Apple expects Leopard to bring in US$140 million between its launch this week and the end of the year, a company executive hinted. That number, however, is lower than analysts' estimates.

In a conference call this week after the release of Apple's fiscal fourth quarter results, chief operating officer Tim Cook answered a question about what part of the Mac installed base will be able to upgrade to Mac OS X 10.5.

"Let me give you two numbers," Cook said. "The vast majority of Macs shipped in the last four years are able to run Leopard. Specifically, the number is about 21 million. When we announced Tiger, there were 15 million units that were eligible to run Tiger and we did US$100 million of revenue on Tiger at the first quarter of launch."

If Leopard sells to current Mac owners at the same rate, Apple will collect about US$140 million in revenues during the current quarter, which ends December 31.

Cook's back-of-the-envelope estimate, however, is low compared to analysts who cover Apple. Ezra Gottheil of Technology Business Research, for example, pegged Leopard's contribution to the quarter at US$157 million. "Add in about US$50 million in the two ensuing quarters, too," he said.

Like others, Gottheil came up with his projection by looking at the current installed base of eligible machines and comparing it to what existed when Mac OS X 10.4, a.k.a. Tiger, debuted in April 2005.

"It really is a function of the installed base," said Chris Swenson, analyst with the NPD Group. "Each upgrade sold better than the previous one, which is pretty impressive. The first two months after Tiger was launched, it ran 30% higher [in sales volume] than [Mac OS X] 10.3, and more than twice as high as 10.2."

NPD's data, gleaned primarily from retail but with a scattering of online sellers -- including Apple's online store -- showed a nearly-identical sales graph for the Mac operating system upgrades. Each edition roared out the gate the first month, but sales plunged by 60% or so the second month as pent-up demand was exhausted. From there, the line fell, then flattened.

While Swenson wouldn't put a number to Leopard's first-quarter, he wondered if Apple could top itself. "It's going to be really hard to top the Tiger launch," said Swenson. "It was such a successful launch. Apple had a well put-together strategic plan in terms of promotion and training and support."

On the other hand, Apple has an edge in 2007: Its retail chain now counts 197 stores, nearly double the 104 that had been opened by Tiger's roll-out. "Apple's retail stores helped drive sales from a customer education perspective with Tiger," said Swenson. "The customer education efforts in the stores [were] a primary reason for the success of Tiger." The additional stores could pump up Leopard's numbers even more.

Gene Munster of Piper Jaffray & Co. is even more bullish than Gottheil. He's calculated the potential bottom-line contribution of Leopard at around US$240 million, based on an installed base he estimated at 23 million -- slightly larger than Cook's number -- and an uptake rate similar to Tiger's, approximately 15%.

No matter what the numbers, analysts are expecting a warmer reception for Leopard than Microsoft's Vista received from Windows users earlier in the year. "Apple still makes its OS so you can easily upgrade older machines," said NPD's Swenson. "[The upgrade] is actually pretty impressive. There may be a performance hit on old Macs, we won't know that until it's out, but I think there's only a remote possibility Leopard will fail."

Apple will start selling Leopard at 6pm, Friday, October 26.