No, that's not true. As I said, it's a combination of factors.
CF-Root is a *Samsung* kernel and is thus (depending on version, and like many other kernels) vulnerable. What we have done is patched out the binary calls that may trigger the brick.
However, update-binaries built with a kernel source tree that do have the MMC_CAP_ERASE enabled can still trigger the brick.
On a truly safe kernel (full recompile from source, with the relevant code portions fixed) the bug cannot occur even with a faulty update-binary.
As stated before, this isn't PA's fault - BUT, PA can implement a really simple fix that will prevent this from happening again.
The "format" command from update-binary being able to trigger the bug on a not-fully-safe kernel has been known to devs on this platform for quite some time ...
