Top ten predictions of semiconductor industry in 2

2022-07-29
  • Detail

Top ten predictions of semiconductor industry in 2008

happy New Year! 2008 has just begun to appear in front of the electronics industry. According to the latest industry data, uncertainty has already permeated the air. Industry forecasters seem to have different views on the outlook of the whole semiconductor and IC equipment in 2008 and beyond. To help clarify the specious views in the market, I have released my own chip market forecast and other predictions for 2008

1. Please forgive my pessimism, but I think that compared with 2007, the semiconductor industry will achieve zero growth in 2008. In 2008, there were some ominous signs, such as the subprime debt crisis, soaring oil prices and the US presidential election, which may bring uncertainty to political parties

regarding microeconomic issues, this is my real concern: DRAM market is still oversupplied. I don't think there will be any signs of improvement in DRAM market in 2008. There are too many capacity lines, especially in Taiwan. Meguiar and Qimonda are suffering losses. Hyundai expects a deficit and Samsung will stop growing. The problem is that Microsoft Vista has not become the driving force of DRAM market demand as previously thought. In addition, memory, in particular, was originally supposed to push DRAM to mass consumer goods. In fact, this did not happen at all

meanwhile, the NAND flash market remains uncertain. I think the sorrow of ASP will continue. The PC market, which drives the demand for microprocessors, is still boring and will show dismal growth in 2008. However, virtualization is still of interest

the market for driving DSP, nor flash memory, RF and other devices will also be calm. How many people love the iPhone these days? There is also a lack of change in the consumer electronics market. I'm still looking forward to some new and interesting killer apps. There are some highlights: the simulator market will revive, as will WiMAX and 4G. However, we need more killer applications

2. Please forgive me again, but the semiconductor equipment market is expected to have another downturn in 2008. Contract factory expansion and capital expenditure - in my opinion - are likely to decrease sharply

my prediction is that the IC equipment market may decline by 20% in 2008

I hope my judgment is wrong. But that is the reality. Wafer factories do not have so many expansion or purchase plans. DRAM players are suffering. NAND manufacturers are expanding their production, but the market will grow slowly. Intel may buy more devices, but the company has learned to use them more effectively

3. Here is a brainless idea: more mergers are expected in the field of semiconductor devices. In the past few years, we have seen a lot of mergers. This year, I think we will see some really amazing news

for the past twoorthree years, I have predicted that applied and Canon will form a joint venture in the lithography field. This year, my prediction may come true. Canon lags behind in the competition for immersion in the lithography market. Applied wants to enter the lithography market, so I'm afraid this pairing can only be realized in the Underworld (impossible). Applied also pays high attention to solar energy these days. No wonder it has grown in this area. In 2007, applied merged some solar equipment companies. In 2008, it may try to seduce Amtech, BTU, spire or Oerlikon for this matter

I think the time is ripe for Novellus to be taken over. Tokyoelectron Ltd is expected to compete for Novellus. For the long-awaited Lam Novellus merger, that is impossible. Forget it! Lam has just acquired SEZ. Therefore, that tells me that the wafer cleaning industry has been crowded. Akrion, FSI and other companies may be the targets of acquisition. I think KLA Tencor may acquire nanometrics. Rudolph may also compete to acquire nanometrics. Has the takeover war begun

in the field of automatic test equipment, credence has no hope. I believe Teradyne will swallow them up to gain market share of logic automated test equipment. Verigy may also compete for the acquisition of LTx

4. When it comes to chip manufacturing, far ultraviolet (EUV) lithography will die slowly and painfully in 2008. EUV was originally intended to process wafers at 65nm nodes. Then it was kicked out by the 45nm process. Then came the 32nm process technology. Now, the latest progress target is 22nm. I know that ASML and Nikon are trying to develop EUV technology, but it seems that EUV technology has not made real progress

at present, the whole industry is facing a huge problem: do we need to invest more money in the expensive EUV technology? Another question is: what will chip manufacturers do during the production of 32nm and below process nodes? Unfortunately, there is no clear answer. I have no answer. EUV is a bad guy! Nano printing is not ready yet. Multi beam electron beam is still a scientific research project that can not be applied

it is hoped that 193nm immersion and dual-mode devices can solve the urgent problem at least at present. Reality of the industry: we need to make a breakthrough in the high-scale mobility of 193nm immersion devices

5. When will IBM spin off its semiconductor business unit? This is what I and many others have been trying to figure out. At the same time, 2008 may be a year of major changes

in the past few years, IBM has transformed itself from a hardware company to a software and service provider through reform. Is the semiconductor division suitable for this strategy? The answer is ambiguous

on the one hand, IBM is still the main ASIC, processor and agent, which clearly stipulates that outdoor tobacco advertising workers shall not be set up. The secret "condiment" of the blue giant is that the chip department is also a major IP provider. In fact, some people call IBM "fabless foundry". However, on the other hand, I predict that IBM may spin off its chip unit and use the investment of private equity companies to form a joint venture. I know that the financial market is not healthy. How about IBM and franciscopartners holding 50% equity of ibmmicro respectively

6. Speaking of IBM, it was quietly released during the holiday curve display season. At that time, the blue giant licensed its 45nm process technology to China's foundry, SMIC

what does this tell me? SMIC cannot develop its own 45nm process. It does not want to be sued by TSMC for alleged intellectual property piracy

the real answer is that it also means that SMIC positions itself as the receiving target of a company. SMIC, which has been losing money repeatedly, has hinted at privatization

in 2008, I predicted that licensed semiconductors and SMIC may form a joint OEM company, which may threaten TSMC and UMC, the big players in Taiwan. Chartered Semiconductor is a member of the OEM Club of IBM. At present, SMIC is trying to get involved in the club

7. I can't see what benefits Freescale has gained by taking the path of private equity. Freescale is running a deficit because it has become a private company. There are many such things

Freescale is seeking another public offering. My prediction is: I am looking forward to the collapse of Freescale to save the company. Infineon may acquire Freescale's wireless and/or wired and/or automotive chip division. Or, for that kind of thing, Infineon may buy the whole company. NXP or ST may also acquire some of its divisions

8. I still don't see whether amd can get out of trouble in the downward spiral. It is losing a lot of money and market share

the company recently received some cash from Mubadala Development Co., Ltd. in Abu Dhabi, but that was only a temporary measure. So who will save AMD in 2008? One website recently suggested that Apple might be the suitor. That's a lie! What a disgusting match? Why does Apple need a losing chip company

I think amd has three possible prospects:

a) IBM buys amd shares. IBM is AMD's technology partner, and IBM may want to compete with Intel again. A few years ago, IBM tried to compete in the x86 processor market, but the blue giant ended up with bruises all over the process

b) amd follows the path of private equity. Private equity is a practice of the past and cannot guarantee success. Ask NXP or Freescale for advice

c) Samsung may want to re-enter the processor business and acquire AMD. For some time, Samsung has tried to compete with alpha chips in the processor business. Remember that failed? I suspect that even if Samsung wants to compete with Intel in the processor market, Samsung is a memory manufacturer, which seems logically unreasonable for Samsung

therefore, my prediction may be: A or B

9. I think something will happen between micron and Qimonda, the two major memory manufacturers

first, let's talk about micron. The company is at a loss, and micron has hinted that it is seeking some kind of partnership. Like AMD, micron may take the path of private equity. Perhaps micron may combine its memory operations with the German company Qimonda. This is a strange combination. However, the DRAM market is unpredictable, and the two companies can not afford the risk of deficit throughout 2008

if the combination of the two companies is not an option, Qimonda may finally choose to merge with Nanya, a DRAM partner in Taiwan. May take these types of speculation to compete with Samsung

10. Finally, we should pay attention to some things that have happened in Japan. In a difficult business environment, there are too many slow-moving IDMS. However, Toshiba is an exception to this rule. They do very well

Sony is more or less a loser in the semiconductor field. Who's next? Epson, NEC Electronics, Oki, Renesas and Sanyo cannot afford another recession in the semiconductor industry. Sanyo has tried to sell its semiconductor business unit before, but it failed. I expect Sanyo to abandon its semiconductor unit. OKI and Epson will maintain their semiconductor business, but only as small players

what left me was stern who said, "the mold capital is the driving force of Renesas and NEC Electronics. In the distant future (after 2008), I think Toshiba will merge NEC Electronics, and Renesas will survive as a fabless company for a long time.

in the next two years, the professional DRAM manufacturer Elpida and its partner Powerchip in Taiwan will merge

Copyright © 2011 JIN SHI