Computex has come and gone, and what we learned from it is that Intel seems to be doing OK at the moment, and AMD isn’t.
Let us be clear; Barcelona and Company have not one but two technical issues:
Will these chips get faster? Well, if they don’t get a lot faster than 1.6GHz, AMD will become the next Cyrix, period, end of story, thanks for the memories.
That won’t happen. They’ll get faster, but that just brings Green to the next game level. The next issue will be whether or not they’ll get fast enough to break the Intel price chokehold. That isn’t a given, or even too likely.
What was important about the Cinebench benchmark is that it indicated that while a highly cranked up K10 could match current Intel chips; it wasn’t going to beat the Penryns at any likely speed.
Yes, it’s just one benchmark, but one that if anything slightly favored AMD. Just a sign, but not a good one.
To put it mildly, a slightly tweaked K8 is not what most people think AMD has been saying about these chips, and it will be a dreadful day of reckoning when/if K10 looks like Second Banana 2.0.
But the crisis isn’t really technical; it’s not really a matter of who leads the parade on Geek Pride Day the next few quarters. It’s financial; which means the money benchmarks matter more than the software ones.
This time is different than the other times AMD has been down because AMD is bigger and can’t cut their expenses like they could before. It is truly becoming “to be or not to be” time.
Given this, for the next few months, K10 speeds can be put on the back burner. Non-technical items will matter more.
The first, small indicator will be whether or not AMD can get its 65nm mobile chips out the door. They’re supposed to show up by the end of this month, no maybe, ifs, ands or buts.
If that doesn’t happen, people are going to wonder even more whether Green can get anything new out the door, and why.
The next, much bigger indicator will be AMD’s second quarter financial results. The numbers are bound to be better than last quarter (there shouldn’t be any more ATI “acquisition costs,” nonessential spending will have been cut, there will be some asset sales this quarter or next, and probably more layoffs). However, nobody thinks AMD will make a profit from business activities any time soon.
Still, the rate at which AMD’s core businesses lose money will have direct bearing on how patient financial types will be about K10 delays and other problems.
Finally, there is the possibility of AMD being bought by some private-equity firm. Let’s say a few things about that.
Going private is no panacea for any and every company. Going private makes sense when you have a financially solid company (or at least one you can make sound with your money) that has solid long-term goals that may cost the company in the short-run, but make the company a lot more profitable in a few years.
This is not AMD, at least not at current stock prices and debt levels.
In a nutshell, we think the financial picture is going to have to get a lot worse before this becomes a serious possibility: either the stock price needs to tank, big lenders will have to be willing to write off a big chunk of current AMD debt or convert it to equity, or some combination of all the above.
What we don’t see happening is anybody spending eight billion dollars to buy a company four billion dollars in debt that’s losing money every quarter, then pump billions more in to stabilize the patient. Cut the “eight” and/or the “four” enough, and that’s when it’s likely to happen.
Private-equity firms have been getting bolder/more desperate lately, so it’s not impossible that one of them might be foolish enough to buy the hype and buy the company under current conditions.
But we doubt it. The best time for private equity to jump in would be if and when K10 has a less-than-awesome debut.
And based on what we see today, that moment ought to come this fall/winter.