Your archived photos may be slowly disappearing – but Lexar wants to do something about it

Last month I travelled to China as part of a small group of European journalists on an exclusive press trip hosted by Lexar – the storage brand behind some of the most widely used memory cards and SSDs in photography, now owned by Longsys and this year celebrating its 30th anniversary. Over the course of nearly a week, we were given extensive access to the company's operations: production facilities in Zhongshan and Suzhou, innovation and testing labs in Shenzhen. But the moment I keep coming back to didn't happen in a factory or a lab. It happened in a press conference room in Shenzhen, prompted by a question that had nothing to do with speeds, capacities or market share.

A fellow journalist was thinking about a physical photograph she owned: her great-grandparents' wedding, printed on paper, more than a century old and still entirely legible. She wanted to know whether her own photographs would be accessible to her great-grandchildren. Whether Lexar was thinking about storage technologies built for genuine longevity. The engineers in the room gave her an answer she probably wasn't expecting.

The back story

First, some context. On the second day of the trip, we'd been taken to the Memory Museum; a small but carefully assembled exhibition within the Zhongshan Storage Industrial Park that traces the history of flash memory from its earliest commercial incarnations to the present. It was actually more fun and interesting than you might imagine.

CompactFlash cards the size of small biscuits. USB drives that look like props from a 1950s sci-fi film. The first Lexar products, sitting behind glass like ancient artefacts. It's the kind of display designed to make you feel the pace of progress. But it also prompted a startling thought.

All of those early formats are, for practical purposes, unreadable now. Not because the data degraded, but because the interfaces to access it no longer exist in any mainstream device. The cards themselves might be perfectly intact. But the photographs on them are simply inaccessible without hardware that no one makes any more.

That's one kind of digital preservation problem. The question raised in that press conference pointed to a different and perhaps more immediate one.

The technical challenge

The NAND flash used in virtually all consumer memory cards today comes in several varieties, distinguished by how many bits of data are stored in each memory cell. The most common types in current consumer products – TLC (triple-level cell) and QLC (quad-level cell) flash – offer more storage at lower cost than older formats, which is why they dominate the market. But they carry a big limitation that doesn't feature prominently in any product's marketing material: when left unpowered, they can begin to lose data within as little as one year.

The mechanism is worth understanding briefly. Flash memory stores data as electrical charges in tiny cells. Yet those charges are not permanent: they leak away slowly over time, like air from a tyre. While a card is connected to a powered device, its firmware can monitor the charge levels and top them up before they fall below the threshold needed to read the data correctly. When a card is in use, this happens automatically. When a card is sitting in a drawer, it doesn't happen at all.

SLC (single-level cell) flash, which stores one bit per cell and is far more robust, extends that window to around a decade under good conditions. But virtually no consumer product uses SLC flash throughout, because the cost and capacity trade-offs make it commercially unworkable.

I sat in that press conference room in Shenzhen – a building I'd been shown around that morning, full of test equipment worth tens of millions of renminbi, designed to push memory to its limits – and absorbed the implication of what was being said. The cards in your camera bag, the drives on your desk, the microSD sitting in your drone's slot right now? All of them are on a clock that you may never have thought to look at.

Risk to the archive

The engineers were careful to contextualise this, though. In normal use (cards regularly connected to cameras, computers, and phones) the firmware does its job and the risk is low. The problem is specifically with archival storage: cards or drives that are left unpowered for extended periods because their contents are considered finished and safe. That is precisely the use case that most photographers would describe as "backing up"... and it's the one where flash memory is most vulnerable.

Lexar said, though, it was developing a direct response to this: a physical device designed to periodically refresh the data on stored cards and drives, re-writing it before the charge levels can degrade to the point of data loss. You would connect your archived media to this device every few months, or perhaps set it to run automatically, and it would cycle through everything stored on it, effectively resetting the retention clock.

No release date or pricing was given: I get the sense they're a long way off this. But the fact that a company of Lexar's scale is building such a product tells you something about how seriously the engineering teams take the underlying risk.

Key takeaway

Walking back through the Memory Museum later on, past the obsolete memory devices in their glass cases, stretching from ancient times to the present day, the point landed differently than it had the first time. Those old cards aren't just inaccessible because the interfaces changed. Some of them, if they were left on a shelf long enough, may simply no longer contain what they used to. The data didn't go anywhere dramatic. It just slowly stopped being there.

The photo of someone's great-grandparents, printed on paper and stored in a shoebox, will almost certainly outlast any flash drive currently on the market. Not because it's higher technology, but because it doesn't need power to remain legible. Analogue media degrades, certainly, but it degrades passively and predictably and slowly. Digital storage requires active maintenance to stay intact, and most of us have been treating it as though it doesn't.

The practical upshot? Treat memory cards as working storage rather than archival storage. Copy your files to at least two separate locations, and make sure at least one of those locations is connected to power and actively maintained. If you do keep data on flash media long-term, plug those cards or drives into a computer periodically, even just for a few hours, to give the firmware a chance to do its job.

If Lexar's data refresh device makes it to market, it would make this process significantly more straightforward. Until then, the responsibility falls on the photographer. In short, your great-grandchildren's ability to see your work may depend less on the quality of your camera than on whether you remembered to plug in your hard drive.