Forgotten Intel Patent Paves the Way for Ultra-Fast Processing

I’ve worked with computers for 28 years, and I’ve spent many image years following the trends and staying on top of new developments to keep my home servers and workstations at peak performance.  I found the race less interesting over time, but this article my brother  passed long peaked my interest.  Intel found a way to integrate DRAM onto the L2 cache using only two transistors and no capacitors, and they run with blazingly fast throughput.  Current chip technology uses SRAM which typically uses 6 transistors to store a single bit.  SRAM currently has about 8-20GB/S throughput, whereas the new DRAM method can get up to about 204BG/S throughput. That’s an order of magnitude increase in processing, and could virtually eliminate cache misses, which will revolutionize the efficiency of each CPU clock.   Read more here: 

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.