If you are considering becoming a 1M/1M premium member and would like to join our mailing list to receive ongoing information, please sign up here.

Subscribe to our Feed

SiP is a major discontinuity

Posted on Wednesday, Apr 6th 2005

Okay, so I should explain what is the big deal about System-in-Package (SiP) …

You must have noticed the trend in cellular handsets of packing in dramatically diverse functionality? Samsung’s latest model has a 5 Mega Pixel camera, and if this is any indication, consumers will further reap benefits of such integration trends in the coming years.

What makes this possible? For a miniscule gadget to pack in various kinds of RF modules, DSP blocks, memory modules, etc. – the technology being used is SiP. It is at the heart of electronics miniaturization, along with Chip Scale Packaging (CSP). The differentiating phenomenon in very fast turn-around, affordable hyper-integration is SiP.

And what’s even more attractive for the IDMs is the fact that the SiP design paradigm allows them to mix dies from different process nodes. They can mix three 130nm modules with one 65nm module, thereby amortizing the old processes.

Finally, SiP will put a further brake in the slowdown of Moore’s Law. Now that old processes can be mixed in with the new, IDMs should rethink their entire product strategies, and Foundries should do capacity planning assuming that not all parts of a chip will require the newest process.

By all measures, SiP is a major discontinuity in the semiconductor business. Those chip houses who will figure out the implications and take advantage of it, stand to win big.

Hacker News
() Comments

Featured Videos


Agree with all of that but few howevers are still up there.
Given that a SiP is a chip that is created by integrating other chips, the following are issues that SiP integrators will have to deal with.

– Markup that will be asked for by the providers of those chips might make the end design a challenge, economically speaking.
– If one of the chips that is assembled in the SiP is not working all the rest of the chips that are already placed, might have to be thrown away, something that increase the cost of the total solution.

As a result of the concerns above, I believe, only large companies who own a lot of IP are using this implementation style, were the small players are still not participating.
Given that the SOC (System On Chip) style is doomed to fail for integration of diverse technologies, companies that will solve the problems described above can harvest huge returns.

Lavi Lev Saturday, May 7, 2005 at 9:43 PM PT

Yes, true, Lavi. I have heard about some smaller companies though, that are designing the entire chip as a SiP, including using all homegrown modules or licensed IP instead of lose die. I believe Fyrestorm ( is doing that – have to check – as a case in point. This means, that if people are smart about leveraging the benefits of SiPs, and can work around the problems like Known-Good-Die and lose die sourcing, they stand to win. Sramana

Sramana Mitra Sunday, May 8, 2005 at 9:39 PM PT

[…] o go from prototype to commercial production, that remains to be seen. System in a package is a big shift in terms of making and developing silicon solutions. How they are different from traditional ASICs […]

Om Malik’s Broadband Blog » Intel develops an 802.11 everything chip Sunday, June 19, 2005 at 12:48 AM PT

[…] An extremely key technology for India to build core competency in, if it wants to play in electronics will be System-In-Package. Most of East Asia has mastered this, and are rolling out mid-volume gadgets at a fierce speed. […]

Sramana Mitra on Strategy » Blog Archive » Concept Arbitrage: Juniper Monday, June 26, 2006 at 6:54 PM PT

[…] chip design and manufacturing capabilities, especially in SiP design, used by major vendors as a key hyper-integration and miniaturization […]

Sramana Mitra on Strategy » Blog Archive » iPhone and the Future of Samsung Thursday, May 3, 2007 at 5:41 AM PT

[…] There is one catch in this company’s future, however. That is a looming question mark over how the company navigates the convergence device trend, and manages to defend its PND position as Smartphones start to absorb PND functionality more and more. Anticipating this trend, Garmin has started making certain moved already. One such is releasing a $200 BlueTooth accessory that turns a smartphone into a GPS receiver. The question, however, is whether the Smartphone vendors need Garmin in the middle, as the era of hyper-integration marches forward, fully capable of absorbing a GPS chip into the fold… […]

Garmin Looks Great, But … - Sramana Mitra on Strategy Thursday, August 16, 2007 at 7:40 AM PT

[…] “off-the-shelf” leadframe configurations to leading-edge chip scale, flip chip and system-in-package (SiP) solutions incorporating highly customized designs. The company is of interest because of the surge […]

Amkor Worth a Look - Sramana Mitra on Strategy Wednesday, July 9, 2008 at 6:06 AM PT

[…] Indeed, SiP “is a major discontinuity in the semiconductor business,” wrote Sramana Mitra back in April 2005. “SiP will put a further brake in the slowdown of Moore’s Law.” And from Wikipedia: […]

Why It’s the Megabits, Not the MIPs, That Matter Friday, May 22, 2009 at 2:31 PM PT

[…] Indeed, SiP “is a major discontinuity in the semiconductor business,” wrote Sramana Mitra back in April 2005. “SiP will put a further brake in the slowdown of Moore’s Law.” And from Wikipedia: […]

Why It’s the Megabits, Not the MIPs, That Matter Monday, January 10, 2011 at 11:19 AM PT