The most accurate TVs in ZDNET's lab were an LG and a Hisense, not a Sony or Samsung

ZDNET ran seven flagship TVs through its Louisville lab and measured color reproduction with Portrait Displays' Calman software, and the two accuracy awards went to the LG C6 for SDR and the Hisense U8 for HDR ^[1][2][4]. The publication's own framing is that the winners were not Sony or Samsung ^[11], which is the part worth sitting with if you buy displays by brand tier or by the numbers printed on the box.

The measurement is straightforward and unforgiving. Color accuracy here is the gap between the color a pattern generator sends and the color the panel actually produces, scored as Delta E 2000 for SDR and Delta E ITP for HDR, with lower numbers better and zero being perfect ^[7]. The LG C6 posted a dE2000 of 1.84 in SDR ^[6]. That is the only per-model figure in the material available to us, so the sensible read is the ordering, not a scoreboard.

ZDNET split the awards because SDR and HDR are distinct formats that are calculated slightly differently ^[5]. It also measured a second thing, color gamut, defined as how much color a set can display as a percentage of the broadest gamut, rec.709 for SDR and rec.2020 for HDR ^[8]. Gamut is capability. Accuracy is whether the panel lands on the color it was handed. The awards were decided on accuracy ^[4], and those two questions do not have to answer the same way, which is exactly the gap a spec sheet built on coverage percentages leaves open.

Two winners across two categories means the other five sets in the field took nothing ^[13]. That field included Samsung, LG and TCL among the brands tested ^[1], plus the Hisense U8 ^[4], Sony sets, and a Sunbrite model ^[10].

There is a real asterisk, and ZDNET flags it. Winners were determined from Filmmaker mode, which the publication describes as generally the most accurate color mode on a TV ^[9]. The Sony sets and the Sunbrite did not have Filmmaker mode, so the lab tested each brand's equivalent ^[10]. That is not a like-for-like comparison. It is closer to a test of what a given manufacturer ships as its most accurate preset, which is arguably the more useful question for anyone who will never hire a calibrator, but it is not the same as holding the mode constant.

The operator conclusion is narrow and still useful. If out-of-the-box color fidelity is the requirement, whether that is a conference room, a retail wall, a video review station or a QA bench, the brand hierarchy is not a proxy for it, and neither is a gamut percentage. One lab, seven units, one mode per set, and the ranking already breaks the assumption. ZDNET discloses that it earns affiliate commissions on purchases made through its links, and says the arrangement does not influence coverage and that neither the publication nor the author is compensated for the reviews ^[12].

What to watch: whether ZDNET publishes the full per-model dE2000 and dEITP tables, since two names without the spread underneath cannot settle a purchase. Watch also whether the Filmmaker-mode gap closes, because as long as some manufacturers ship an equivalent instead of the standard preset, cross-brand accuracy tests stay approximate. And watch whether anyone selling into procurement starts quoting a dE figure rather than a gamut percentage.