Industry related on topic, page-402

Hi @vishesh839

(The system wouldn't let my reply so we'll do it this way)

"how does WBT matrix compare to 4ds? Does WBT has better endurance and write/read speed based on their today's results??"

No, the announced results from 4DS are far better than WBT's ReRAM. The endurance @ 2 x 10⁹ is in the ballpark of 20,000 times better than WBT and 200,0000 times better than WBT's current qualification level at Skywater. Furthermore, this is only as far as they have tested. The ultimate endurance is yet to be established.

About speed, I'm not so sure. I have never seen any hard data on the read/write cycle times at the level of individual cells, for WBT's ReRAM or any other memory type for that matter (before today). WBT simply say they are 100 times faster than flash on write cycles. Performance is probably similar on read cycles.

Manufacturers' specs on speed tend to be given in terms of latency and transfer rate at the level of whole devices (chips, memory cards, and drives), where many other factors come into play, such as architecture, error checking and correction, temperature and voltage.

However, my guess (and it's just a guess so don't quote me on it) is that 4DS is in the order of 10 times faster than WBT's ReRAM.

The important thing to remember here, is that 4DS is not a direct competitor of WBT (Coby has said this on many occasions). Their product is going to be considerably more expensive to manufacture and they are targeting a different market. I could be wrong, but I don't think their tech is particularly fab friendly. It is a FEOL tech as opposed to WBT being a CMOS compatible BEOL technology. As I said above, I think UltraRAM has more to worry about than WBT (or TSMC).

All IMO