep and hearing. For attention, Nigg and colleagues (Journal of the American Academy of Child and Adolescent Psychiatry, 2024) found a small benefit of white or pink noise on laboratory attention tasks in young people with ADHD or elevated attention problems, and no benefit outside that group.

Blue noise has none of this. No systematic review covers it, no trial has used it as a sleep or attention stimulus, and there is no reason to expect one — it is not a sound anyone has proposed as an intervention. A page that lists blue noise among sleep sounds with a description of its benefits is completing a set, not reporting a finding. Nothing here is medical advice.

On tinnitus the position deserves care rather than a recommendation. Tinnitus percepts are frequently high-pitched, and a bright masker overlaps with them where a dark one cannot — but some people report that a bright sound sitting directly on the percept makes it more prominent rather than less, and there is no way to predict from a web page which group you fall into. Most people using sound for tinnitus end up somewhere darker, and the page on masking and habituation sets out why the pitch of your tinnitus decides the colour and why raising the volume is the wrong response to a mismatch. Persistent or one-sided tinnitus is worth an audiologist's assessment rather than an experiment with a noise generator.

The Verdict, Which Is Not Balanced

For sleep, for long sessions, for speech, for household noise and for anything you intend to leave running: white noise, without hesitation. Blue noise has no low-frequency weight to settle behind, it fatigues faster, it has no research base at all, and it puts more energy than any other colour into the band your ears are most sensitive to.

For one narrow job — a high-pitched whine that white noise leaves standing — blue noise is the right tool and the only one here that will reach it. Use it in short, task-bounded stretches at a modest level, and put it down afterwards. That is a genuine use and it is not a bedtime sound.

Both are generated procedurally in the browser rather than looped, so a fair A/B takes about a minute; free sessions run 30 minutes. The free Android app carries all five colours including blue, so if it does turn out to be the answer to a specific machine in your house you can run it offline and on a timer like any of the others.

Frequently Asked Questions

Related pages

• Brown Noise for Sleep — The free web player, plus the full guide to brown noise for sleep, focus, and tinnitus.
• Blue Noise — The bright end of the spectrum at +3.08 dB per octave — an engineering signal that is genuinely useful against high-pitched noise, and a poor choice for sleep.
• White Noise — The flat reference colour, measured at 0.00 dB per octave — the best of the five at making speech unintelligible.
• Brown vs White Noise — The widest gap on the site — 5.98 dB per octave, about 38 dB apart by 8 kHz — and what each colour can and cannot cover as a result.
• Brown Noise for Tinnitus — Masking versus habituation, why the pitch of your tinnitus decides the colour, and where sound therapy actually sits in the 2014 clinical guideline.