Originally posted by You Beaut
It gets back to my earlier post, filamentary ReRam can't get the DRAM like speed (without error correction) required for SCM.
Hence, not surprising that wbt has explicitly changed focus to join all the other "early mover" filamentary ReRam companies who, despite aspirations of conquering the world, ended up producing embedded memory.
Based on current mkt cap, the market would appear to share your skepticism regarding 4ds achieving storage class memory. However, Western Digital and imec clearly see something, and the industry is keenly waiting/watching for the data/results. I suppose we will know soon enough.
“However, high-speed operations cause high power consumption and chip temperature increase, which can deteriorate NAND reliability. Hence, reduction of operating voltage is inevitable to achieve the future NAND. Opportunities for the use of 3-D as well as polymer memory design in modern electronic circuits are rapidly expanding, based on the very high performance and unique functionality. However, their practical implementation in electronic applications will ultimately be decided by the ability to produce devices and circuits at a cost that is significantly below that needed to manufacture conventional electronic circuits based on, for example, silicon.”
“An embedded nonvolatile memory with superior performance to Flash could see widespread adoption in system-on-chip (SoC) applications such as smart cards and microcontrollers.”
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4182445/
https://weebit-nano.com/wp-content/uploads/2018/05/may-presentation.pdf
All that a successful replacement for Flash needs to do for the next decade or two is be 1000 x faster at the lowest cost.
The above slide is from May 2018.
WBT has achieved its 1MB array and is now moving to 28nm.
Ability to scale to 1GB is implied according to TMT Analytics https://www.weebit-nano.com/wp-content/uploads/2017/10/WBT-company-report-191017-1.pdf
WBT is strides ahead with development and does not fail the cost benefit analysis like others. Remember, if Intel can’t do it cost-effectively without silicon perhaps no-one can (for the next 2 decades).
A lot get caught up in the theory of what’s better but theories not grounded in commercial reality will leave investors kicking and screaming like 4DS investors were when its first iteration did not succeed in Sep.
Opinion only, dyor etc
(20min delay)