Note: This was all done before the recent revelations at the Sony conference, so none of that was taken into consideration when they were doing this analysis. I'm not sure if this was ever posted, so if it was I apologize.
There's WAAAAAAY more in this document, but there's no way I can post it all. So here's a summary of the important part, very fascinating read.
Thanks to: snatches
Summary
Sony, Microsoft, and Nintendo all unveiled their next-generation consoles at the May 2005 E3 game show in the US and at the Tokyo Game Show 2005 in September 2005the
generational change in game consoles will begin, at last, at the end of 2005. The focus in
next-generation consoles will likely be on 1) convergence between game consoles and
non-game devices, 2) the use of computing power and dramatically improved 3D imagingcapabilities, 3) online game services for consoles, and 4) the online distribution of games for casual users and non-game content.
We have investigated the specs of the Sony PS3 and Microsofts Xbox 360 and performed
a cost analysis. We conclude that the PS3 will be an expensive machine to make, as it
comes equipped with high-performance components such as the Cell chip and a Blu-ray
disk drive, but we think costs could be cut massively in the future if chip yields rise and
mass production of Blu-ray disk drives lowers unit production costs. The Xbox 360s
advantage is that initial costs will be low because of the use of commodity components
found in PCs.
We expect the Xbox 360 to make major market share inroads in Europe and the US. We
forecast that the winner will be decided by the 2007 Christmas shopping season, for which
we expect leading software titles to be launched. We feel the Sony camp has to line up
some exclusive titles for the PS3 and strengthen its first-party title offerings.
There are limits to the usefulness of a single-item business model in digital consumer
electronics; tech firms need multiple earnings drivers, and the need for more earnings
drivers in the games business, with its current dual-driver structure, is especially acute. The industrys urgent task now is to establish a third earnings driver, such as online game
services or distribution, to complement hardware and software.
Implications for the electronic components industry
We estimate that the number of FC packages needed for the new game consoles will reach
more than 82mn units/year in the initial year. This is a fairly sizeable market, even when
compared with the Intel MPU package substrate market (approximately 200mn units/year), the Intel chipset package substrate market (approximately 200mn units/year), and the AMD MPU package substrate market (50mn units/year).
We surmise that the suppliers of FC packages for the high-end LSI used in the new game
consoles are as follows. For the PS3, listed in order of technical manufacturing difficulty:
Shinko Electric Industries and Ibiden for the GPU; Shinko Electric and Kyocera for the
CPU; and Ibiden and Nan Ya Plastics for the chipset. For the Xbox 360 and Nintendos
Revolution, we surmise that Shinko Electric and other Japanese makers will play a central
supply role, supported by some Taiwanese makers.
Implications for Sony
Sonys current earnings structure is such that profits from games and finance are offsetting weakness in the electronics segment. We think FY3/07 will be a grim year for the games segment and for the company as a whole, with operating losses on PS3 consoles. We are more optimistic on FY3/08 earnings, however, as the electronics giant works to cut PS3 costs and restructuring benefits emerge.
What Sony needs to do now, we feel, is to spin out businesses that bear no relevance to
network consumer electronics and devote funds and resources to technological
development that will help it differentiate itself from peers and to alliances and M&A in related industries so that it can become the new dominant force in network consumer
electronics. We will be watching with interest the reforms that CEO Howard Stringer
undertakes from here on out.
Implications for Toshiba
The semiconductor segment is a core one for Toshiba, generating more than half of its total OP, and the ramp-up of next-generation system LSI, chiefly for game equipment, together with growth in NAND flash memory, is one of the growth drivers in the semiconductor segment.
We see three key points with regards to game console chips: 1) Depreciation on logic chip
lines could be absorbed if company begins turning out game-console chips in substantial
volume, 2) the RSX graphics processor for the PS3 is a foundry product for Toshiba (i.e., it
is the subcontracted manufacturer), so if yields rise, earnings will get a boost (we expect
this to happen from late FY3/08), and 3) the problem for Toshiba is getting to grips at an
early stage with next-generation processes at the 45nm node and below. Next-generation
processes will require immense R&D expenses, and the time has come for the company to
start thinking about where it will raise the necessary funds and which firms it should
partner with.
We performed a cost analysis after examining each of the specifications. First, the PS3s Cell CPU is made from 12-inch wafers and uses 90nm processes, and we think initial production phases will entail considerable volumes of dummy wafers.
Next, we assume the effective usable area per wafer at 57,000mm2 having taken edge
exclusion (the narrow band on the outside of a wafer cannot be etched with a circuit pattern) into consideration. As we assume the Cell chip size at 235mm2, we calculate that around 243 chips can be derived from each wafer. While yields need to be additionally considered here, we forecast initial phase yields at around 30% as logic yields are typically low when lines are newly started up. We thus take 30% of an assumed 243 chips on a front-end process basis in arriving at an estimate of around 73 good chips. The back-end process comes next, and we forecast yields of around 75% during packaging
and testing processes.
We assume the number of chips derived will thus further decline from the 73 good chips yielded from front-end processes, ultimately amounting to 55 good chips. We calculate the final cost per chip to be $329 after aggregating processing costs, packaging costs, and testing costs, and dividing this total by 55, the number of good chips.
Using a similar methodology to that for the Cell, we also calculate the cost of producing each RSX (Realistic Sound eXperience) graphic chip in the initial mass production phase to be $93. Figure 7 shows a summary of costs, including the above plus around $60 for memory modules such as XDR and GDDR3, $45 for communications and other semiconductors, and $180 for a Blu-ray drive. (We have not included the cost of detachable HDDs in our calculations here, but estimate amounts of around $45 and $50, respectively, for 20GB and 40GB HDDs.) We estimate that direct material costs, mainly for components, will amount to $827. Adding overheads such as assembly and R&D expenses to this, we estimate that the initial phase cost will amount to $1,002 overall. This is more than ¥100,000 in Japanese yen terms, and we believe it highly likely that the manufacturing cost of the PS3 will be extremely high considering that it will come installed with a Cell chip that performs on a par with supercomputers and a next-generation Blu-ray disk.
That said, while we acknowledge that initial phase costs will most surely be high, we
anticipate a rapid decline in component manufacturing costs in line with improving yields and expansion of mass production. In particular, we expect that the cost of the Cell will fall
considerably from FY3/08 when 65nm processes are slated to begin. Furthermore, we
anticipate that the cost of Blu-ray disks will nearly halve in the next year and a half owing to mass production effects. In sum, we expect that costs for the PS3 will be extremely high in its first year of launch in FY3/07, and that this will likely have a substantial impact on Sonys profitability. However, if 65nm processes begin as expected and get through the first year without a hitch, then we believe that it may be possible to achieve profitability at an early stage from a hardware perspective.
Figure 8 shows a comparison of direct material costs for the PS3 and Xbox 360. In terms of the Xbox 360, using a similar methodology to that for the Cell, we calculate the cost per CPU to be around $180. This cost is much lower than compared with the PS3 Cell to the extent the CPU does not incorporate SOI (Silicon On Insulator) or new architectures.
We assume an overall cost of around $100 for graphics chips, around $48 for GDDR3
memory, $45 for a 20GB HDD, and expect that costs for DVD-ROM drives can be kept at
around $20 given their commodity-like nature, and thus estimate that direct materials costs for the Xbox 360 will come in close to 40% lower than those for the PS3.
Summing up our cost analysis from a hardware perspective, we believe the Xbox 360 has an advantage considering the potential to lower initial costs as the console uses parts based on PC components. On the other hand, the cost of the PS3 is high given that it employs high performance components such as the Cell and Blu-ray disk, but we believe there is considerable scope for ongoing cost reductions provided that the effects of higher
semiconductor yields and mass production of Blu-ray disks emerge.
Reforms Sony needs to work on
Until now, the games business model largely hinged on making hardware profitable as early as possible and generating royalties from game title software. This essentially is also the business model underpinning the PS3 and Xbox 360. That said, each generational changeover tends to be accompanied by a jump in hardware performance and increasing costs overall, including R&D. Furthermore, bolstered graphics performance also leads to sharp increases in development costs for software suppliers as they endeavor to create content that suitably exploits improved hardware performance. However, the reality is that higher software costs cannot simply be passed onto consumers via selling prices. In particular, this situation becomes more noticeable the closer to a generational changeover the industry is.
We believe the games business is approaching a critical turning point. To recap, we doubt the consumer electronics business will be able to continue generating stable earnings from business models based on single products. We thus believe companies need to have several earnings drivers. The games business is a typical business with two earnings drivers, but we believe even the games industry will face increasing pressure to develop additional earnings drivers. In other words, we believe this industry is approaching a time when it will need to swiftly establish a third earnings source in addition to the hardware
and software aspects of the business.
The Xbox 360 has a built-in Media Center Extender for Windows XP, which allows it to
playback video images (such as TV programs and movies) stored, for example, on a PC. In addition to its existing business model, Microsoft, as part of its efforts to expand Xbox Live functionality, is working to establish a new earnings driver through the bolstering of its online services. By connecting to Xbox Live, users can do such things as shop online at the so-called marketplace, download game content, or exchange data with other users.
Other services that are possible include video chat and online tournament features, and we expect over time that the company will also increasingly look to incorporate sales of various digital data. Should such a situation eventuate, we believe it highly possible that the business could develop substantially to encompass the advertising industry (if the number of hits increases) and settlement functions for the financial services industry, for example.
While the Xbox 360 camp is saying that it aims to create a new network-centric culture, we
believe rather that this could ultimately develop into a fight for supremacy over who takes the initiative of pushing this business further to seize profits from the retail supply chain and furthermore encompass online shopping and the financial services and advertising industries. One feasible strategy for Japans consumer electronics manufacturers is to adamantly continue radically whittling down manufacturing costs. However, we believe Sony needs to aggressively pursue this kind of networking business and try to get a handle on these supply chain trends in view of the resources it has and its corporate culture.
We believe what Sony needs to do right away is to cut loose businesses that are not necessary for network consumer electronics and devote funds and resources to 1) technological development that will help differentiate itself from its peers and 2) to alliances and M&A in related and outside industries so that it can become the new dominant force in network consumer electronics. We therefore think Sony is at the stage of reviewing the businesses and supply chains that have supported it hitherto. Such reforms cannot be executed by personnel that have an emotional attachment to the past. The companys recently announced restructuring plans represent a half-hearted attempt in our view and did not come as a surprise.
We intend to keep a close watch on reforms that CEO Howard Stringer implements going
forward.
There's WAAAAAAY more in this document, but there's no way I can post it all. So here's a summary of the important part, very fascinating read.
Thanks to: snatches