Hey /g/ i have a tinfoil hat theory about Intel, what do you think, that Mesh is going to be Netbrusrt 2.0
>Intel up until 2000 relies on transistor shrinkage without and major architecture changes
>In 2000 they come up with Netburst in order to reach 20ghz speeds.
>Fails, Chips ran very hot, and where less efficient then the Pentium 3s at launch
>Rather than pushing speed they follow AMD and go the more core count in order to keep performance and progress moving forward.
>AMD shits the bed with bulldozer
tinfoil hat part
>Intel finds during it's near total domination of the laptop and desktop market that Ring Bus won't scale very well
>they start buying up large entities in the hopes they can find talent to fix the problem
>This is why during this time we don't really see much improvement over core count, not due to Intel's greed, but do to ont being able to come up with something better.
>Amd brings back Jim Keller to solve the same problem
>intel Doesn't really change much to their CPUs until they can figure something out, realize on smaller transistors in order to get small gains each generation, not in too much of a hurry.
>They get some AMD silicone in their labs to see what they came up with.
>They now have to use what they got in the form of Mesh even though it's not ready
>Ryzen launch
>Mesh launch
We know for a fact that Intel CPUs with Mesh run very hot and use more power compared to AMD at this time, You can debate on which is better performance but that's not the point. I think they won't be able to solve the heat problem while pushing more cores, similar to how they couldn't push more speed on Netburst without running into heat problems. AMD I think will scale better at 10nm and 7nm due to Jim's design and we will see difference in performance close that of the Pentium 4 and Athlon 3000+ days.
Thats is happening right now.