AMD: Zen Server Sales Alone Could Be Worth $1 Profit Per Share In 2018

| About: Advanced Micro (AMD)


CEO Su says double-digit market share for servers. A 15% share in 2018 could be worth $2.5 billion in revenue.

Using a 55% gross margin percentage for Zen servers gives profit over $1 billion or more than $1 per share just from servers.

This does not include the large revenue increases to be expected on desktops and laptops or the new Polaris and Vega GPUs.


As AMD's (NYSE:AMD) price continues to go up, we need to analyze what AMD executives are saying in regards to "growth." We can look at their predictions, and do some comparisons with Intel's (NASDAQ:INTC) margins to see where AMD could be in 2018 with all systems "go."

In previous articles, I have suggested that AMD could earn $1 per share in 2017 if delivery dates are met - "When AMD's Earnings Projections Hit $1 Per Sh...". In this article, I will expand the horizon to 2018.

What does "double digit" server market share mean to AMD? It means growth.

In a recent meeting AMD's CEO Lisa Su made this statement:

I think, going forward, the way you should think about it is, we are a growth company. "

So the question is how much "growth" is coming for AMD in 2017 and 2018 just from servers?

Precedent would say a 15% market share increase (from zero) in a year is not impossible. In Q1 2005, AMD's server share was 7%, and one year later it was 22%, an increase of 15% in one year.

AMD has announced that its server market share jumped to 22.1% for the first quarter of this year, according to numbers provided by Mercury Research. This is an impressive number, considering AMD's server market share was a paltry 7.4% exactly one year ago"


Both CTO Mark Papermaster at the Deutsche Bank 2016 Technology Conference and CEO Lisa Su have said that "double digit" market share is expected:


We expect to attack over 80% of the TAM of the server market, the x86 server market and likewise in high performance desktop, you're going to see us just go head-to-head on high performance and again have scalability and the ability that we have in AMD to put our high performance graphics and pair it with the CPU."


we want double-digit share of that server market"

Lisa Su:

From here, it's a $15 billion market. It's very, very strong margin contribution. If I weren't aspiring for double-digit share then I don't think that would be aspirational enough."

Of course "double digits" run from 10 to 99, so I think 15% is very reasonable in 2018.

If AMD gets 15% server share in 2018, how much revenue does that represent?

Using Su's $15 billion number as a starting point and increasing revenue by say 9% for 2017 and 2018, we get $17.8 billion server market, which at 15% equals about $2.6 billion. But importantly, what will the margins be for server sales?

Well to see what that may be, we have to go to AMD's x86 duopoly partner Intel to determine what gross margins might be available to AMD from Zen servers. Intel does not break out margins by product group (segment), so we have to use some other method to come up with a number. That method would be "Operating income" by segment. This should be a reasonable proxy for gross margin, though certainly not exact as is obvious in the following chart.

We can see that DCG (Data Center Group i.e. servers) only had 30% of revenue, but that resulted in 47% of the operating income. Therefore, it is safe to assume that DCG had a higher GM percentage than CCG (Client Computing Group). We also know that Intel said the GM for all products in 2015 was about 63% and that is shown in the last column "GM$ @ 63%." The total GM for all groups at 63% is a little over $33 billion. The 2nd to last column is my estimate for GM by segment that shows DCG with 70% GM. The total GM dollars for my estimates just about matches the $31 billion calculated from the straight across the board 63%. Personally, I think the DCG margin is above 70% and CCG under 60%, but I am trying to be conservative here. The other two segments, IoT and Software are too small to make much difference.

So if we use our 70% GM from DCG and apply it to AMD, we have to include a discount for AMD's low volume and the fact that it sub-contracts its fab work to Global Foundries and others. Let's take a 20%+ haircut to say 55% GM for AMD's Zen server chips vs. Intel's DCG server chips at 70% GM.

When we apply the 55% margin to our $2.6 billion in Zen server revenue (assuming 15% market share in 2018), we get a gross margin from Zen servers of $1.4 billion. That's billions folks.

Last year (2015), AMD's gross margin was $1 billion, meaning Zen servers alone in 2018 could increase GM by 140%. Subtract maybe $500 million for increased SG&A expense (that may be high), and you have about $1 per share earnings (900+ million diluted shares) just from server sales compared to today's $10 price.

That's not including any profit contribution from desktops, notebooks, new semi-custom deals, the huge and growing VR market or new IP licensing deals. And all based upon a server market share 2/3rds (15% compared to 22%) of what it was 11 years ago.


AMD's history is one of bad news and more bad news and some terrible news after that. But under the guidance of CEO Lisa Su, the next 18 months or so are going to be good news, then great news eventually turning into greatly increased sales and profits. But there are many who are still looking backward as I described here "AMD: Don't Look Back." They are wrong.

Over the next 9 months or so, new products will roll out, hopefully to good reviews, and many purchase orders will be issued by customers. Then from 9 to 18 months out, those POs will turn into high sales and profits, much to the surprise of AMD's critics. And along the whole route, AMD's price should go up, albeit not without the occasional dip.

So the investment question is this: do you agree with Dr. Su that AMD is a "growth" company? If you do, then the question becomes how much growth? I think a doubling of revenue in 2018 (from 2015) is entirely feasible, and if/when the market comes to that realization too, the stock price will be much higher than it is today.

As for a buyout possibility, sure that's OK if the price is right. If not, no problem, I will wait patiently for the future to develop, as I have described many times in the past. Either way, I will make a nice profit.

AMD is a strong buy with a price target of at least $15 in 2017 and $20 in 2018.

If you found this article to be of value, please scroll up and click the "Follow" button next to my name.

Disclosure: I am/we are long AMD.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

About this article:

Author payment: $35 + $0.01/page view. Authors of PRO articles receive a minimum guaranteed payment of $150-500. Become a contributor »
Tagged: , , , Semiconductor - Broad Line, Authors Pick, Expert Insight
Problem with this article? Please tell us. Disagree with this article? .