As each day passes, it seems that Intel (INTC -2.10%), the world's leading CPU and semiconductor manufacturer, should scoop up NVIDIA (NVDA -8.96%), the world's leading graphics processing unit company -- or, as NVIDIA likes to call it, "visual computing."

Many may disagree, contending that Intel's own in-house GPU teams may be the demise of NVIDIA. But the results so far haven't pointed to that at all.

The obvious: what Intel gets
If Intel took out some relatively cheap debt to finance an acquisition -- at a 50% premium, NVIDIA would cost Intel about $13.5 billion, or just under a year's worth of free cash flow -- it would immediately get the following:

  • The world's best graphics IP, bar none, and it would no longer have to pay royalties to use NVIDIA's patents.
  • A business that generates about $650 million per year in free cash flow. This would be immediately accretive even if Intel had to borrow $13.5 billion at 3% interest to do it.
  • A near monopoly in the high-performance accelerator space. AMD is no threat.
  • A patent IP that may be more easily monetized under Intel's wing. It's easier to go to war with the likes of Qualcomm when you're Intel than if you're NVIDIA.
  • One fewer competitor in the mobile system-on-chip space, as well as the in-vehicle infotainment space. That could lead to higher margins.

The downside is that if Intel did this deal with debt rather than with stock, it would probably leave the company's balance sheet in a meaningfully worse position. Given the threats to its PC-chip cash cow, that may not be the best plan at this time.

What about NVIDIA?
Assuming that Intel took the company out for a 50% premium to the most recent close of $15.73, NVIDIA shareholders would get about $23.60 per share. Since NVIDIA would likely perform its fiduciary duty and try to make a sweet deal for all of its recent shareholders, it would probably really fetch more along the lines of $26 per share, or a 65% premium. What's a few billion to Intel, particularly if it is able to do this deal in stock?

The downside to doing this deal is that some investors may believe the shares are worth well north of $26. Indeed, during the 2007 peak, the company's shares actually traded for $37 apiece -- a 42% premium to this potential takeover price, which itself would be a 65% premium to today's price.

While the shares could very well get there over the next few years if the stars truly align, it'd take some pretty aggressive -- and unexpected -- free-cash-flow growth to justify that on a discounted-cash-flow basis. However, NVIDIA's patent portfolio and its brand have some very clear value that is not necessarily captured in a DCF model.

Foolish bottom line
From a strategic perspective, Intel buying NVIDIA makes perfect sense. The only issues that stand out against this deal stem primarily from Intel's inability to pay for it without either issuing a bunch of stock or taking on a hefty amount of debt. It may turn out to be easier on Intel's balance sheet to continue developing its own technology, but a deal would offer strategic benefits beyond NVIDIA's world-class IP.