A Minute On Manufacturing Access

You may have seen an announcement today from privately-held Achronix, which plans to start manufacturing a future line of FPGA products on our forthcoming 22 nanometer chip-making process.

As many of you know, Intel enjoys a multi-year lead on manufacturing and process technology. Our factories or “fabs” are our prized possession. In fact, we just recently announced another $6-8 billion multi-year investment in manufacturing in the US alone. As Senior VP Brian Krzanich told CNET, we now have line of sight to 22, 15, 11 and 8 nanometers, with first 22nm products expected to be in production in the second half of 2011. Mind you, a nanometer measures in at one billionth of a meter. These days, more than 10 million transistors can fit in the period at the end of this sentence, and we can now make chips the size of a fingernail that have about one billion transistors inside them.

Being ahead and relentlessly pursuing Moore’s Law means smaller and lower-power Intel® Atom®, Core® and Xeon® chips about every two years. Yet these chips will still increase performance and provide more chip ‘real-estate’ to add features such as graphics or virtualization inside. But the advantage isn’t just a time-to-market one. We are also increasingly adding unique features to our manufacturing capability, such as our 2007 Hafnium-based high-k, metal gate technology, which created a new formula for reducing current leakage, and thus power consumption, of transistors. We’ve shipped more than 500 million chips with high-k, and we’ll be on our third generation of that technology when we begin producing 22 nm processors.

With Achronix, we are selectively offering access to our 22 nm fabs. For perspective, this deal would only make up a tiny amount of our overall capacity, significantly less than one percent, and is not currently viewed as financially material to Intel’s earnings. But it’s still an important endeavor for us that we’re committed to deliver on. I can tell you the folks over at Achronix are very excited about the opportunity and the expected performance boosts they will see in their Intel manufactured products. We are too.

So what say you? Assuming Intel (and our customers) can find alignment to benefit and profit from a relationship like this, what’s your view of opening up our manufacturing doors to others? Let us know.

5 thoughts on “A Minute On Manufacturing Access

  1. Hi Bill,
    I cover the foundry industry so this interests me. I also have work experience with FPGA’s as a replacement for ASICs with GateField which was acquired by Actel.
    I see five possible reasons for this move:
    (1) Intel enters the FPGA business
    (2) Intel enters the foundry business
    (3) Intel ramps advanced processes with FPGA technology (TSMC uses Altera)
    (4) Intel adds FPGA muscle to Atom (ARM and Altera)
    (5) Intel wants Achronix asynchronous logic IP
    So which is it?

  2. I highly doubt you would let AMD use your fabs, but what about Nvidia? Their latest Fermi-line could use some shrinking.

  3. Big question: Why does this make financial sense now?
    In the past, there was never enough leading edge capacity. Any wafers given to anything other than a high margin cpu would be wasting fab capacity.
    Can you charge enough for the wafer starts?
    Or is this a hedging of the bets?

Comments are closed.

Comments are closed.