Intel’s Mark Bohr on the push into multi-chip solutions and issues at 5nm/7nm

&quot;Our 14nm delivers a logic transistor density of about 37.5 million transistors per square millimeter. Our 10nm technology delivers about 100 million transistors per square millimeter.&quot;<p>So go draw a square millimeter. For us US people, it&#x27;s a little bigger than 1&#x2F;32 of an inch (closer to 1&#x2F;26th of an inch). If you do any woodworking, you can have a dovetail too big by that much and just wack it in. So it&#x27;s sort of the allowable slop in a woodworking joint (in softwoods).<p>They&#x27;ve got 100 million transistors in that tiny amount of space? I know, it&#x27;s not news, but wow, stop and think about the tech that they have to have to do the layout. Quite an accomplishment, especially considering that the transistor was invented in 1947. In 70 years we go from just having a transistor to having 100 million in a square millimeter.<p>Other than computer stuff, is there anything else that has seen that sort of scaling up?

&gt; For the other companies just shipping 10nm, we expect what they call 10nm to have a logic transistor density measured close 50 million transistors per square millimeter, not the 100 million per square millimeter that we offer on our 10nm. It’s almost a full generation difference.<p>Something to keep in mind. I&#x27;ve read in multiple articles that other foundries have caught up with intel. Apparently not quite. Though it&#x27;s probably closer than it has been for a while.

&gt; <i>ARM is part of our foundry offerings now. We offer ARM IP, which can be put on Intel chips and be part of an Intel chip, whether it’s a 14nm or 10nm chip. That’s an IP block, not necessarily a separate chip.</i><p>Where would this be needed?