2 ms·
Nintendo chooses to use old components in their hardware, though. It's not like they don't have the industry contacts to source more powerful stuff if they want
by derefr 6y ago
Nintendo chooses to use old components in their hardware, though. It's not like they don't have the industry contacts to source more powerful stuff if they want. They just want to have a positive profit margin on selling console hardware. They want hardware sales (and sales of first-party "killer app" software enabled by the hardware) — not third-party software license fees — to be their main revenue stream.
Each console or handheld Nintendo puts out follows a formula:
* Choose key components (CPU cores, screens on portables, etc.) that are two-to-five year-old, one-or-more generations-behind tech at design time, such that they're in wide manufacturing and easy to source as parts / cheap to license to build themselves.
* Add in novel in-house-developed IP cores/ASICs, which allow developers to build games that "do" a lot more than the key components would seem to allow.
* Wrap the whole thing in a novel interaction method that makes the console's hardware performance relative to its "console generation" kind of illegible (e.g. the Wiimote, the two screens of the DS, the portability of the Switch, etc.); which de-facto locks in games developed to target the hardware, making them much harder to port away; and which acts as a neato whizz-bang selling point in the form of enabling new "killer apps" (e.g. 3D first-person games with the N64 analog stick; Wii Sports with the Wiimote; Brain Training and other pre-smartphone "apps" with the DS stylus.)
Nintendo are at their best (i.e. most profitable) when they double down on this strategy, releasing a seemingly-new hardware "generation" that's really just the same existing hardware but with new/updated IP cores. The Gameboy Color is just an overclocked Gameboy with a better PPU. The Wii is just an overclocked Gamecube with an ARM coprocessor embedding wi-fi/bluetooth/SD card support. Etc.
The Switch is a departure from this strategy, in that the Tegra X1 wasn't actually all that old at the time of the Switch's release, and so likely bit into Nintendo's profit-margin on the Switch pretty heavily. I'm sure Nintendo would rather have used a slightly-older part. But there just wasn't anything like the Tegra X1 (in terms of GPU power per watt) until the Tegra X1 — no viable competitors, and no previous generation of the chip to lean on. And they were constrained by wanting to port Wii U games to the Switch, which demanded basically the X1's level of power.