The page is a historical price series for computer memory, normalized to dollars per gigabyte and shown on a log scale from 1960 through 2026. It mixes very old memory technologies with modern DRAM, NAND flash, and storage, and the main signal is obvious even before the comments: memory got exponentially cheaper for decades, then that trend bent hard in the 2010s and snapped upward recently. People kept coming back to two basics that many readers missed. First, inflation changes the early decades far less than the dramatic shape suggests because the chart is logarithmic. Second, using dollars per GB is not a category error. It is the standard way to compare a commodity resource across time, even if it says nothing about how much memory a normal machine needed in a given year.
The useful criticism was elsewhere. Several readers pointed out that the recent DRAM points include older
DDR3 parts, even a 2 GB stick in 2025, which likely makes the present look cheaper than what buyers of current
DDR4 or
DDR5 systems actually face. That lines up with the broader complaint that this is a good chart for industry cycles but a weak chart for practical upgrade cost. If what you care about is building a current PC or provisioning developer machines, memory feels worse because software footprints, minimum viable configurations, and the floor on what counts as a usable system have all risen.
The discussion also converged on why the cycle looks so violent now. Memory has always been cyclical because producers add capacity together, crash prices, pull back, then repeat. What stands out this time is that demand from
AI and accelerator-heavy systems has stayed strong enough to keep prices elevated longer than people expected, while manufacturers remain cautious about overbuilding after prior busts. A few readers tied that to physical scaling limits in DRAM itself. Unlike logic chips, DRAM is constrained by tiny capacitors that are getting harder to shrink, which helps explain why the old decades-long decline has slowed. The net read was not that RAM is back to 1960s absurdity or even necessarily back to 2010 in real terms. It was that a resource everyone had treated as steadily cheaper is acting cyclical and scarce again, and that changes both hardware planning and software discipline.