Apple has had great success in the MP3 player market. They have released multiple versions of MP3 players. We had previously reported on "Find out what's really inside the iPod" a year ago, which identified some of the components in various different generations of the iPod family. On August 12, Apple release new versions of their line up, most notably the inclusion of an 8GB iPod nano.
Compared to the previous incarnation, which was a 4GB version, the new system stacks up quite well. While the density has doubled, allowing for approximately 2000 songs, the form factor is quite comparable.
|| Previously released 4GB iPod nano
|| Newly released 8GB iPod nano
||3.5" x 1.6" x 0.27" (H x W x D)
||3.5" x 1.6" x 0.26" (H x W x D)
| LCD Screen
| Estimated Battery Life
The best place to start the component discussion would have to be around the "mystery chips" that have been widely speculated about over the past weeks. There have been a number of speculations regarding the contents of three Apple packaged components in the latest iPod nano. It is not uncommon for OEMs to get components packaged with markings that are not the same as those available in the open market. There are various reasons for this, one being that they are able to get discounted prices as, technically, the rest of the world cannot buy the same device. Another is that you cannot find information without first identifying what the device actually is, which requires a level of technical expertise.
In the EETimes article "Mystery chips, no Portal Player, in iPod nano"), Wedbush Morgan Securities makes claims regarding what the devices are - Samsung, Wolfson, and Phillips.
By comparing die markings with other chips that have been analyzed the manufacturers of these three devices can be unequivocally proven. If the die markings are not sufficient for identification there are other features in the device's layout or cross sections that can be used to identify a particular foundry, design library, the lithography generation, which can ultimately be used to determine the company that designed the chip. It is easy to identify the Philips power management device from die markings.
While Wolfson's name is not prominent on the CODEC, the part numbering nomenclature matches many other Wolfson devices. The telltale map of Scotland is a fun feature on some Wolfson devices. The WM8750 has been used in various systems, including the LG KP4000. Other Wolfson parts with the same die have been packaged as WM8973.
The Samsung S5L8701 is a little bit trickier to identify, but by comparing the die marking to previously analyzed devices it can be determined that Samsung does indeed manufacture the device. This is the device that replaced PortalPlayer in Apple's products. Cross sections on the Samsung S5L8701 indicate an older 0.18μm CMOS lithography node with seven copper layers. In fact, this device, at 28mm2 is about 1.3 times the size of the Portal Player chips we have seen in previous iPods. Aggressive process technology nodes do not always result in the lowest cost parts. The cost improvements achieved through die size reduction do not always offset higher development expense and wafer costs associated with the leading edge process generation.
Another component of interest is the flash memory. Apple has an agreement with Samsung to use their memory devices, but given the demand, Apple has sometimes used second sources, such as Toshiba and Hynix. In the iPod nano that was torn down, Samsung's K9MCG08U5M NAND Flash was used. In this case it is a 64Gbx8 MLC device. Given the density, to accommodate the 8GB of storage capacity, this must be a stacked die.
Samsung also provides 256Mb of mobile SDRAM. Linear Tech's 4066 (Data Sheet, Technical Paper) was used. Both of these devices are the same parts that were used in the previous 4GB nano.
About the author
Gregory A. Quirk is a Technology Analyst for Semiconductor Insights. You can reach him at: firstname.lastname@example.org