Breaking News
Blog

Samsung’s 3D V-NAND 32L vs 48L--Just Vertical Expansion?

NO RATINGS
2 saves
Page 1 / 3 Next >
View Comments: Newest First | Oldest First | Threaded View
Page 1 / 2   >   >>
Sang-Yun
User Rank
Rookie
Re: Price parity at 64-layer and what's next?
Sang-Yun   7/2/2016 9:24:07 AM
NO RATINGS
Thank you for sharing useful information!

resistion
User Rank
Author
Expensive stairs
resistion   7/2/2016 7:08:59 AM
NO RATINGS
The staircase cost is a concern. Everyone had assumed one stair mask for all layers. That was Toshiba's story actually. It's actually one mask for 8 stair steps. http://www.semiconchina.org/downloadFile/1460699158931.pdf

resistion
User Rank
Author
Re: Price parity at 64-layer and what's next?
resistion   7/2/2016 7:02:49 AM
NO RATINGS
Micron's 3D NAND looks more promising but it was reported at ISSCC that their block size could be too large for some customers. That could lead to degraded die area efficiency. Most likely their next generation will benefit more from 2D lithographic shrink than vertical enhancements like stacking more layers.

realjjj
User Rank
CEO
Re: Price parity at 64-layer and what's next?
realjjj   7/1/2016 8:14:13 PM
NO RATINGS
That's the market willing to pay more for 3D today and i guess that won't change before the majority of the bit output is 3D. Ofc it is a fair concern as demand is very elastic and prices need to decline.

Sang-Yun
User Rank
Rookie
Re: Price parity at 64-layer and what's next?
Sang-Yun   7/1/2016 6:14:32 PM
NO RATINGS
Here is $/GB comparison at the market

Planar TLC: $0.21

Planar MLC: $0.26

Samsung 32-layer TLC 3D NAND: $0.32

Micron 32-layer TLC 3D NAND: $0.27

Sang-Yun
User Rank
Rookie
Re: Price parity at 64-layer and what's next?
Sang-Yun   7/1/2016 6:02:23 PM
NO RATINGS
Thank you for sharing the info. I aware of Micron's claims. It it works like that, Micorn should the dominant winner in the NAND market. At this moment, I don't have enough data to compare Samsung 3D NAND and Micron's 3D NAND.   

realjjj
User Rank
CEO
Re: Price parity at 64-layer and what's next?
realjjj   7/1/2016 5:41:11 PM
NO RATINGS
*384Gb not GB obviously

realjjj
User Rank
CEO
Re: Price parity at 64-layer and what's next?
realjjj   7/1/2016 5:39:04 PM
NO RATINGS
Micron said their 3D NAND bit output will surpass 2D this fall and they also said costs are lower than their 2D 20nm and 16nm - hard to say if there are any catches here like comparing 2D MLC with 3D TLC. They also claim that their second gen will reduce costs by 30% over the first gen.

Their 16nm die wasn't the smallest ( i think 173mm2 for 128Gb MLC) and their first gen 3D TLC is 32 layers 384GB at 168.5mm2 so maybe they did better than others with fewer layers, floating gate and a bigger die. Assuming their second gen is what they showed at ISSCC, it would be 64 layers maybe 2x32 string stacked, 768Gb TLC at 179.2mm2. Maybe some have more problems but Micron seems to be doing better, the fact that such a high % of their bit output will be 3D this year makes that hard to doubt.

Sang-Yun
User Rank
Rookie
Re: Price parity at 64-layer and what's next?
Sang-Yun   7/1/2016 4:56:17 PM
NO RATINGS
I think 3D NAND will be unable to reach the price parity with planar NAND regardless technology maturity and demands. If we see the technology in detail and the big picture of 3D NAND roadmap, we cannot find out low cost solution at all. At Amazon.com, 32-layer TLC 3D NAND SSD is about 50% more expensive than planar NAND SSD. 32-layer 3D NAND is already matured with more than 90% yield and more than $30Bn NAND market is already wide open for 3D NAND if it is cost competitive. 48-layer TLC 3D NAND will be more efficient. However, the price gap between 3D and 2D NANDs will be more than 35%. Considering techical difficulties and negative effect of cell efficiency (i.e. WL staircase), 64-layer 3D NAND has low chance to make price parity with planar NAND. After 64-layer, multi-string (which is multi-stacking of 64-layers) is considered. It will make worse in terms of Cost-per-Bit. Manufacturing cost of multi-string will be easily double. In addition, yield will drastically drop down. If we consider 90% and 80% yields at single-string, multi-string with 2 stacked layers will have 81% and 64% yeidls, respectively. NAND industry faces limitation of device scaling, which generates panic. So, NAND industry jumps into 3D NAND and creates "3D NAND Syndrome" And then, this syndrome has been amplified by media. During the Flash Memory Summit in August, we should ask clear roadmap of "Cost-per-Bit" reduction to NAND vendors. Competition of memory layer stacking in vertical direction is meaningless without cost savings.     

msporer
User Rank
Author
Re: Price parity at 64-layer and what's next?
msporer   7/1/2016 3:58:52 PM
NO RATINGS
Planar NAND has reached the end of scaling.  3D NAND is the only way to continue to increase density, and the cost will become optimized as volume production increases.  Whether or not it ever reaches price parity is immaterial because demand for NAND bits is increasing.

Page 1 / 2   >   >>

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)
Like Us on Facebook
EE Times on Twitter
EE Times Twitter Feed