Intel Corp (INTC)

[DavidA2]

Again, I don't see any process advantage for Intel as of today.

It's a little bit more complicated than that.

Traditionally Intel lost out in density compared to foundries. Their 22nm was only about 30% more in density compared to TSMC's 28nm. They only told us when they had 14nm out, when they started focusing on density.

If that was true before, that means Intel's 45nm was 30-50% larger than TSMC's 40nm process. Intel's 65nm would have been 30-50% larger than TSMC's 65nm process.

The key for Intel was that not only they came out with it faster, they had the transistor performance lead. That makes sense since Intel makes very high performance processors.

In the Ryzen vs Skylake comparison, yes the core is bigger on the latter, but so is the performance. Skylake versus Ryzen core difference is much smaller if you compare it to the consumer Skylake which lacks AVX-512 and the extra L2 cache.

On the consumer side, Kaby and Coffee can clock 5GHz or more, while Ryzen struggles beyond 4GHz. That's in no insignificant way helped by the better performing transistors. Higher performing circuits also cost more in terms of die area.

In the purest sense of "Who's process is the densest?", yes Intel doesn't have a practical advantage. But consider that they have a nonexistent foundry business, and all their lines take full advantage of their process, it doesn't really matter.

It's all marketing talk. They used to not do this before they "went into foundry business". Now its all talk for them like its all talk for rest of the foundries.