Learn this article by heart, know well the difference between a micro controlller and dsp who makes what and which are the one's that are going to make big going forward, knowing who does what in semi's business is the basic art many of us would consider a sector as important enough.
On a lighter note come out of self imposed hibernation, the reason I requested not to get involved on this recent issue and make comparisons on my bad and good or vey good calls was simple, no call or no post is bad or good if you have the guts to come out and say what you need to say with your identity known to the world, for your future only one hting counts either trade in your name post in your name or forget it, that whole thing of making yourself known and exposing your self to criticsm is the essence it keeps on gaurd and that is what I wanted for myself and have aksed you guys to do.
These kind of flaring up is normal part of the art and the training we going through, you are going to be in and taking heat rather getting worrried about obtaining insulation from the third degree burns, this is the art you got to learn. No honor is ever lost neither respect is compromised if someone is persistent to deny the reality and chase shawdows or highlight out of context part or reads from the post, only one honor that of total integrity and keeping your thoughts free of charge is important, on that count if you loose, you loose everything.
The honor of integrity and dedication to a concept, go back four years on this thread, on that count you would see that most if not every one have appreciated this work. You could be proud of your father but on these minor skirmishes never be defensive, these kind of rmearks only make me stornger and also these will help you, this is the real world and you got to know to survive you need to gain strength from knowing your weaknenss.
Markets are a treacherous business and the lines between a good call and a bad call are quiet blurred. I hope back to uni does not mean out of the markets and rest assure I know you are guys reading every post here I saw your post on your favourite site, great call... hope to see you back here ..love from dad...
<<Semiconductors: Where the Growth Is
By Thomas W. Smith, S&P Technology Analyst
NEW YORK, Sep. 15 (Standard & Poor's) - The semiconductor industry expansion that began meekly in late 1998 -- and accelerated dramatically in 1999 -- is continuing at a high intensity in 2000.
From a plant utilization perspective, things could hardly get any hotter. The utilization rate for all integrated circuit wafer fabrication plants, or "fabs," worldwide was 95% in the second quarter of 2000, according to the Semiconductor Industry Association (SIA).
Utilization was up slightly from the first quarter of 2000, and up significantly from a level of 89% in the second quarter of 1999 and from the 81% level that marked the cycle's low point in the third quarter of 1998.
STILL STRONG. At present, the chip cycle also looks strong from a sales perspective. In July 2000, global sales of semiconductors were $17.32 billion, up 4.0% from June 2000 sales of $16.65 billion, and up 50.3% from July 1999 sales of $11.53 billion, according to SIA data.
By region, Asia/Pacific and Japan showed stronger year-over-year growth than did the Americas and Europe. Fortunately for the chip industry, economic growth is generally robust in all major markets and likely to stay that way in the near future. Of course, strong economies mean strong sales of electronic goods and therefore high chip sales.
I expect fab utilization rates to remain high into mid-2002, when the next generation of fabs begins to produce at high volume. The new fabs began to be built in late 1999 and early 2000. It takes about two years to build a fab, and perhaps a couple more quarters to work the kinks out of new production lines.
Right now, though, there is a lively debate on the Street as to when significant new capacity comes on line and utilization rates fall. Some say it may occur as early as the fourth quarter of 2001; others argue that broadening end-markets for chips may lift demand sufficiently to prolong the period of tight capacity to 2003. Whenever the new capacity finally does come on line, it is apt to mean lower chip prices, as supply begins to exceed demand. This will be an important turning point in the chip cycle.
HIGH-GROWTH. In June 2000, the SIA released its midyear forecast for global chip sales through the next four years. As shown in the table below, the trade association estimates worldwide total semiconductor sales will grow 31% this year, to a level of $195 billion. Sales are then expected to rise 25% in 2001, 14% in 2002, and 12% in 2003 to a level of $312 billion. Generally speaking, private consultants' estimates run a little higher than SIA estimates, at least for 2000.
For the benefit of investors seeking chip categories that are estimated to grow above the overall trend in chip sales, I have chosen to include the numbers for several kinds of chips that are seeing exceptional growth. All of these hot categories have estimated growth above the overall trend in 2000; all except analog are above the general trend in 2001; most categories are superior in 2002 and 2003 as well.
Estimates for year-over-year worldwide sales growth in dollars for all semiconductors and for selected high-growth categories:
2000e 2001e 2002e 2003e
CHIP TYPE
Optoelectronics 32.2 29.1 25.5 23.0
Microcontrollers 37.4 28.5 18.0 15.6
DSP 55.0 36.0 29.0 20.0
Analog 34.5 23.2 17.3 10.7
Flash EEPROM 116.0 34.0 9.0 1.0
MOS DRAM 41.5 44.3 5.0 13.0
FPLD MOS 42.9 38.5 33.5 26.4
Worldwide total semiconductor 30.6 25.3 14.1 12.1
e-Estimated. Source: World Semiconductor Forecast 2000-2003 (June 2000), by Semiconductor Industry Association.
TOP PICKS. Given that there is at least a year-and-a-half of strong chip pricing conditions ahead, I believe investors should consider overweighting the semiconductor sector generally, and consider picking a few companies from the hotter categories.
Optoelectronics: Briefly, this category includes displays, lamps, couplers, and other opto-sensing and emitting semiconductor devices (excluding liquid crystal displays, incandescent lamps, etc.). The term "optoelectronic" signifies a combination of using photons for data transmission and electrons for switching. These chips are used in communications equipment. Many of the companies operating in this area have small operations and carry high valuations, so we hesitate to recommend them on a value basis despite the good looking growth potential.
However, note that a high-performance analog chip company we like, Analog Devices (ADI; S&P STARS rank , buy), with a market capitalization of $32 billion and an annual revenue run over $2 billion, gets about 3% of its revenue from optoelectronics. If that portion of ADI's operations were valued at multiples comparable to those for some of the standalone optoelectronic chipmakers, ADI's share price would run noticeably higher.
Microcontrollers: These small, relatively inexpensive chips perform embedded control functions within an overall electronic system. For instance, they may perform control functions in TVs, VCRs, kitchen appliances, or automobiles. The average household has about 150 microcontrollers, and that number is rising as our gadgets proliferate and get "smarter." While Motorola is a big player in microcontrollers, it is not a pure play. For that, we like Microchip Technology (MCHP; ), the number two player that is consistently gaining market share. Because microcontrollers are used in a very broad range of end products, Microchip Technology displays steadier earnings performance than many chip companies.
Digital Signal Processors (DSPs): As their name implies, these chips perform digital signal processing tasks in wireless phones and many other settings. In a cell phone, a DSP chip is typically surrounded by a set of analog and mixed signal (analog and digital) chips in order to create a system that receives a voice signal (analog), gets it converted to digital format, processes the data in digital form, and converts and resends the signal in analog format.
Texas Instruments (TXN, ) aims to be the biggest player in DSP. Lucent, Motorola and Analog Devices are also players in DSP. We like the strategic switch Texas Instruments has made toward DSP and away from commodity memory, and maintain an accumulate recommendation on the company. TXN is a large cap stock, offering investors liquidity. However, TXN usually trades at a premium valuation compared to its peer companies and has some slower growth legacy businesses.
Analog: This category includes amplifiers, interfaces, data converters, and other chips typically used in processing analog signals such as sound, video, pressure, and acceleration. As noted above, analog chips are critical in converting real world analog signals to the digital signals of the computer world. There is strong demand for high-end analog chips to handle complex tasks in communications, industrial, personal computing, and consumer markets. High-end analog chips have a lot of proprietary design content and carry high margins and have long product life cycles.
These factors contribute to outstanding growth and financial performance at Maxim Integrated Prod (MXIM; ) , as well as at Analog Devices. Texas Instruments is the leader in low-end analog chips and will be competing more vigorously for high-end business with the August 2000 acquisition of Burr-Brown, a company with expertise in high-performance analog.
Flash EEPROM: These chips are a type of EEPROM (electrically erasable programmable read only memory) in which the memory pattern is erased by large arrays of bits rather than by smaller aggregations, such as bit by bit. They are nonvolatile memory devices that retain stored information when the electrical power is interrupted. Flash memory is used in a variety of wireless appliance functions, and has experienced very high demand in 2000 as a result of a surge in demand for cellular phones.
One of our buy recommendations, Atmel Corp (ATML; ), has about a fourth of its revenue from flash memory. The company also makes logic chips and, overall, has a broad range of products. We expect Atmel to experience significant gross margin improvement over the next few quarters.
MOS DRAM: Metal oxide semiconductor (MOS) dynamic random access memory (DRAM) is a cheap and popular type of memory chip. End markets include personal computers. Capacity for DRAM was severely overbuilt in the mid-90s. Subsequently, the industry consolidated, making it easy to identify Micron Technology (MU; , accumulate), and Korean manufacturers Samsung and Hyundai as top players. When pricing for DRAM, which has commodity characteristics, goes up, Micron shares can really fly up. Likewise, the shares can drop swiftly if pricing deteriorates.
Alas, it is devilishly tricky to forecast DRAM pricing trends. Micron consistently has an exceptionally wide range of analyst estimates, reflecting uncertainty about DRAM pricing direction. Judging from SIA forecast numbers for this category, DRAM producers may have some good innings in 2000 and 2001, but new capacity may spoil the party by 2002.
FPLD MOS: These logic devices are a special class of "semi-custom" chips that are sold as standard catalog products that can receive final programming by the customer rather than the chipmaker. They permit an original equipment manufacturer (OEM) of electronic equipment to keep chips on hand that can quickly be programmed to handle a rush job, thereby reducing the OEM's time to market. OEMs will pay up for the time savings capability. In the FPLD area, we like Xilinx Inc (XLNX) as a and Altera Corp (ALTR) as a recommendation. Both are fabless semiconductor companies. That is, they are design houses that outsource chip production to foundries such as Taiwan Semiconductor (TSM; ), which we also like. In fact, Taiwan Semiconductor recently issued a financial forecast report indicating earnings for 2000 ahead of consensus.
>> |
