The Indilinx-based drives would usually just disappear, never to be detected again. Total data loss in all cases. I could often bring the drive itself back to life using the programming mode jumper (no data is accessible in this mode, but it would detect) and a downflash to 1.1, but that's a destructive flash and would sometimes only last a day or two before the drive would just fail again.
The Intel failure presented with several weeks of slowness and hesitation, causing the user to make repeated complaints. We were able to back up all data and RMA'd the drive. -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Zulfiqar Naushad Sent: Tuesday, February 28, 2012 12:31 PM To: [email protected] Subject: Re: [H] Bit the bullet on ssd What were the symptoms during regression to failure. Sent from my iPhone On Feb 28, 2012, at 9:26 PM, Bryan Seitz <[email protected]> wrote: > Agreed +100 on the Intel drives, any of them. > > On Tue, Feb 28, 2012 at 10:58:49AM -0600, Greg Sevart wrote: >> What drives were they? Anything based on the first gen Indilinx >> controllers (e.g., OCZ Vertex 1) were garbage. We bought around 50 of >> them at work, and I think they've almost all failed. The Intels have >> been rock solid (one failure out of around 50, and all data was >> recoverable), and I wouldn't feel uncomfortable with Marvell or >> Samsung based drives either. I'm still a little wary on SandForce >> after seeing a number of reports, but they're infinitely better than >> the Indilinx-based drives. Intel qualified them for their new 520 SSD, but only after a year of validation and custom firmware. > > -- > > Bryan G. Seitz
