Dynamic random-access memory: Evolution
Timeline
The milestones from first idea to today
| Year | Milestone |
|---|---|
| 1966 | After an IBM review conference in the autumn, Robert Dennard wonders whether a capacitor could replace core memory's magnetic ring [3] |
| 1967 | Dennard records the one-transistor cell and files his patent on 14 July [1][3] |
| 1968 | Lee Boysel builds a 256-bit dynamic RAM at Fairchild; Dennard's patent is issued [1][2] |
| 1970 | Intel's 1103, a 1024-bit DRAM at 1 cent per bit, starts to replace magnetic core memory [1][2] |
| 1974 | 16K DRAMs follow 4K chips of 1973 [2] |
| 1976 | Mostek's 16K MK4116 adopts the single-transistor cell [1] |
| 2024 | Server memory and HBM lift DRAM sales above 28 billion US dollars in the last quarter [5] |
as of 2026-11-10
Lineage
The technologies it came from and the ones that came from it
Builds on- MOS MOSFET Each DRAM cell uses one MOS transistor as its switch [4].
- IC Integrated circuit DRAM is an integrated circuit with millions to billions of identical cells on one chip [4].
- Fl Flash memory Flash keeps data without power but is slower; computers use DRAM for working memory and flash for storage.
- PCo Personal computer Cheap semiconductor main memory, standard in almost all computers by the mid-1970s, is counted as a step toward the affordable personal computer [4].
as of 2026-11-10
Sources
- 1970: MOS Dynamic RAM Competes with Magnetic Core Memory on Price Computer History Museum, The Silicon Engine
- 1970: Semiconductors compete with magnetic cores Computer History Museum, The Storage Engine
- Thanks for the Memories IEEE Spectrum (Sally Adee), 2009
- Robert Dennard Lemelson-MIT Program
- TrendForce Findings: Server DRAM and HBM Continue to Drive Growth StorageNewsletter, reporting TrendForce, 2025