Sometimes it pays listen to the voice in your head at the expense of what numbers and company managements may have to say. There have been plenty of calls out there in 2012 to buy semiconductors in the hopes of a recovery, and in particular to buy Atmel (Nasdaq:ATML) in the prospects for good growth in the touch controllers that help drive the user-interaction experience for smartphone and tablet users. And yet, worries about the ongoing commodization of this category have kept me on the sidelines and have cut the value of Atmel roughly in half this year.

Discount Brokers Comparison: Your one-stop shop for finding the perfect broker for your investments.

Not Much Good News in Third Quarter Results
Atmel's numbers didn't look all that bad relative to published estimates, but there wasn't much in the data that was especially good either. While some high-profile tablet slots could keep hope alive in touch and ongoing profitability speaks well to the company's expense structure, this is still a "show me" story.

Revenue fell 25% this quarter compared to the same quarter last year, with a 2% drop in sales from the second quarter. The weakness was pretty even; microcontroller sales fell 25% from last year, while memory dropped 33%, ASICs fell 21% and RF/auto dropped 17%. The bright spot for Atmel was actually in touch; mid/high single-digit sales growth actually fueled another quarter of sequential growth for microcontroller sales.

Margins continue to come under pressure with the weakness in sales and utilization. Gross margin fell about seven points from last year (on a GAAP basis), and nearly a point from the second quarter. Operating income was a little more complicated; Atmel logged a big year-on-year decline in profits (and a mid-teens sequential improvement on an adjusted basis, or 6% on the company's non-GAAP basis), but the company cut expenses more deeply than expected and actually did all right relative to consensus at the operating margin line.

SEE: Understanding The Income Statement

Want to Understand Touch? Good Luck
Atmel's prospects in touch continue to be pretty murky. Rival Cypress (Nasdaq:CY) reported that TrueTouch revenues were "up slightly" in the third quarter, but it's looking like this company's second-half growth is going to come in about half of what was originally expected. That puts Atmel and Cypress in more or less the same boat, with touch revenue down somewhere in the low teens for the year.

As much as there's a lot of Atmel-versus-Cypress talk concerning touch, I don't think these companies are their biggest respective problems. Both are losing smartphone share to Samsung (largely to Melfas), and both have a lot to fear from the commoditization of touch controllers that is pushing ASPs in smartphones below $1 (tablet ASPs are higher). Likewise, both stand to lose if companies such as Broadcom (Nasdaq:BRCM) continue to integrate touch control into more a comprehensive system on chip offerings.

Atmel still seems to have an edge in terms of its up-market position (where ASPs are higher). Atmel controllers are in both the Microsoft (Nasdaq:MSFT) Surface and Amazon (Nasdaq:AMZN) Kindle Fire HD, but I worry that these are a large percentage of Atmel's touch revenue and vulnerable to cheaper replacements down the line (likewise for Cypress and Amazon). Elsewhere, I'm not so confident anymore about the potential for Windows ultrabooks, and I think Atmel could struggle to find designers willing to pay up for the benefits of XSense technology.

SEE: A Primer On Investing In The Tech Industry

The Bottom Line
With all of that said, there could still be a value call to make on Atmel. Although the semiconductor industry has had a pretty rough year and third quarter, Atmel's downward guidance revision wasn't so bad. Netting out the sale of the flash memory business, it looks like Atmel management believes revenue could fall about 3% sequentially, with possibly two more points of gross margin at risk. That's not great, but it's not bad relative to what other chip companies have been saying. What's more, Atmel has been active on the product development/launch front with microcontrollers and MICUs, and a general industry recovery in the second half of 2013 would be a big help.

Right now I'm looking for Atmel to log compound revenue growth of negative 1% through 2017 and only about 2% growth out to 2022, with a substantial improvement in the free cash flow margin into the mid-to-high teens. That's an ambitious target, but if Atmel can do that (and produce the resulting free cash flow growth of about 10%), fair value is above 8%.

Keep in mind, however, that that would represent record levels of free cash conversion, and a more sedate outlook for low teens free cash flow margin would drop the implied growth in the mid-to-high single digits and the fair value down to about $6.25. Even that fair value is about 25% above current prices, however, so Atmel may appeal to aggressive investors looking for a badly-beaten chip turnaround play.

At the time of writing, Stephen D. Simpson did not own any shares in any company mentioned in this article.