[HH] anyone have experience with fake flash (microSD) memory?

[Kurt L Keville]

Was just reading about this topic... Bunnie Huang suggests that much if not most SD media is made after hours when the boss has gone home...

http://www.bunniestudios.com/blog/?p=3554

-----Original Message-----
From: hardwarehacking-bounces+kkeville=mit.edu at blu.org [mailto:hardwarehacking-bounces+kkeville=mit.edu at blu.org] On Behalf Of Bill Bogstad
Sent: Monday, July 14, 2014 3:30 PM
To: hardwarehacking
Subject: [HH] anyone have experience with fake flash (microSD) memory?

I've recently been adding (micro) SD cards to various android devices
and decided to go all the way with 32GB class 10 cards.   For some reason,
I remembered hearing about fake products a few years ago and decided that
I should test my new purchases.   I found that there are two programs out there

h2testw for Windows:

http://sosfakeflash.wordpress.com/2008/09/02/h2testw-14-gold-standard-in-detecting-usb-counterfeit-drives/

http://www.heise.de/download/h2testw.html

and f3 for Linux/OS X:

http://oss.digirati.com.br/f3/

They both work by filling up the drive to its maximum capacity with psuedo-random data and then reading the data back to compare what was written with the same
stream of pseudo-random data.   Interestingly, the most
recent version of the f3 program now uses the same file format/pseudo-random data stream, so you can fill a disk with h2testw and verify it with f3read.

Unfortunately, one of the three microSD cards that I
purchased in the last month or two is in fact a fake.   The
verify log from h2testw was as follows:

Warning: Only 31978 of 31983 MByte tested.
The media is likely to be defective.
7.7 GByte OK (16344913 sectors)
23.4 GByte DATA LOST (49146031 sectors)
Details:0 KByte overwritten (0 sectors)
0 KByte slightly changed (< 8 bit/sector, 0 sectors)
23.4 GByte corrupted (49146031 sectors)
0 KByte aliased memory (0 sectors)
First error at offset: 0x0000000002067800
Expected: 0x0000000002067800
Found: 0x0000004002067800
H2testw version 1.3
Reading speed: 2.51 MByte/s
H2testw v1.4

The bad microSD card was labeled as an ADATA card.
A different ADATA and a G-Skill card, both turned out
to be fine.   The bad card was purchased from the
Amazon Marketplace while the good one was
purchased directly from Newegg.   All three of the cards
cost between $14 or $15 dollars, so there was no way to tell by the price.

I'm still figuring out how to pursue this with Amazon and if anybody has experience with this I would appreciate it
hearing about it.   Assumming that Amazon doesn't want
the card back for verification, would any real "hardware hackers" be interested in it?

Thanks for reading this far,
Bill Bogstad
_______________________________________________
Hardwarehacking mailing list
Hardwarehacking at blu.org
http://lists.blu.org/mailman/listinfo/hardwarehacking