It does not detect them at all. Or crashes on reading the MBR.JonD wrote:Hi mimi,
not broken, but perhaps not quite right. It passes the utility that is supposed to check for fake SD cards, so nothing is actually broken, but even Linux distros on this card seem to freeze for several seconds at some points when writing. The test didn't freeze, but linux does.
Just out of interest, what is the issue with RISCOS and bad SD cards? Does the filesystem not repair itself well?
It didn't say bad SD card, but bad SD card *slot*!JonD wrote:what is the issue with RISCOS and bad SD cards?
Oh? Is that a "doesn't work AT ALL with class 2 cards", or a "works so slowly that it's unbearable"?mimi123 wrote:It's worth noting that RiscOS doesn't work with class 2 sd cards.
On RISC OS, timeout issues so doesn't work at all.AndrewS wrote:Oh? Is that a "doesn't work AT ALL with class 2 cards", or a "works so slowly that it's unbearable"?mimi123 wrote:It's worth noting that RiscOS doesn't work with class 2 sd cards.
I always thought that the host OS had no 'knowledge' of the SD class, and (ignoring the UHS modes) the cards operated identically, just taking longer to return data from a read/write call?
(trying to learn as much as I can about the different quirks of SD cards, because of e.g. https://github.com/raspberrypi/noobs/issues/174 and https://github.com/raspberrypi/documentation/pull/88 )
Yes. Exactly that. So RISC OS can be fixed for the OP but RISC OS is propretary with the source code. Just read the Castle license.AndrewS wrote:Ahh, so if the OS was modified to have bigger IO timeouts, then it would work on class 2 cards?
Not that I want to run RISC OS on class 2 SD cards, just curious