News
 

Bookmark and Share

(5) 

In a completely surprising move, Intel Corp. on Monday said that it would produce field-programmable gate arrays (FPGAs) to privately owned Achronix Semiconductor Corp. using 22nm process technology. For the first time in history the world's largest maker of microprocessors actually becomes a foundry for a smaller company. Moreover, the giant welcomes other developers to follow.

"With Achronix, we are selectively offering access to our 22nm fabs. For perspective, this deal would only make up a tiny amount of our overall capacity, significantly less than 1%, and is not currently viewed as financially material to Intel’s earnings. But it’s still an important endeavor for us that we are 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," said Bill Kircos, a director of product and technology media relations at Intel's global communications group.

Intel's motives to manufacture certain third-party chips on its state-of-the-art manufacturing facilities are not completely clear. One of the explanations is that the company needs certain products that it cannot make itself to exist in order to sell its microprocessors.

According to the developer, the Achronix Speedster22i FPGA family will shatter existing limitations of FPGAs, allowing cost effective production of high performance devices over 2.5M LUTs in size, equivalent to an ASIC of over 20 million gates.

Taking advantage of the performance and power savings of Intel’s 22nm process technology, Speedster22i will also extend the boundaries of FPGA speed and power efficiency, enabling as much as 300% higher performance, 50% lower power, and 40% lower cost than any other FPGA in any other process technology.

Achronix Speedster22i will be suitable for a wide range of applications in the telecommunication, networking, industrial and consumer markets and will enable emerging applications such as 100G, 400G Ethernet networking and LTE mobile communications. Additionally, Speedster22i will be the first commercial FPGA family that can be manufactured in the United States of America, making it the ideal platform for military and aerospace applications requiring “on shore” silicon.

“Intel has the best process technology in the world and we are privileged to have formed this strategic relationship, which enables simultaneous improvements in speed, power, density and cost. The combination of the advanced 22nm process from Intel and the advanced FPGA technology from Achronix enables Speedster22i to eclipse other FPGA solutions expected to hit the market in the next few years," said John Lofton Holt chief executive officer of Achronix.

Tags: Intel, Achronix, Semiconductor, 22nm

Discussion

Comments currently: 5
Discussion started: 11/01/10 09:52:08 PM
Latest comment: 11/03/10 07:19:48 AM
Expand all threads | Collapse all threads

[1-4]

1. 
>>Intel's motives...are not completely clear
No mystery here: Intel is concerned about fully utilizing the 22nm production lines. With 32nm they were forced to produce cheaper mainstream chips sooner than desired since sales on higher-end CPUs couldn't consume the production capacity. This also has negative effect on amortization of older fab line costs so must be avoided.

FPGAs don't compete with any Intel business, and may well rather profitable to produce since they are kind of a niche. In the (unlikely) event that 22nm CPU sales are so strong that production becomes constrained, Intel no doubt has the contractual option to prempt FPGA production for their own CPUs.
0 0 [Posted by: Phlogiston  | Date: 11/01/10 09:52:08 PM]
Reply
- collapse thread

 
"With 32nm they were forced to produce cheaper mainstream chips sooner than desired since sales on higher-end CPUs couldn't consume the production capacity."

Dont you worry'bout them. If they retained those pretty insane prices for those quasi-premium chips they could move more of them

But that in fact wasn't intel's intention at all. They hide problems with their 32nm node ramp up under the carpet and claim first to the market 32nm node. They heavily pushed Westmere-EP into server market trying to move AMD from there, but i dont see all those scrubbed Gulftown (desktop parts) that has been sold out due high demand rather thru obscure availability dictated thru extreme prices.

Intel's intention imnsho just to skip from 32nm node to 22nm as soon as possible and if they deploy it 2yrs after last node as usual. Then they'll probably be more than 1.5yr ahead from any foundry competitor.

So it's not that they couldn't ramp up 32nm they just never wnted to when 22nm was the rel goal. 40nm most of Fabs have GF/TSMC (UMC?) is too close to 32nm while 28nm to 22nm is huge jump considering TSMC still doesnt eevn mention it's 28nm after epic 40nm troubles, and GloFo seems like it'll have troubles ramping up NY Fab production and have to postpone it's date in planned schedule.

So 1.5yr with far most advanced processing node is something Intel eagerly expecting.

FPGA after all is nowhere so complex as cpu, that couldn't be properly salvaged. And then there's bragging rights with "possibly lowest power consumption" in that market if you use most advanced node.
0 0 [Posted by: OmegaHuman  | Date: 11/02/10 09:19:10 PM]
Reply

2. 
Why doesn't INTEL manufacture VIA's CPUs ? Yeah, the ones competing with Atom.
0 0 [Posted by: East17  | Date: 11/02/10 06:08:21 AM]
Reply

3. 
Intel's first foundry customer. Cute. They feeling threatened from GF, must of been.
0 0 [Posted by: OmegaHuman  | Date: 11/02/10 09:03:20 PM]
Reply

4. 
intel has the worlds best fabs, always has... this allowed them to survive competition with superior architectures in the past (although now they have the best fabs AND best architecture for CPUs)

leasing said capacity makes a lot of sense, there are many applications that are not competing with them in any way shape of form, this is one of those. It doesn't compete with intel, yet utilizes their fab capacity (providing another source of income), there is no reason NOT to do this on intel's part.

The theory that intel needs this technology is legitimate as well. It will allow a variety of other technologies to advance which increases demand for intel's own core products.
0 0 [Posted by: taltamir  | Date: 11/03/10 07:19:48 AM]
Reply

[1-4]

Add your Comment




Related news

Latest News

Monday, July 28, 2014

12:11 pm | Intel Core i7-5960X “Haswell-E” De-Lidded: Twelve Cores and Alloy-Based Thermal Interface. Intel Core i7-5960X Uses “Haswell-EP” Die, Promises Good Overclocking Potential

Tuesday, July 22, 2014

10:40 pm | ARM Preps Second-Generation “Artemis” and “Maya” 64-Bit ARMv8-A Offerings. ARM Readies 64-Bit Cores for Non-Traditional Applications

7:38 pm | AMD Vows to Introduce 20nm Products Next Year. AMD’s 20nm APUs, GPUs and Embedded Chips to Arrive in 2015

4:08 am | Microsoft to Unify All Windows Operating Systems for Client PCs. One Windows OS will Power PCs, Tablets and Smartphones

Monday, July 21, 2014

10:32 pm | PQI Debuts Flash Drive with Lightning and USB Connectors. PQI Offers Easy Way to Boost iPhone or iPad Storage

10:08 pm | Japan Display Begins to Mass Produce IPS-NEO Displays. JDI Begins to Mass Produce Rival for AMOLED Panels

12:56 pm | Microsoft to Fire 18,000 Employees to Boost Efficiency. Microsoft to Perform Massive Job Cut Ever Following Acquisition of Nokia