OCZ VERTEX 2 90GB not work

[Doomer]

If I had asked this question in a group of my peers, it would have been answered within the day.

Why don't you do that then

I'm asking you.

Just measured Samsung 7PC256D: NAND voltage is 2.75V

Last edited by Doomer on August 21st, 2013, 18:59, edited 1 time in total.

[Doomer]

If you look at the previous roundups, then OCZ fares consistently worse than the others:

I tried to look to the numbers and they don't looks straight
for example
Table says
8.57% OCZ Vertex 3 Series 120 GB
6.37% OCZ Vertex 3 Series 240 GB
At the same time description says
"Note that over the coming period, the Vertex 3s are doing much better thanks to developments in the firmware, with a rate of just 1.01% for the Vertex 3 120 GB as things stand."
So when does their statistics lie then or now?

Maybe you have a better source?

Last edited by Doomer on August 21st, 2013, 18:55, edited 3 times in total.

[Doomer]

If I'm not mistaken NAND voltage can be regulated by the ASIC on the fly and that's part of the error recovery process on SSDs
But would it be true for this SSDs can be checked with schematics

You're mistaken, at least in the case of the last Vertex 2 I looked at.

Could you elaborate?

[fzabkar]

Vertex 2 240GB

[inkub]

Where is located the SSD software.
Inside SF1222, or in an external NAND?

[Doomer]

on the NANDs

[fzabkar]

If I had asked this question in a group of my peers, it would have been answered within the day.

Why don't you do that then

I'm asking you.

Just measured Samsung 7PC256D: NAND voltage is 2.75V

AFAICT, these SSDs use Samsung's K9PFGY8U7A-HCK0 NAND parts, and they don't have backup capacitors.

The "U" indicates a Vcc range from 2.7V to 3.6V, so ISTM that low Vcc is a non-issue. :o

[Doomer]

AFAICT, these SSDs use Samsung's K9PFGY8U7A-HCK0 NAND parts, and they don't have backup capacitors.

The "U" indicates a Vcc range from 2.7V to 3.6V, so ISTM that low Vcc is a non-issue.

So 2.75V on Samsung is not an issue but 2.9V on OGZ/Hynix is?
Interesting logic you got there

[fzabkar]

AFAICT, these SSDs use Samsung's K9PFGY8U7A-HCK0 NAND parts, and they don't have backup capacitors.

The "U" indicates a Vcc range from 2.7V to 3.6V, so ISTM that low Vcc is a non-issue. :o

So 2.75V on Samsung is not an issue but 2.9V on OGZ/Hynix is?
Interesting logic you got there

I didn't say that at all. I meant that low Vcc appears to be a non-issue ... period.

AISI, if Samsung doesn't care about the contents of cache memory, then it must contain nothing that is of vital system importance. It would appear that all cache contents, including unflushed user data, are simply sacrificed when power is lost

Of course this then begs the question, what are backup capacitors used for? My calculations suggest that the capacitor bank on an Intel SSD, for example, is nowhere near big enough to allow even a small portion of the SDRAM to be saved to flash. AIUI most of the RAM cache is used for the FTL and other system related structures, but if they're not affected by power loss, then should we bother trying to save a little bit of user data? No HDD manufacturer does this.

Having said the above, it still goes against the grain to see a device being operated at its design limit. :?

[Doomer]

My calculations suggest that the capacitor bank on an Intel SSD, for example, is nowhere near big enough to allow even a small portion of the SDRAM to be saved to flash. AIUI most of the RAM cache is used for the FTL and other system related structures, but if they're not affected by power loss, then should we bother trying to save a little bit of user data?

Yeah, that's an issue on majority of customer-grade SSDs, I think