When shares of Advanced Micro Devices (AMD) traded in the $2's, my thesis was essentially, "this isn't a great company, but it's a stock with a story". That story was a return to profitability in the second half of 2013, and as it seemed so implausible at the time, made for a good contrarian play. Well, shares have more or less doubled since that time as CEO Rory Read announced on the most recent call that Q3 would see a return to profitability. While the initial part of the thesis has played out as the bulls have anticipated for quite a while, there were some serious red flags in this report that make me hesitant to recommend buying the shares here.
Graphics Division Now Unprofitable
The first problem I saw with the report is that in the year ago period, AMD's Graphics and Visual solutions segment reported net revenue of $367M and operating income of $31M. In the most recent quarter, the company reported net revenue of $320M and zero operating income. While I have been critical for some-time of AMD's strategy of giving away a non-trivial amount in game bundles in an attempt to gain market-share against Nvidia (NVDA) in the discrete GPU business, it is clear that this strategy is not particularly sustainable.
Further, it seems that AMD's strategy in this area is to build up significant hype vis a vis its game console wins in the enthusiast community. That is, the company tends to try to perpetuate the notion that because the next generation of game consoles is using AMD's GPUs, games will now be optimized exclusively for AMD's products (and of course, AMD pays the developers to bundle their games with their GPUs). This, while it sounds nice in theory, is unlikely to play out in practice.
The fact is, PC gaming is growing, and Nvidia has the lion's share of the discrete GPU market, which would mean that it would be incredibly foolish for game developers to attempt to lock Nvidia out of this space. Further, I have spoken with Nvidia's developer relations team on this matter and they see no signs that AMD's console wins will have a material effect on how developers treat Nvidia cards against, say, AMD's cards. Finally, much of Nvidia's R&D on the graphics side (as well as AMD's) is actually spent working directly with the game developers to optimize the code and to introduce relevant, user experience enhancing algorithms.
While AMD's console wins may help it to build some mind-share, I believe that the competition here will be largely as it has been in the past, and that the PC graphics chips will have to stand on their own in the marketplace. The fact is, AMD isn't making a dime on its graphics division, and I don't believe that "hype" is going to sell here for much longer.
Computing Solutions Is Up Q/Q, But Looks Temporary
The computing solutions segment was up 11% Q/Q, although still down 47% on a Y/Y basis, which is partly due to the broad weakness in the PC market, but largely due to market share loss to Intel. The Q/Q result is encouraging, however, but it may turn out to be a flash in the pan as the boost that AMD is getting from Temash/Kabini sell-in at the OEMs will come under pressure during Q3 and Q4 as Intel rolls out its Bay Trail platform.
I further believe that the company will continue to see market share declines in its server business (its Opteron 6300 chips are not particularly competitive, and Intel's upcoming Ivy Bridge EP/EX parts will only further push AMD's products here into irrelevance), and I expect that the high end of the PC market is likely to continue to be difficult for AMD to gain meaningful share in (I expect further losses on notebook and all-in-ones, with some stabilization in the legacy desktop as the company competes on price with its FX series of enthusiast chips).
There is hope that the firm's next generation "Seattle" SoC will reverse the company's fortunes in the server market, particularly as this is widely viewed as a green field opportunity, but there is significant evidence to suggest that Intel (INTC) will be first to market in this space as well with "Avoton" launching by the end of the year, and a 2014 part based on the 14nm "Airmont" design should further make life difficult for AMD's 28nm part (the issue of being 2 generations behind Intel is a recurring theme).
Game Console Wins Are Low Margin ... As Expected
The most frustrating part of the report is that while many of the AMD perma-bulls expected the game console wins to be AMD's savior, it turns out that the consoles are just yet another low margin business. While guidance called for 22% increase Q/Q, this is likely driven largely by the console wins, which is why we are seeing gross margin guide of 36% (although margin tailwinds in the last few quarters have been a result of selling previously written off Llano inventory). It's nice that the volume pickup due to the console wins will drive enough raw gross margin dollars for the company to breakeven, but for the semi-custom gig to really pay off, we'll need a lot more of these types of design wins.
Low margin businesses are fine and dandy, and AMD really needs to be taking whatever business that it can get at this point, but I believe that tempering expectations might be prudent for longer term investors here.
Conclusion: Back To Profitability, But Still A Lot Of Red Flags
AMD will be profitable if Q3 goes as planned, and I think that the company will hit its targets. However, it's still a long, uncertain road for the company, particularly as the fundamental story of being out-gunned and out-manned in most of its target markets has not changed. AMD, at the end of the day, is a stock for swing and day-traders, and news flow is very important; the company has been doing a great job with press releases and building hype at various events, but at the end of the day, I still don't think the fundamental question of how AMD will be competitive in its end markets going forward has really been answered, and that's why AMD still remains a speculative stock for traders, rather than a serious investment. There's nothing wrong with that, but know what you're buying.