The best desk lamps for a home office

This is a two-lamp list, and before anything else it is worth saying why it is a two-lamp list: almost no desk lamp publishes the numbers this page ranks on. At the budget end, the category is a wall of lamps with no CRI figure, no Kelvin range, and no rated life anywhere on the box or the brand’s own site. There is nothing to compare. We could pad this page to eight entries by copying retailer bullet points, and it would be the same page every other site already published.

So: two lamps that actually publish specifications, ranked on colour temperature, CRI and rated life. And the result is more interesting than we expected.

The lamp with the lower CRI won

Dyson publishes “CRI 90 or above.” BenQ publishes “Ra≥95.” The Dyson costs roughly two and a half times as much. On the sticker, that looks indefensible — you are paying a large premium for a smaller number.

Except CRI is a much weaker specification than the marketing around it suggests. Ra is an average of eight low-saturation test colours, and it excludes R9— the deep-red sample. R9 is the one that governs how skin tones, wood and anything red actually render. A lamp can average Ra95 across the eight and still handle red badly, and nothing in the Ra number will warn you.

Dyson is the only lamp on this entire site that publishes R9 at all: 48 minimum. That is not a flattering number, and they printed it anyway. A published 90+ with a stated R9 tells you more than an unpublished 95 with the awkward sample averaged out of the frame.

The deeper problem belongs to CRI itself. The CIE — the international body that definedCRI — has published a position statement saying Ra “does not agree well with overall perceived colour rendering” for LED sources, and that the metric needs replacing. CRI was first published in 1965 and last improved in 1974. Every Ra sticker on a 2026 lamp is a 52-year-old formula that its own authors have said in writing is not fit for the light source it is printed on. The statement is free to read and it is in our sources below.

The longevity numbers are not the same kind of thing

This is the part that decided the ranking, and it is the clearest example on the site of why we bother.

Both brands express lifespan the same way — years at eight hours a day. BenQ’s Genie page says the lamp “can stay on for up to 17 years, 8 hours per day.” Dyson says its LEDs maintain light quality for at least 60 years. Superficially these are the same class of claim.

They are not. Dyson states the figure as 181,000 hours, names the metric (L70— the point at which output has faded to 70% of original), names the standard (IEC 62717), and names the assumption (8 hours per day). Every element of that claim can be checked, including the arithmetic: 181,000 ÷ (8 × 365) = 62 years, which supports a marketed 60. It holds up.

BenQ’s 50,000 hours comes from a single blanket FAQ that covers the lamp range and cites nothing at all. No standard, no metric, no method. Run BenQ’s own “17 years at 8 hours a day” and you get 49,640 hours — so the 17-year claim and the 50,000-hour claim are one unsourced figure stated twice, not two pieces of evidence.

We are not accusing BenQ of anything. 50,000 hours is an entirely ordinary figure for an LED lamp and it may well be conservative. The point is narrower and it is not about BenQ: one of these two claims can be verified, and one cannot, and you deserve to know which is which before you weigh them against each other.

What a cited standard does not buy you

Worth being careful here, because it would be easy to overclaim on Dyson’s behalf. L70 measures lumen maintenance — how long until the light fades — not when the lamp dies. The US Department of Energy’s LED Systems Reliability Consortium says this class of data “can predict lumen depreciation but not lifetime,” because the projection assumes the driver, power supply and electronics never fail. In real luminaires, those are frequently what fail first.

And a caveat on our own caveat: IEC 62717 is a different standard from the LM-80/TM-21 pairing that guidance is usually aimed at, and we have notverified its extrapolation limits. So we will state the general warning and stop there. We are not alleging Dyson has stretched anything — we have no evidence of that and would not print the insinuation without it.

The colour temperature gap is real

The third published spec is the one people skim past, and it is the only place on this page where the two lamps differ in what they can physically do rather than in what they tell you.

Both start at the same place: 2700K, which is a warm, lamplight-coloured white. The difference is the top. Dyson reaches 6500K; the BenQ Genie stops at 5700K.

That 800K sounds like a rounding error and is not. 6500K is roughly the colour of indirect daylight from a north-facing window — which is the light most desks are set up around during the day, and the light this site is named after. If you work beside a window and you want your lamp to sit alongside that rather than fight it, 5700K cannot get there. Your desk ends up with two different whites on it, and the warmer one reads as slightly yellow by comparison. The Genie is a 13-step lamp across a genuinely useful range; it simply has a ceiling, and the ceiling is below daylight.

Going the other way, Dyson publishes a second, separate range: 1800–3400K when docked. That is much warmer than anything the BenQ offers — candle-ish at the bottom — and it is a different mode rather than an extension of the main range, intended for when the lamp is parked rather than lighting a task. Whether that is genuinely useful or a feature looking for a job, we cannot tell you. We have not switched it on.

One thing we are deliberately notdoing here. BenQ’s own page attaches a claim to its warm end about the effect of light on melatonin. That may well be sound — there is real science in that area — but it is a physiological claim, we are not qualified to evaluate it, and repeating a manufacturer’s health claim as though we had checked it is exactly the move this site exists not to make. It is BenQ’s claim. We are telling you they make it, and stopping there.

Why the famous BenQ isn’t here

The BenQ e-Reading lamp is the one people expect on a list like this, and it is deliberately absent. Its Amazon listing dates to roughly 2015, while BenQ’s live page now describes a newer “E-Reading Intelligent” revision. Those are not reliably the same lamp.

We could have put the current spec sheet next to the old ASIN and nobody would have noticed. That is exactly the mismatch this site exists to catch — you would have clicked through expecting the specs on our page and received whatever the decade-old listing actually ships. The Genie has no such problem: its listing and BenQ’s current page describe the same lamp, down to the 13 colour modes and the auto-dimming sensor. So the Genie is the BenQ we ranked.

What we can’t tell you

We have not switched either of these on. We cannot tell you whether the Dyson’s arm moves nicely, whether the Genie hums, how either handles a year of being knocked about, or whether the Dyson’s docked 1800–3400K mode is genuinely pleasant or a gimmick. Those need hands, and we do not have them — our methodology sets out exactly what that does and does not buy you.

What we can do is read both spec sheets properly and notice that the cheaper lamp’s headline advantage rests on a metric its own standards body has disowned, while the expensive one quietly publishes the number that metric leaves out. That is worth knowing before you spend anything.