Over the past month, the rumor mill surrounding the next version of Amazon's (AMZN) Kindle Fire low-cost tablet has heated up in a big way. A number of Internet sources have predicted that an updated Kindle Fire could be introduced any day now (and no later than September). The successful launch of Google's (GOOG) Nexus 7 tablet (also priced at $199 for the base model) has been a major catalyst for Kindle Fire rumors. Widespread reports that Apple (AAPL) will release an "iPad Mini" in September have added fuel to the rumor mill.
While design specifications are typically closely held ahead of new product launches, leaks are also rampant, and some reports have claimed that Nvidia's (NVDA) Tegra 3 processor will power the next-generation Kindle Fire. It's always prudent to take these reports with a grain of salt. For instance, there were numerous reports last spring and summer claiming that Nvidia's Tegra 2 would power the original Kindle Fire, but the design win ultimately went to Texas Instruments (TXN). It's likely that Amazon chose TI's OMAP processor because it was able to get better pricing from TI than from Nvidia. Given Amazon's desire to minimize the cost of the tablet, this was probably a good decision on its part.
However, I think this year's Tegra-Kindle Fire rumors are more credible. The Tegra 3 clearly offers superior value/performance compared to TI's OMAP 4. The proof is in the market outcomes, as Tegra 3 has won numerous design wins while OMAP 4 has really struggled recently (despite being the "reference design" for the Ice Cream Sandwich version of Android). And while TI is coming to market soon with a much improved OMAP 5 processor, it's at least 50/50 that it won't have the supply available this fall to meet Amazon's requirements. The OMAP 5 is being manufactured on the 28 nm process at United Microelectronics (UMC), meaning that supply will be affected in the near term by the industry-wide shortage of 28 nm capacity. While UMC is better off than larger rival TSMC (TSM), neither company is able to meet demand fully, according to Qualcomm (QCOM).
Thus, I would be surprised to see Amazon choose a 28 nm processor for the new device. That also rules out Qualcomm's S4 "Krait" processors. On the other hand, Amazon cannot afford to compromise on performance this time around. When the Kindle Fire came out last fall, it defined a new low-cost tablet segment. This allowed Amazon to be successful with a design that involved clear sacrifices (e.g., the lack of a camera). Today, the Nexus 7 has grabbed the spotlight as the clear choice for consumers in the sub-$300 tablet market. It offers high performance at the same price as the current Kindle Fire models. There are other competitors bringing devices to market at low price points, too, with Apple looming in the background as well. In order to maintain a leading position in the non-Apple tablet market, Amazon needs to vastly improve the Kindle Fire's specs, while ensuring sufficient supply. Tegra 3 is the clear choice for Amazon to produce a worthy rival to the Nexus 7.
Of course, a Kindle Fire design win is no sure thing for Nvidia. However, as I have argued elsewhere, the company is already a good value based on the success of the Nexus 7 and other Tegra design wins, combined with surging market share in the discrete GPU segment. Recent reports have suggested that Google could sell 8 million units of the Nexus 7 between now and the end of 2012. This already puts Nvidia well on the way to its target of at least $540 million in Tegra sales for FY 2013. A revamped Kindle Fire would send Nvidia flying past that level.
If Nvidia does get the Kindle Fire design win, the stock is likely to move up toward its 52-week high near $17. If Nvidia can meet the high end of its Q3 guidance, then the stock has upside to around $20. On the flip side, given Nvidia's strong balance sheet (more than $5/share in cash and no debt) and improving fundamentals, I see minimal downside from current levels.