"Regarding our ASIC business based in France, as you may recall, we announced in February that we were pursuing strategic alternatives with respect to our ASIC business and related manufacturing assets, including a potential sale.

We are far along in the sales process and while there is always a great deal of uncertainty regarding M&A transactions particularly in this economic climate, we do expect to present a proposed transaction to the appropriate workers council representatives this quarter. Based on this timeline, we are hopeful to complete a transaction in the early part of 2010."

Is Atmel finally going to address the "Elephant in the Room" as the analyst Craig Berger said last year.

Some Data Points:
1) the gross margins by Business segment assuming normal Fab utilization rates 80%-85%

Non Volatile Memeory - high 30's.
RF/AUTO - 40%
Microcontroller - 50%+
ASIC - 15% to 30%

2) Atmel still has too much Fab capacity. Rousette fab is horribly underloaded. The Colorado Springs fab can handle all of the Non-Volatile and Microcontroller demand for the forseeable future plus the new RF/AUTO designs as the 3 year German Foundry wafer purchase agreement winds down.

3) Gross margins on New Microcontroller chip designs. Reference the new MaxTouch chip-MT224. It's in a 5mm x 5mm package so the die size is probably something like 4.5mm x 4.5mm conservatively. On a 6 inch wafer this would be 1100 potential die. On a mature fab process like Atmel's Colorado Springs process, I would expect a probe yield of 90% and package yield of 95% yielding 940 shipable parts. Price is $4.75 each in million unit orders. This would generate approx. $4500/wf. A huge fab almost completely depreciated like Colorado Springs loaded normally is bound to generate a wafer cost of no more than $500.00. Even when you through in a lot of other accounting stuff the gross margin has to be north of 70%.

4) Microcontroller growth-SIA states Atmel is the fastest growing 8 bit Microcontroller company in the world. GAGR of 21%. Last qtrs sequential growth was 18%. 32 bit sequential growth was 36%.

Does Atmel have a buyer for Rousette Fab and the ASIC business. With the low margins for this business it would have to be a low price, but the Fab is a decent Fab. Middle Case .8X sales-$250M plus $250M for the building and the equipment. It would have to be from someone that could fill the fab to capacity which has never happened. 5000K wafers per month. A foundry?

Atmel's Total revenue without the ASIC division last qtr was $238M. It's not unreasonable to expect growth to $300M by mid next year. Microchip has stated that they intend to grow 26% next year and Atmel is growing at a slightly higher rate. At gross margins of mid 40's. And an OpEx of 30%, this would generate operating income of 45 million a year.

With 450 million shares outstanding this would be $0.40 cents on a yearly basis.

What if they applied the purchase price of $500M to a stock buy back. At $4.50 a share that would reduce the shares outstanding to 340 million yielding operating income of $0.53 on a yearly basis.

What if they have no buyers and simply exit the Asic business and sell the fab equipment to TSMC and the building to someone else....i.e. Like N. Tynside. In my estimate this would generate $250M. Rousette Fab is more than twice the size of N.Tyneside and is fully facilitized. Knock off $50M for worker's payments required by French law and you still have $200M. Atmel has almost $400M in net cash, so they still could buyback at least $250M in stock. This would leave 400 million shares outstanding and $0.45 annual earnings per share.

Worst case-$0.45 minus 0.05 tax would be $0.40 x a PE of 15 would suggest a stock price of $6.

All my Opinion of Course.