Amdahl Corp, with figures due tomorrow, is on target to shoot through the billion dollar mark for its first nine months of 1987. The IBM-compatible mainframer, 46%-owned by Fujitsu Ltd, is tipped to report profits of 62 to 70 cents a share on sales of $350m to $363m for the third quarter, compared with just 15 cents a share on sales of $220m in the 1986 third quarter. For the whole of 1986, it just missed the billion dollar target, doing $966m for the year, but in the first half of 1987, sales totalled $659m, which suggests that the barrier will be breached this year with a whole quarter to go. Amdahl’s business is extremely cyclical, typi-cally reaching a peak just after IBM announces a new generation of machines and before it starts to ship them in large numbers, before tailing off sharply as it brings its own response to market, at which point it usually sees at least two very dull years of marginal profitability before its own new generation begins to make its mark. And all the while, it is having to invest heavily to prepare for whatever IBM has up its sleeve. This time ar-ound, however, IBM has had so many problems of its own to struggle with that it has perforce paid less attention to the competition in its pricing policies, being more concerned to maximise the return on its own machines. And Amdahl has struck lucky on its own account: the original 580 machines had an uncertain and difficult genesis, and were slow to become established, although they finally made their mark during the first year or so after the 3090s were announced. But this time around, the 5890, with a higher Fujitsu content, has made a big impact much earlier in the cycle, and also turned out the perform to a higher specification than Am-dahl had announced. With IBM not expected to an-nounce the successor to the 3090s before 1990, Am-dahl looks set fair for three more good years, and should make it to two billion dollars in 1988.
This article is from the CBROnline archive: some formatting and images may not be present.
CBR Online legacy content.