hcl64 (mario.smarq@gmail.com) on 4/30/12 wrote:
---------------------------
>David Kanter (dkanter@realworldtech.com) on 4/30/12 wrote:
>---------------------------
>>hcl64 (mario.smarq@gmail.com) on 4/30/12 wrote:
>>---------------------------
>>>David Kanter (dkanter@realworldtech.com) on 4/30/12 wrote:
>>>---------------------------
>
>Sorry again Mr. Kanter
>
>Facts:

Actually fiction...

>NovaThor ARM chipset for smarthphone has been on the market >even before the end
>of 2011 in 1 or 2 HTC products mostly sold on China
>
>Wireless transceiver of ST, SPIRIT 1... is now on the market, or will be very soon (can't confirm).
>
>all based on ST 28nm FD-SOI, so we can say BEFORE 22nm >FinFET.

Let's try this again. Here's Ivy Bridge for sale:
http://www.newegg.com/Product/Product.aspx?Item=N82E16819116504&cm_sp=DailyDeal-_-19-116-504-_-Homepage2011

Here's a discussion of the FD-SOI from STE:

"LT: FD-SOI will be introduced into next generation products from ST-Ericsson. At this time, our first 28nm FD-SOI products are scheduled to tape out in Q3 2012 with production start anticipated in 2013."

The bold is mine, for emphasis.

http://semimd.com/hars/2012/04/24/st-ericsson-28nm-fd-soi-smartphone-soc-q3-tape-out-interview/

So according to STE, they haven't even taped out 28nm FD-SOI products and there won't be products until 2013.

In contrast, Intel is selling millions of FinFET products. And in 2014/2015 TSMC should have FinFETs as well.

>About other issues they can be matters of opinion,

Actually there were a lot of facts, e.g.:

"Also, many of the claims about FD-SOI assume that customers will use body biasing techniques to further enhance power/performance. Body bias adds cost to chips, because it requires additional voltage rails to apply the bias and it also adds complexity to the chip by distributing more power rails throughout the design.

I know that TI and Samsung are using body bias, and STMicro clearly intends to do so. However, high performance products generally don't. Body bias isn't used by Intel, AMD, IBM, Oracle or Nvidia for CPUs/GPUs. I don't believe it is used by Qualcomm either."

>but all in all, SOITEC group
>as source, **can be as much or as less reliable as Intel or related sources**,
>at least they don't have the "issues" of marketing of how >any of their particular
>products is "perceived" by the broad retail market, because >they don't have any products in a broad retail market.

That makes no sense. SOITEC talks up their technology because they need to convince companies to use it. Frankly, they have a poor history of doing so. There was a huge hyping of PD-SOI (and ZRAM). Neither of which panned out really. PD-SOI is a niche technology, and ZRAM is dead.

Just look at Advanced Substrate News. That entire website is a giant shill for SOITEC. It's all copyright SOITEC (if you look at the bottom of the page). So anything from Advanced Substrate News should be treated as if it were directly coming from SOITEC's website.

Of course Intel is trying to puff up their technology as well...but when it comes to process technology, they tend to be fairly honest. Moreover, FinFETs are not just an Intel technology.

TSMC is doing FinFETs and SOITEC is getting ready to do FinFETs.

The latter is particularly interesting. Using FD-SOI + FinFETs is guaranteed to be much more expensive than bulk FinFETs.

If FD-SOI is truly a replacement for FinFETs, then why are people designing FD-SOI with FinFETs?

The answer is that FD-SOI is suitable for applications where performance is a secondary concern. Where performance is a key criteria, you need FinFETs.

David