Disclaimer: I don’t believe Lenovo had anything to do with this card. I think this is an unlicensed knock-off — hence why “Lenovo” is in quotes.
Ok, let’s be clear: I didn’t want to test this card. I had already one of the 128GB and one of the 2TB variants of this card and knew that they were fake and performed poorly. However, my decision to test at least three samples of each card came back to bite me in the ass here: when I ordered two more of these cards, the seller actually sent me two 256GB cards instead.
Well crap.
So…I went to a different seller and got two 128GB cards. But what to do with these cards? I know they’re fake. I know I have no use for them outside of this project. Might as well get a third one and add it to my test set.
This card bears a number of hallmarks of fake flash: (a) they bear the name of a well-known electronics manufacturer, even though that manufacturer is not known for selling flash memory (in this case it’s Lenovo, but I’ve seen Sony and Xiaomi caught up in the crossfire here as well); (b) two samples had the manufacturer ID and OEM ID zeroed out from the CID register in an attempt to mask their true origin; and (c) performance across the board was pretty abysmal. This particular seller also attempted to be sneaky by not including any of the standard speed class marks. Instead, they opted to include the Nintendo Switch icon — which, last I checked, wasn’t an indicator of speed. (They did include the UHS-I mark, but this mark by itself also doesn’t make any assertions as to the card’s read or write speeds.)
On the endurance testing front:
- Sample #1’s first error was a series of bit flips during round 378. However, once it got to round 399, it began experiencing I/O errors on pretty much every sector I tried to read from. In this instance, the card wasn’t completely dead — but I ended up letting endurance testing go on far too long. It stayed stuck in this round for probably 8 months before it finally decided to die.
Sample #2’s first failure was a series of bit flips during the very first round of endurance testing. It only made it 95 read/write cycles before hitting the 50% failure threshold. Here’s what this sample’s progression looked like:
Sample #3’s first failure was a series of bit flips, that affected two adjacent sectors, during round 1,496. It survived until round 1,912, when all sectors started reading back as all
0xff‘s. Here’s what this sample’s progression looked like:
Overall? They’re fake flash, they scored below average in just about every metric, and not a single one of them made it to the 2,000 read/write cycle mark without major issues. They’re trash — don’t buy them.
June 10, 2025