Despite all their caveats about the lack of a convincing operating system apart from Apple Computer Inc’s Macintosh System for PowerPC (CI No 2,4XX), all but one of the analysts we talked to displayed the usual inconsistency of their trade and were generally optimistic about the PowerPC’s prospects. Paine Webber’s Stephen Smith is the exception. He says the whole PReP party will fail if Apple does not buy into the PowerPC Reference Platform. He perceives Workplace as being very late and says his biggest concern is that IBM will use the only lever it has got and start cutting prices on the Power Personals early next year. At the other end of the spectrum sits Richard Buchanan of Forrester Research Corp, who says: PowerPC will be very successful both on the desktop and embedded systems. Successful, in this context, means he is betting that PowerPC will be found in a third of personal computer desktop devices by the end of the decade and will dominate the embedded market. Competing RISC chips, he says, will basically get blown away. While admitting that Alpha has arguably better technology, he is dubious about the sales and marketing strategy Digital Equipment Corp is following with the chip. Sparc and MIPS are both in trouble because they are failing to infiltrate the right markets, according to Forrester’s Buchanan. Their success, he says, can only be achieved if they find strong strategic partners to team up with in much the same way that Apple Computer Inc, IBM Corp and Motorola Inc did. However, the companies concerned are strangely reluctant to do the innovative alliance work required, he says. He nominates Hewlett-Packard’s Precision Architecture RISC as the most serious competitor, though it is hobbled by its current 32-bit limitations.

Buoyant

Buchanan is even buoyant when it comes to software. Microsoft Corp has repeatedly said that Chicago (Windows 4.0) will be an iAPX-86-only operating system, however, he believes that at some point over the next six years we will see Windows spread to PowerPC. In the long run it is in Microsoft’s clear interest to have all their operating systems running on all chips, he points out, suggesting that over the next six years there will be a general merging of Windows and Windows NT technology, perhaps culminating in a hardware-independent version of Windows. Despite their difference in outlook, Smith and Buchanan’s views converge on one point: the importance of Intel Corp in all this. Buchanan believes that, in the short term, Intel price cuts and the size of the software base will slow the progress of PowerPC. But it won’t stop it, he says. In the long term he acknowledges that the Hewlett-Packard-Intel alliance is a big unknown. Smith is blunter: The real issue is the pace of change of the Intel architecture, he says. But in general, it seems the hardware question has already been answered. We have warned against believing anyone who said they knew how the battle between Intel and the PowerPC would turn out. Now, at least, some of the parameters have been tightened. Manufacturers have shown that there is no black art involved in designing and building PowerPC motherboards. Bench tests have shown that PowerPC can deliver improved price performance compared with Pentium systems, though Dominic Ricchetti of Dataquest Inc is concerned that we have not actually seen this translated into finished systems yet. Finally, the number of manufacturers that have signed up to build boards or machines is really rather impressive. What they are waiting for is the software, and that still remains an unknown. However, the battle should enter full flow during the first quarter of 1995 and we will report again then. And of course, our delivery schedules never slip. – Chris Rose, PowerPC News