I'm still kind of tied up so don't have time to do an in depth analysis on Intel's earnings right now. A couple of things that caught my attention:
1. They exceeded targets slightly which was kind of in line with my overall prediction that they would be on track to hit their target. They exceeded the mid point by 100 mln $ which is ok. But the bigger take away is Intel is focussed on executing again and we will begin seeing mometum return to them in Q4.
2. Market share - Otellini is now reserving market share comments till folks like Gartner/IDC confirm in 2-3 months. I guess he learned his lesson from the Q1 fiasco not to trust his sales guys on this number. He did however say he believed they gained market share in DP server (which is the volume) with Wodcrest. Let's see if Hector provides an indication or we'll have to wait some more for this data.
3. Q4 gross margins - this was the most disappointing piece that margins would remain around 50%. This either means the price war will continue or incremental costs in the factory...possibly due to the 45nm costs kicking in. There was some discussion on this during the analysts call but I didn't have time to hear it completely.
Come back on the weekend. I'll be reviewing Intel and AMD's results in some more detail. Meanwhile, I've had to help Sharikou by correcting some basic flaws in his calculations on capacity and cash flows.
My comment on capacity can be found far down here:
http://sharikou.blogspot.com/2006/10/charlie-showed-some-wisdom.html
My comment on cash flows will show up here if the post is authorized. In all fairness, I think only one of my posts was not put up so I don't want to make it appear like all my comments are being filtered:
http://sharikou.blogspot.com/2006/10/another-look-at-intels-balance-sheet.html
Hopefully, see ya on the weekend.
Subscribe to:
Post Comments (Atom)
14 comments:
AMD gross margins tanked..Stock is 13% down after hours.
AMD missed on earnings as well, once you take out the one-time tax benefit. Don't expect that to come up on that other blog.
AMD unit growth 18%, market growth under 7%. Means your gonna have to wait a little longer for that market share loss prediction.
My prediction: AMD exits 2006 with >26% market share.
Sharikou certainly needs your help, he can't help but fudge numbers in AMD's favor. Keep up the good work, nice to see news from the other side.
After have read AMDs finacial report I can safely say that this was a horrible quater for AMD.
Maybe this will get SHakira to shut his fundamentalist mouth.
/mork
PS:Keep blogging 180*
Actually, I can say that AMD is doing better than Intel. My concern was that the volume share for AMD was just a temporary bump caused by reduced volume from Intel. However, it is clear that AMD has now consolidated this share and is not likely to lose it. It is true that AMD's margins are down but they are only down 12% from the their high in Q1 06 while Intel's are down 20% from their high in Q4 05. AMD's processor revenue increased 10.1% from Q2 06 while Intel's only increased 8.8%. Compared to last year Intel's processor revenues are down 21% while AMD's are up 33%.
Simply put, this is the worst year that Intel has had since 2001. These are also the highest inventories that Intel has ever had. Intel has been steadily piling up inventory since Q3 05. Today, Intel's inventories are 54% higher than a year ago while AMD's are less than half.
Intel's gross processor revenues for 2006 will be much lower than what they were in 2005 and only about equal to what they were in 2004. However, AMD's gross processor revenues will be double what they were in 2004. Frankly, I'm not sure how nearly zero growth is better than 100% growth. This makes the outlook for Intel in 2007 even worse. Intel will likely regain the revenue that it had in 2005 while AMD is already ahead of the 2005 revenue. AMD's costs will steadily drop as capacity on FAB 36 ramps which is not a cost benefit that Intel enjoys. This should be enough to pull up AMD's margins even if the cpu prices stay the same.
Overall, I am certain that Intel has hit bottom and is now slowly growing again. However, I do expect AMD to grow faster in 2007 and continue to take share. I would guess that AMD will be around 25% by end of 2007.
I'm all in favor of healthy debate. It seems that we're in a situation where we're going to go quarter to quarter to see who is on top of whom. That being said, I'm sure seeing an awful lot of jaws yammering about how AMD will make mighty gains with Fab36 transitioning to 65nm. What nobody is talking about is the bomb they dropped by announcing that Fab30 will be going partially off-line in Q2'07 for a year. It completely shocked the analysts evidenced by the transcripts of the AMD Q3 2006 Earnings call found at the following link.
QUOTE:
Joe Osha - Merrill Lynch
I'm wondering if I can drill down a little bit on the manufacturing output. Can you give me some sense roughly for the current quarter and then for the fourth quarter, what portion of your output is coming from Fab 30 versus Fab 36 versus Chartered? Obviously, units you probably don't want to talk about, but maybe just in 200mm equivalents?
Dirk Meyer
I can certainly give you a feel for where the wafers are coming from, from our internal factories. We started ramping Fab 36 in Q4 of last year, and it's roughly a two-year ramp to get to the 20,000 a month point. We're on a schedule that says that we are at about the 10,000 a month output level now.
Joe Osha - Merrill Lynch
That's 300mm you're talking about?
Dirk Meyer
That's right. It's 300mm. Fab 30 is really pretty stable; 200-mm starts --
Joe Osha - Merrill Lynch
So it's around 28,000, 30,000 a month?
Dirk Meyer
28,000 to 30,000 200mm. We don't, I think, want to be to specific on how we are loading Chartered at this time.
Joe Osha - Merrill Lynch
But it's a much smaller number, yes?
Dirk Meyer
Very much smaller.
Joe Osha - Merrill Lynch
I knew you talked about Fab 30 maybe being kind of incrementally brought down, or at least partially brought down and converted to 300mm starting in late 2007? Is that still on the table?
Dirk Meyer
Actually, we will start taking Fab 30's 200mm output down in about the second quarter of 2007.
Joe Osha - Merrill Lynch
I'm sorry? Second quarter of 2007?
Dirk Meyer
2007. Then, we will be getting 300mm output back out of that facility in very early 2008.
Joe Osha - Merrill Lynch
Will Fab 30 continue to produce for that? Will it completely go away, or --?
Dirk Meyer
The total utilization of the floor space does not really dip any lower than about 40% over the course of the transition.
Joe Osha - Merrill Lynch
But I could see Fab 30 drop to, say, 10,000 or 12,000 a month while you were in the midst of this, yes?
Bob Rivet
A quarter.
Joe Osha - Merrill Lynch
I'm sorry, a quarter.
Dirk Meyer
Absolutely.
UNQUOTE.
This definately changes Sharikou's projected output for AMD next year by 4o,ooo to 48,ooo wafers. Also, somewhere in the transcipts it mentions that Fab36 is only at half capacity and won't reach full capacity until Q2'07.
Competetively speaking, I believe Intel will be ramping 45nm in about that time and by the time AMD's Fab30 has been converted to 300nm and re-named Fab38 in Q2'08, Intel will have 3 factories pumping out 45nm chips. Forward looking, it appears that Intel will be easily able enjoy similar cost advantages through 2009 that AMD had needed to put itself in front of Intel.
Scientia.
I believe your wrong in one aspect. And that is the growth of AMD, I wanna see the cost for integrating ATi into the buiessnes before I can agree with your assessment.
/mork
Thnx for that info Sheepshagger. If you have any additional info please post the links, ive actually not heared anything about this until now.
Why would you compare Intel's highest gross margins compared to their latest and AMD's highest compared to their latest? Wouldn't quarter over quarter be a more fair assesment or at least both having the same quarters? If Intel goes back to 2005 numbers then they'll be filthy rich again regardless if AMD also gets richer.
Quarter over quarter doesn't work because of seasonal demand and seasonal constraints. These differ from year to year and quarter to quarter, so a quarter to quarter assesment cannot accurately take into account these adverse effects.
"Bob Rivet
A quarter.
Joe Osha - Merrill Lynch
I'm sorry, a quarter.
Dirk Meyer
Absolutely."
It sounds like dirk is either misunderstanding the context of the question and thus his answer or is failing to explain his context. "Absolutely" is a confident remark, meaning dirk has something up his sleeve or is saying that it will only drop by a quarter. However, you are right that Sharikou makes no mention of this in his analysis, and he probably wont budge even with this involved because he seems to be very stubborn.
Sheepshagger, you do fail to factor in that chartered has been ramping its own production up and will be capable of much larger production values soon. Especially as demands for xbox 360 processors level off as demand evens out and fewer units have to be replaced. Also, considering AMD is planning on producing 32 nm from new york, they'll have production coming from that fab at least by 2008 or 2009. Considering how much of the market AMD addressed with 1 fab, and is addressing now with basically 1 and a half, I'd say it's looking like it'll be doing okay by the time intel rolls out 45 nm. Also seeing how AMD generally stays 1 process level behind Intel and yet still stays competitive (market-wise) I would also say that this might actually put AMD ahead of Intel only has 3 fabs at 45, yet has 4 at 65 and is still only using that for 50% of its production right now. (as a side note, about 65% of Intel's 65 nm production is 9xx dual core processors).
What will really matter as to what puts AMD in front of Intel is all of the aspects of marketing, production, and performance. I suggest we all sit back and enjoy the low prices if we are really that concerned about what will happen.
Oops... scratch that last prediction.
New prediction, this one's just for you Sharikou 180°, AMD exits 2006 with +28% share.
This should be no problem with mobile their largest unit growth before Dell starts shipping notebooks in Q4. As long as they get them out the door alright, with higher ASP, increased mobile sales should raise AMD's margins next quarter. If not, it would indicate just how little Dell are paying for those chips.
Randy - you're implying that I should re-consider my prediction on market share reversal based on the Q3 results and you would be right...I am thinking about it. But have not made up my mind yet. I said it earlier - I'm not afraid to say I mis-called it. I'll let you know what I think in a couple of days since I have not yet listened to the AMD earnings call.
However, I'd suggest you think carefully too. To grow 5-6 points of market share is an incredible feat. Even if AMD grows, 5 points in a qtr is unlikely.
"Also, considering AMD is planning on producing 32 nm from new york, they'll have production coming from that fab at least by 2008 or 2009"
AMD is saying 45nm in mid 2008; I'll be impreseed if the have 32nm prodution that same year as typical node scaling is 2 years...
Also 2008 production would mean they're breaking ground sometime this year. I believe AMD has stated production starting in 2010. Maybe they can pull that in a bit but 2 years?!?!? on 32nm?!?!
Post a Comment