'As long as you have the twist, I think you're okay,' a Hong Kong-based fund manager said.

There is, for example, little-known SIM-BCD, which describes itself as the mainland's only six-inch 'bipolar' plant. The Shanghai-based company, expected to raise between US$200 million and $300 million, aims to become the world's largest provider of bipolar foundry services.

Its six-inch line is capable of producing 15,000 wafers a month. A planned second line would increase capacity by 30,000 a month, while an eight-inch wafer facility is also on the drawing board.

SIM-BCD makes chips mainly at line-widths of one to four microns in size - large by sector standards - using technology that advanced manufacturers such as Intel Corp gave up generations ago. The smaller the line-width, the more chips that can be etched on to a wafer.

But using older technology is not necessarily a drawback. Analysts at Citigroup, which is taking six-inch player CSMC Technologies public, note most mainland factories running five and six-inch lines are profitable because they mostly operate used equipment which has been fully depreciated.

Gartner analyst Dorothy Lai said: 'They have mastered the low-end stuff and they compete at very low prices. Foreign companies don't even want to deal with it. [The low-end chipmakers] don't have that much competition from foreign players.'

Another competitor in the five and six-inch segment is Advanced Semiconductor Manufacturing Corp, which is expected to raise between US$200 million and $400 million in an offer led by Goldman Sachs.

Perhaps the most talked about public share offer is the listing plan from Semiconductor Manufacturing International Corp (SMIC), which analysts say could challenge the world's third-largest contract chipmaker - Chartered Semiconductor of Singapore - for its spot.

SMIC supplies global clients such as Infineon Technologies and Elpida Memory of Japan. This pits it against the likes of Taiwan Semiconductor Manufacturing Co (TSMC) and United Microelectronics Corp (UMC), the largest and second-largest contract chipmakers respectively.

Unlike its smaller rivals, SMIC had had to spend huge sums for new equipment, posing a threat to profitability. The company has a 12-inch wafer facility under way in Beijing and wants to build two more. Such plants can cost up to US$3.5 billion each.

If the company can be said to have a market niche, it is in its lower prices compared with rival TSMC. Another selling point is global customers who manufacture semiconductors at SMIC can enjoy rebates on the value-added tax for chips sold into the mainland market.

But should SMIC eventually wrestle the No?3 position from Chartered, this would not necessarily guarantee profits. Chartered, supported by the Singapore government and the long-time holder of the position, has been bleeding red ink for three years.

Because the semiconductor industry is capital intensive, requiring continuous investment in new technology to churn out ever faster and smaller chips, market leaders such as TSMC tend to attract the bulk of capital at the expense of their smaller rivals.

And while the semiconductor market is enjoying good times - with capacity utilisation at or nearing 100 per cent among the top players - the cyclical industry could peak as early as next year. Companies such as TSMC tend to keep spending going into the downturn, preparing for when the market turns upward again. Smaller players usually lose money during cyclical downturns and rarely have the resources to spend on cutting-edge manufacturing technology.

SMIC will need to continually pour funds into new capital equipment to compete. But as the company noted in its US listing documents, the funds 'may not be available on acceptable terms or at all', and raising funds could prove difficult when the next downturn hits in two or three years. The company plans to spend US$3.32 billion over the next two years and acknowledges its share offering will not raise enough to cover expenses, meaning further issues will be necessary at the cost of share dilution.

'TSMC 10 years from now will basically have killed off all its competitors,' the fund manager said.

Still, SMIC has cemented its position as the leading mainland foundry catering to global demand, leaving local competitor Grace Semiconductor Manufacturing far behind. Grace, which also is planning an IPO and competes for global orders, had capacity to make about 10,000 eight-inch wafer-equivalents in the fourth quarter of last year, against 58,000 for SMIC.

'[Grace is] like an SMIC also-ran,' the fund manager said. 'If there is any rationality in the market, Grace should not get funded.'

Chinese plants receive government support via land concessions and easy bank loans. They typically have debt-equity ratios exceeding 50 per cent, against about 33 per cent for their Taiwanese counterparts, the fund manager said.

While investors have three choices if they want to invest in the foundry industry - TSMC, UMC and Chartered - this would not necessarily hurt SMIC's chances in the capital markets.

The fund manager said the listing could be a success on a 'greater fool basis'.