G3 What gives?

[idata]

So OCZ are now shipping SSD's with IMF 2Xnm NAND. What gives? What has happened to the G3?

[idata]

Scott mention there will be a SSD announcement at the Game Developer Conference which runs 2/28-3/4.

No way to know it is about the G3 or not...

[idata]

redux, I read an article that said Intel has delayed the G3 SSD due to "longevity issues" with the 25nm NAND. I can't find the link at the moment, but I'll keep looking.

You likely know that as the size of the NAND cells decreases, the amount of times they may be written to or erased also decreases. As I recall from the article, the author assumed Intel was not happy with the lifespan of the 25nm NAND chips, and delayed the release of the G3's until that could be resolved.

Apparently, OCZ has a work-around or some technique that solves the issue, or they don't care?

EDIT: Ok, found an article, not the one I was referring to, but this does explain things... interesting:

http://www.storagereview.com/ssds_shifting_25nm_nand_what_you_need_know http://www.storagereview.com/ssds_shifting_25nm_nand_what_you_need_know

Message was edited by: parsec

[idata]

OCZ seem to have messed up completely the way they have introduced 2Xnm. I would not expect Intel to follow suit.

The leaked G3 specs indicate that random write lifespan capability will be increased considerably in comparison to the G2 drives.

I'd guess this is possible due to an increase in over provisioning and more sophisticated controllers, which will more than make up for the loss of erase cycles with 2Xnm.

According to the specs leaked:

Total 4KB Random Writes (Drive Lifespan)

G2 = 7.5TB - 15TB

G3 = 30TB - 60TB

http://www.anandtech.com/show/3965/intels-3rd-generation-x25m-ssd-specs-revealed http://www.anandtech.com/show/3965/intels-3rd-generation-x25m-ssd-specs-revealed

[idata]

I agree, and as we can see, Intel is not following suit.

I am surprised that Intel supposedly was shooting for those G3 specs, but how could they not know of the issues? Well, easy for me to conjecture knowing zero details. Actually it's probably a cost issue, since the increased over-provisioning space adds to the cost, while a reduced final price to the consumer is a major goal. It would be great if Intel can attain those longevity specs.

IMO, what OCZ did (they use part of the advertised SSD's capacity as over-provisioned space instead of adding more separately, and the post-formatting size of the SSD is smaller than it should be) is indicative of their priorities in the marketplace. That is, get that "new, bigger, better" product out there first for those folks that must have cutting edge components, which also gives them the appearance of a leader and innovator in the marketplace.

Alas, their customers and the reviewers are more savvy than their marketing people give them credit for, and are unhappy with the loss of capacity. That may be only the beginning of their problems, if these SSDs don't last very long.