CHF1999.–
Samsung QE85QN85D
85", QN85D, NeoQLED, 4K, 2024
Question about Samsung QE85QN85D
QN90D is clearly better <-> QN85D Ah, then the values on the shop page are not quite right, are they? ChatGPT: Exactly - the values on the shop page are formally correct, but: ❗️Sie conceal the crucial differences. Which is true: Both devices are "Neo QLED" → i.e. mini LEDs with full array local dimming. Both have 4K, Smart TV, HDMI 2.1, 2024 model year. 120 Hz vs. 144 Hz is stated correctly. Panel technology = QLED is correct (refers to the colour filter). What is missing (but crucial): Number and quality of dimming zones: QN90D has significantly more (→ better local dimming, black level, less blooming). Panel type: QN90D almost certainly has a VA panel with contrast optimisation, QN85D usually has an IPS-like panel with flatter black levels. Anti-reflective layer (Ultra Viewing Angle): only included in the QN90D. HDR peak brightness: QN90D ≫ QN85D, but not mentioned in the shop. Quantum HDR vs. Quantum HDR+: QN90D has the stronger version, but is not listed separately. 📌 Why is this information missing in the shop? Many retailers only use Samsung's standard data sheets or automatically fill in structured fields. Samsung itself does not make the differences transparent at first glance - deliberately so that marketing labels such as "Neo QLED" can be used widely. 🟢 Conclusion for you: You won't find the decisive quality differences in the shop comparison, but only via data sheet analysis, reviews or direct specs from Samsung US/KR/UK
Helpful answer
ChatGPT is right, as a rule all data sheets are recorded automatically and integrated into predefined fields.
In my experience, it is a widespread misconception that AI can be left to its own devices. I use AI very intensively and also profitably (triple rebound). But I have learnt one thing. Everything the AI generates, including product pages from the same source (Samsung, same line), must be edited.
This support effort, a kind of supervisor, for the AI probably has to be maintained for at least three years - until then it would be possible for the AI to correct itself.
Anything else leads to a loss of quality. --> See Duolingo.
The AI has a slow but clear tendency to become mired in its database. It's almost like a black hole. I wonder what you do about it. My only remedy is to edit and try to make the AI better at my tasks by giving it constant correction instructions.
Can you pass that on? It's relevant to the business. At least for the next three years, it is absolutely necessary to ensure professional editing for quality assurance. Otherwise, we can expect similar results as with Duolingo (deterioration in the quality of product descriptions - until you can hardly tell the difference between an electric toothbrush and a mopping robot - an experiment in Japan).