Withered Technology (Lateral Thinking of Seasoned Technology)
Win on cleverness with cheap, mature parts instead of chasing the priciest chips.
- Difficulty
- Moderate
- Time to result
- ~months to results
- Steps
- 5
- Confidence
- 85%
Coined inside Nintendo by engineer Gunpei Yokoi, 'withered technology' is the principle of applying well-understood, inexpensive, mass-produced technology in creative new ways rather than racing for the newest, most expensive components. Nintendo's NES used a cheap, off-the-shelf CPU but paired it with a custom, purpose-built Picture Processing Unit dedicated to graphics, an early example of specialized co-processing. The result was a machine years ahead of rivals in gaming capability while staying under Nintendo's mandated retail price and remaining profitable per unit sold, rather than subsidized like competitors' consoles.
Origin
Yokoi first applied the philosophy to the Game & Watch line, repurposing cheap calculator LCD technology into a wildly profitable handheld gaming product, before the principle was carried into Famicom and NES hardware design.
Core principles
- 01Innovation is about creative application, not component novelty
- 02Match the priciest component budget only to the function that most differentiates the product
- 03Specialized, narrow-purpose chips can outperform general-purpose ones for a specific job
- 04Negotiate volume commitments to push mature-tech unit costs down further
How to run it
- 1
Identify the one function that truly differentiates the product
For Nintendo, that was graphics quality, since gameplay and story were the differentiators consumers cared about, not general computing power.
- 2
Choose mature, cheap, well-understood core components
Use an off-the-shelf, already-depreciated-cost CPU instead of the newest chip on the market.
- 3
Build a specialized companion processor for the differentiating function
Design a dedicated Picture Processing Unit with its own memory to handle graphics rendering, offloading that work from the general CPU.
- 4
Lock in a hard cost ceiling before development starts
Mandate a sub-$100 retail price years before launch, forcing the engineering team to work backward from cost rather than spec first.
- 5
Commit to volume to unlock supplier pricing
Place a multi-million-unit order with your chip supplier up front to get per-unit pricing low enough to hit the target retail price.
In the wild
Thanks to the dedicated PPU paired with a decades-old CPU, the NES could run near-perfect ports of arcade games like Donkey Kong, something rival consoles like the Atari 2600 had to visibly dumb down because their hardware lacked dedicated graphics processing.
→ The NES became the first affordable home console capable of arcade-quality graphics, cementing its dominance over consoles built with pricier, less specialized chips.
Common mistakes
Chasing spec sheets over player experience
Competitors trying to out-spec each other on raw CPU power missed that specialized graphics handling mattered more for the actual product experience.
Underestimating volume needed to hit price targets
Without committing to millions of units upfront, Nintendo would never have gotten its supplier's pricing low enough to hit its retail price target.
Is it for you?
Best for
Hardware and product teams operating under strict cost or component constraints.
Not ideal for
Products where raw computational power genuinely is the differentiator, such as scientific computing.
From the transcript
“It's not about using the state of the art chips. It's not about using the most expensive hardware. It's about how clever and inventive and…”
“The PPU that Nintendo designs and develops is its own programmable chip that sits next to the CPU within the NES, and it has eight…”
From the episode
Nintendo's Origins