The gaming laptop was supposed to die. It didn't.

Around 2020–2022, the consensus take in consumer tech writing was that the mid-tier gaming laptop — the $1,200–$1,600 RTX 3060/3070 machine — was going to be eaten by cloud. The reasoning was clean: why spend $1,400 on a portable rig when $10/month gets you better hardware in a Frankfurt datacentre?

Cloud-bullish analysts pointed to the streaming-music precedent: physical media revenue cratered as Spotify and Apple Music took share. The argument was that gaming hardware would follow the same trajectory, with cloud playing the role of streaming.

Gaming laptop unit sales in 2025 were higher than in 2020. Not dramatically — roughly flat to slightly up — but the trend is the opposite of the cloud-replacement thesis. ASUS, MSI, Razer, and Lenovo all expanded their gaming laptop lines in 2024–2026, not contracted them.

The mid-tier specifically held up well. RTX 4060 and 4070 laptops sold strongly through 2025, and the RTX 5060 refresh in early 2026 has been a top-three SKU at major US retailers. Cloud did not eat this segment.

The Spotify analogy was wrong in a specific way. Streaming music replaced a product (a CD or MP3 file) with an identical experience delivered differently. Cloud gaming replaces a product (a local GPU) with a similar-but-not-identical experience — there's added latency, there's a network dependency, there's a session model that doesn't behave like a local install.

For a music listener, Spotify is functionally a strict upgrade over a CD collection. For a gamer, cloud is not a strict upgrade over a local rig. Latency-sensitive players notice. Modders notice. Players who travel to spots with bad internet notice.

The forecast also overweighted the price comparison. $10/month for cloud sounds cheap until you realise that a $1,400 laptop also runs every other piece of software a person uses for the next four years. The all-in cost comparison is much closer than the monthly-vs-upfront framing suggested.

Cloud didn't displace the gaming laptop. It displaced the second gaming PC — the one that the gaming-laptop owner used to build for a travel setup, or for the kid's room. A lot of households that would have bought two gaming machines in 2018 now buy one machine and a cloud subscription.

Cloud also displaced the casual upgrade. The person who upgraded from a GTX 1660 to an RTX 3060 every three years now often skips the upgrade and runs cloud when they want to play something current. That's a meaningful share of the GPU upgrade market, but it doesn't show up as 'fewer laptops sold' — it shows up as 'fewer discrete desktop GPUs sold to upgraders'.

Be wary of any cloud-displaces-hardware forecast that uses streaming music as the precedent. Streaming video is the better analogy and it tells a different story — Netflix didn't kill the TV; it killed the DVD player. Cloud gaming is more likely to kill the second gaming PC than the primary one.

If you're a hardware analyst tracking the mid-tier laptop segment, the right question isn't 'when does cloud kill this category' — it's 'which adjacent categories does cloud cannibalise first'. Our reading: discrete GPU upgrades and entry-tier consoles, in that order. Mid-tier gaming laptops are safe for at least the next product cycle.

One scenario could change this: cloud gaming services shipping a high-quality 4K/120 streaming experience on existing Wi-Fi 6E hardware, paired with a workable answer for anti-cheat and persistent mods. If both happen in the same product cycle, the mid-tier laptop replacement argument gets stronger.

We're not seeing the second piece happen on any roadmap. The first piece (4K/120) is now real on premium tiers. So the answer-to-anti-cheat constraint becomes the load-bearing variable for the next five years of this question. Watch that, not the streaming quality.