Doomer wrote:fzabkar wrote:I wonder if the reduced flash Vcc could explain why OCZ SSDs appear to be significantly less reliable than similar products from their competitors?
I don't think so
Moreover lowering voltage of NANDs could improve reliability
I've brought this subject up before, and I've explained why I think lowering the voltage could reduce the reliability of the SSD. You and your colleagues seem to focus on chip reliability, in which case it would be reasonable to assume that lowering the Vcc would lead to increasing chip life. But I'm focusing on SSD failures, not chip failures. There's a difference.
If you watch the storage forums, you'll notice that users generally report that their OCZ SSDs most often fail at power-up, and rarely in-flight. One consequence of a reduced Vcc would be much less time for emergency "housekeeping" after a power loss event, which in turn would make the Flash Translation Layer much more vulnerable to corruption. In this case the Vcc would only need to decay from 2.9V to 2.7V instead of the usual 3.3V to 2.7V scenario. In an earlier thread a Vertex 2 240GB SSD had a Vcc of 2.8V.
viewtopic.php?f=1&t=25650So do those two cases represent genuine failures, or are they an intentional design choice?
Doomer wrote:Also could you please provide statistics on what you based this conclusion "OCZ SSDs appear to be significantly less reliable than similar products". Such statement seems strange to me
Components returns rates:
http://www.behardware.com/articles/881- ... tes-7.html- 40.00% for the OCZ Petrol 64 GB
- 39.42% for the OCZ Petrol 128 GB
- 30.85% for the OCZ Octane 128 GB SATA II
- 29.46% for the OCZ Octane 64 GB SATA II
- 9.73% for the OCZ Vertex 2 120 GB 3.5"
- 9.59% for the OCZ Vertex 2 120 GB
- 6.73% for the OCZ Vertex 2 60 GB
- 5.43% for the OCZ Agility 3 240 GB
- 5.12% for the OCZ Vertex Plus 128 GB
"With such rates, we can justly classify such models [OCZ Petrol and Octane SATA II] as defective and it is shameful that such products have remained on sale in stores!"
Corsair and Crucial have overall return rates of 1% compared with 5% for OCZ.
Notice that OCZ's Indilinx based SSDs are described as "defective". The funny thing is that OCZ bought the company and has preferential access to firmware updates. So this begs the question, just what are OCZ doing differently? Is there a common factor between their SandForce and Indilinx products, and is this common factor a lower flash Vcc? Are SSDs with backup capacitors any less prone to failure than equivalent models without backup caps?
Even in endurance tests where presumably most drives were operated 24/7, OCZ fared worse than the others:
http://www.xtremesystems.org/forums/sho ... nm-Vs-34nmSo what does this tell us? Are chips actually less reliable at lower Vcc? If so, could this mean that worn chips can no longer be reliably erased and programmed at lower voltages?
