The yield math behind the shift to chiplets
Splitting a die improves yield until the packaging becomes the bottleneck.
A large monolithic die loses more of each wafer to defects. Four smaller dies recover most of that, which is why the industry moved.
Subscriber story
Continue reading with a subscription
This piece is part of the subscriber archive. Subscribe to the daily brief for full access — one email, no card, cancel any time.
Already a subscriber?
#repairability#memory#power
More in this section
🔒hardwareSubscriber
Nvidia's next-gen interconnect reportedly doubles bandwidth per lane
Early silicon samples point at a redesign of the die-to-die fabric — and a supply chain that isn't ready for it.
🔒hardwareSubscriber
Benchmark leakage suggests a 20% IPC uplift, with caveats
Numbers from an unreleased engineering sample. We stress-tested the assumptions behind them.
🔒hardwareSubscriber
Thermal paste myths, tested with actual instrumentation
We ran 240 mounts across three coolers. The results are boring and therefore useful.