Sign in with
Sign up | Sign in

AMD First to 22 Nm, Challenges Intel

By - Source: Tom's Hardware US | B 37 comments

Yorktown Heights (NY) - IBM and its chip development partners, including AMD, made a stunning announcement today, apparently beating Intel in the successful production of the first functional 22 nm SRAM cell. 22 nm processors are still three years out in the future, but IBM’s news is a good sign that chip manufacturer will be able to easily scale to this new level by the end of 2011. It appears that, for the first time in several decades, Intel may have to put some extra time into its research and development efforts to make sure it can keep its manufacturing lead at 22 nm and beyond.

SRAM chips are typically the first semiconductor devices to test a new manufacturing process as a precursor to actual microprocessors. The devices developed and manufactured by AMD, Freescale, IBM STMicroelectronics, Toshiba and the College of Nanoscale Science and Engineering (CNSE) were built in a traditional six-transistor design on a 300 mm wafer and had a memory cell size of just 0.1 μm2, which compares to Intel’s 45 nm SRAM cell size (the test chip that was used for today’s 45 nm processors) of 0.346 μm2.

A 22 nm chip is two generations out in the future and AMD even has to catch up with Intel’s 45 nm. Intel presented the first 32 nm SRAM cell wafer in September of last year and in fact is not expected to show 22 nm SRAM cells for at least another year, while first 32 nm CPU prototypes could be shown at IDF this week.

IBM said that it is on track with its 32 nm process and promises that it will use a "leading 32 nm high-K metal gate technology that no other company or consortium can match." IBM did not provide further details to substantiate this claim, however, Intel has been using its high-K metal gate technology since the introduction of the 45 nm Penryn processors in late 2007.

While we are far from actual 22 nm and 32 nm products, it is clear that IBM and its partners are turning up the heat on Intel. For the first time in decades, there could actually be an interesting race towards a new production node.

Discuss
Display all 37 comments.
This thread is closed for comments
Top Comments
  • 10 Hide
    jaragon13 , August 18, 2008 11:44 PM
    spanner_razorMajor failure in the thread title, should be IBM not AMD reaches 22nm.

    bobsmith1AMD needs to join up with IBM.

    Lol,you guys are totally ignorant of the fact AMD works with IBM...
  • 10 Hide
    Anonymous , August 18, 2008 11:46 PM
    IBM is king

    my dad works there :p 
Other Comments
  • 2 Hide
    physx7 , August 18, 2008 11:19 PM
    Cool nice article..
  • -4 Hide
    spanner_razor , August 18, 2008 11:22 PM
    Major failure in the thread title, should be IBM not AMD reaches 22nm.
  • -1 Hide
    Anonymous , August 18, 2008 11:35 PM
    AMD needs to join up with IBM.
  • 10 Hide
    jaragon13 , August 18, 2008 11:44 PM
    spanner_razorMajor failure in the thread title, should be IBM not AMD reaches 22nm.

    bobsmith1AMD needs to join up with IBM.

    Lol,you guys are totally ignorant of the fact AMD works with IBM...
  • 10 Hide
    Anonymous , August 18, 2008 11:46 PM
    IBM is king

    my dad works there :p 
  • 8 Hide
    Anonymous , August 18, 2008 11:48 PM
    IBM and AMD made a deal ages ago for the production of 22nm chips, so this is the culmination of it. They made the achievement together.
  • 0 Hide
    frozenlead , August 18, 2008 11:54 PM
    I wonder what the overclocking margins on such a small chip will be...
  • 1 Hide
    jaragon13 , August 19, 2008 12:11 AM
    frozenleadI wonder what the overclocking margins on such a small chip will be...

    depends on how good the actual process is...

    I myself don't overclock,don't need more waste heat on my other parts
  • 0 Hide
    pogsnet , August 19, 2008 12:13 AM
    They should create 22nm processor next year ^_^
  • 0 Hide
    horendus , August 19, 2008 12:27 AM
    Im just wondering what kinda of clock speed increase can be expected from 22nm because we have seen very little average clock speed increases over the past 3 or 4 manurfacturing gereration (90-65-45) (although clock speed certainly is not everything when it comes to performance)

    I guess this is the kinda manufacturing size that would be weel suited for the CPU/GPU intergrations that are on the horizons...


    and..a few years down the track....
    ...maybe once CPU/GPU intergration kicks off, Physx(pun intended) will be once again sent back to the CPU for calculating allowing for single Vcard setups to make a comeback....
  • -5 Hide
    JonathanDeane , August 19, 2008 12:51 AM
    It always seems IBM anounces they made this or that break through... then its like 5 years before you can actually buy the tech.
    Are they making 45nm Cell chips yet ?
  • -5 Hide
    wrack , August 19, 2008 1:29 AM
    Incorrect title! Nice to attract more traffic though.
  • -5 Hide
    zerapio , August 19, 2008 1:44 AM
    Take a look at this:

    I appears IBM also set the previous record for 32nm back in 2004. Have they released any 32nm processor? Not that I'm aware. I think the first 32nm CPU's will be shown at IDF.

    I think the title is misleading on the major contributor and the implication of the finding.
  • -4 Hide
    zerapio , August 19, 2008 1:49 AM
    (repost with good link)
    Take a look at this:

    I appears IBM also set the previous record for 32nm back in 2004. Have they released any 32nm processor? Not that I'm aware. I think the first 32nm CPU's will be shown at IDF.

    I think the title is misleading on the major contributor and the implication of the finding.
  • 0 Hide
    caamsa , August 19, 2008 1:55 AM
    Hell I would like to see AMD release a 45nm cpu........get your heads out of the future and into the present moment.
  • -3 Hide
    Anonymous , August 19, 2008 4:39 AM
    Well like Tom's said before IBM is a evil vampire that never die :p 
  • -3 Hide
    milk_inc , August 19, 2008 4:40 AM
    Like Tom's said before IBM is vampire that never die.
  • -4 Hide
    lopopo , August 19, 2008 5:01 AM
    Nice title. Credibility = lost
  • 0 Hide
    blackwidow_rsa , August 19, 2008 6:18 AM
    It's called planning ahead, it's kind of needed
  • 0 Hide
    Anonymous , August 19, 2008 8:08 AM
    Good to see some folks drinking the cool-aid. It is H2'2008, correct? Did I miss the 32nm node? Exactly when will 32nm go into production and then 22nm? 2011 seems a bit optimistic for IBM, yet alone AMD which hasn't even sold a SINGLE 45nm part yet! Help us out on this one... AMD 45nm end of the this year 2008... 32nm maybe mid-2010/late 2010... and then 22nm the following year?

    "but IBM?s news is a good sign that chip manufacturer will be able to easily scale to this new level by the end of 2011"

    If only we could fast forward 3 years! "EASILY SCALE"?!?!? Timeline?!? (are you trying to minimize the actual use of 45nm and 32nm capital equipment to throw it away as soon as possible?) This makes no economic or technical sense!

    Finally consider READING the announcement - IBM demo'd/claimed a single SRAM cell (which is 6 transistors), not even an actual test chip (which is what you reference when discussing Intel). A 22nm SRAM cell does not mean a 22nm test chip, yet alone a yielding 22nm wafer. When you say Intel is not expected to show 22nm until next year you are no doubt referring to a TEST CHIP (containing 100's of millions, if not a billion transistors) as opposed to the IBM 6 transistor announcement... a wee bit (maybe 5 or 6 orders of magnitude?) different!

    If you don't understand the technical details, which from the article it is clear that you don't, you shouldn't be making any analysis from it.
Display more comments