An Asus laptop that stopped booting, with the SSD appearing in the BIOS as a one-megabyte device. Alarming to see, and far better news than it looks — because no drive is 1MB, and the number is the controller telling you something specific.
The laptop would not start. In the BIOS the drive was present but reported a capacity of 1MB, and no operating system, adapter or other machine could make anything of it. On it were documents, photographs and work files with no backup anywhere.
No SSD is manufactured at one megabyte. When a BIOS reports 1MB, 2MB or 8MB, it is not reading the drive’s true capacity — it is reading a placeholder the controller is deliberately presenting. The specific figure varies by controller vendor, which is why the same fault is known by different numbers depending on whose silicon is inside.
It is a fallback state, and the sequence that produces it is worth understanding.
An SSD controller has a small amount of code in permanent memory that runs at power-up. Its job is to load the real firmware, which lives on the NAND itself, and then read the translation tables mapping logical addresses to physical flash pages.
If either step fails — the firmware image damaged, the translation tables inconsistent after a power event — the controller has a choice. It can present user data it cannot vouch for, or it can refuse. Well-designed firmware refuses, and rather than disappearing entirely it announces itself at a nonsense size: alive, identifiable, not ready.
The practical consequence is good news. The flash holding your data is generally untouched. This is a firmware condition rather than wear-out, and it is among the more recoverable SSD faults.
Manufacturers build a service mode into their controllers that bypasses the normal interface entirely. Reached that way, the drive can be interrogated directly and the reserved service area read — including those translation tables.
Those tables are the whole job. An SSD does not store data at fixed physical locations: wear levelling moves blocks around constantly, so the physical arrangement bears no relation to the logical one. Without the map, the NAND is undifferentiated data and reading every page reconstructs nothing.
With the map recovered, the drive images read-only exactly as though it had never failed. That is why this took three days rather than three weeks — once the controller was talking, the recovery was conventional.
Removing the NAND packages and reading them directly is a real technique and it is a last resort rather than a first move, for three reasons.
It is destructive to the drive and slow. It requires the translation layer to be reconstructed from scratch afterwards, including undoing the scrambling and interleaving the controller applies across multiple chips. And on many modern drives it returns encrypted data, because the controller encrypts everything written to the flash whether or not the user enabled anything — so a technically perfect chip-off can produce a bit-exact copy of something nobody can read.
Reviving the controller avoids all of that, because the controller performs its own decryption and address translation as part of ordinary operation.
Where a firm reaches for chip-off early on a drive whose controller could have been revived, the job is being made harder, slower and more expensive than it needs to be. It is a fair question to ask which route is proposed and why.
About 98%, returned on fresh media, three days from arrival. The shortfall was a small number of blocks the controller itself reported as unreadable — genuine flash degradation rather than anything to do with the fault that brought it in.
One instruction that matters more than the rest: do not apply a firmware update to a drive showing a nonsense capacity. It is the obvious thing to try, manufacturers publish the tools, and on a healthy drive it is routine maintenance. On a drive already in this state it rewrites the exact service area a recovery needs to read, and it can make the condition permanent.
Stop using the machine, note the make and model — controller work is manufacturer-specific and the model determines whether service access exists at all — and get it assessed. SSD and NVMe recovery is from £300 +VAT after a free 48-hour diagnostic.
Send your device in for a free diagnostic, with a few words about what went wrong — an engineer will look it over and put your exact quote in writing before anything is started.
The route to your data starts with the device reaching us. Box it up securely, tuck your contact details inside, and post it across — after the free diagnostic, we’ll put your exact price in writing before a thing is done.
Posting it? Tracked and insured, please. Bringing it in? Weekdays, 9am to 5:30pm — still packed as above for the trip.
After a little more detail up front? Complete the form describing your issue and an engineer will assess it and send back a tailored quote.
We’ll come back to you soon. If it’s urgent, call 0161 871 0788.
Almost certainly not. No drive is manufactured at that size — it is a placeholder the controller presents when it cannot load its firmware or mapping tables. The flash holding your data is usually untouched behind it.
It is a convention rather than a measurement. Different manufacturers use different placeholder capacities — 1MB, 2MB and 8MB are the common ones — as a way of saying the drive is alive but not ready to serve data.
No. An update rewrites the service area that a recovery needs to read, and on a drive already in a fallback state it can make the problem permanent. The same applies to secure erase and any manufacturer “repair” utility.
A diagnostic mode built into the controller by its manufacturer, which bypasses the normal interface. It allows the service area and translation tables to be read directly, so the drive can be imaged even though no computer will mount it.
Because it is slower, riskier and often pointless on modern drives — many controllers encrypt data at rest, so a chip-off returns a perfect copy of something unreadable. Reviving the controller lets it do its own decryption and address translation.
Where the controller can be revived, typically a few days. This one completed in three. Cases needing direct flash reads take considerably longer, which is another reason the controller route is preferred.
Begin with an instant online quote, or ring us and talk it over first. Either way you’ll know a clear, fixed price before any work starts.