[Beyond3D Article] AMD Q1 2007 Warning Analysis

Just a bit of crazy talk, but i wonder if IBM would be interested in acquiring AMD. There might be some sharing of patents/ideas between GPUs and their supercomputer chips.

IBM's commitment to hardware is also up in the air.
Its profits come from services, not hardware. While it has a strong interest in making more of its chips, it doesn't have a strong one in expanding its hardware stake.

I'm also curious about some of IBM's statements concerning the fate of its process technology research. Last I read, it seemed its continued role in the cooperative alliance with AMD and others was set to be decided soon.

IBM has the option of walking away from the technology deal and its partners, and I'm not sure the deadline has passed.
AMD's current bad news may not make the option of staying look any better.
 
They're talking about on the order of 500 employees attrition in the nearish term, which is fairly minor in the large scale.

It seems very clear that they don't see a reason to panic at this point, or the actions we'd be seeing from them would be much more serious.

Talk is cheap. They were also going to launch R600 last quarter. They were also going to boost gross margins. what else- three months ago they did not even see any need for additional outside financing.

Part of the talk is I'm sure aimed to keep investors from panicking, because they do need financing. But they are having major problems - is there any relief in sight in the next few quarters? If this continues they will not be able to fund operations, so something drastic has to happen, barring a remarkable and fast financial turnaround.

AMD legally can't go fabless and continue to make x86 chips, per its licensing agreement with Intel.

There is no future for AMD's x86 if it goes fabless. If that happens, it'll just be ATI with a whole lot more debt and a bunch of employees it will soon fire.

I haven't listened to the call yet, but news accounts are suggesting a major strategic review and talk of an asset lite (fabless model).

Because you read some Internet document you assume they can't go fabless? They can do whatever they want, even if contractually bound. If Intel really thought they had a case they could sue them, but would they really do that. Thats even assuming htey can;t legally do, which is likely the case considering that there is more than just one document floating around.

As mentioned, if they sell there fab in Dresden, then they can reduce the debt and the employees still work there. There are many examples of foundries buying fabs, and since the Dresden one is probably pretty good, not to mention there might it come relatively cheap with a great customer, then I think is a pretty viable option.
 
Who can they fire, or what can they sell off?

Not talking about whole divisions or such, just thinning out some areas like say marketing, middle management, sales, you know all the useless load making engineers lives worse that every company has.
 
Because you read some Internet document you assume they can't go fabless? They can do whatever they want, even if contractually bound. If Intel really thought they had a case they could sue them, but would they really do that. Thats even assuming htey can;t legally do, which is likely the case considering that there is more than just one document floating around.
I'm just reading the text of AMD's arbitration agreement with Intel.
The max for outside fabs is 20%.

If there's some kind of legal out, it would likely invalidate the technology agreement that requires AMD and Intel share ISA extensions.
That's not a good thing either.

AMD's luck with pushing its own extensions is worse than its luck in buying graphics companies.

As mentioned, if they sell there fab in Dresden, then they can reduce the debt and the employees still work there. There are many examples of foundries buying fabs, and since the Dresden one is probably pretty good, not to mention there might it come relatively cheap with a great customer, then I think is a pretty viable option.

Maybe AMD should sell it to Intel, since the most direct beneficiary of a process customized for x86 chips is an x86 manufacturer.
The process is not very good for any analog or mixed-signal product. It's too hot and leaky for much of that.

AMD would be selling Dresden for the chance to pay someone more money to make its own chips.
Notice how horrid its margins are. Those margins will not improve if there is a middle man.
Notice how circuit performance suffers with foundry designs, they are nowhere near what x86 chips hit.
 
As for Dresden, selling that would be very stupid, since they get loads of funds from the local government there. That would rather make things worse IMO.

What they need is a performance-leader CPU, maybe also such GPU and some really great marketing. Not all is lost, but they need better judgement prior to giving a go to certain too ambitious projects.

For example, no manager in his sane mind would have okayed R520's development as it was, that was effectively half the transistors threwn away, only good for heating (huge trans. budget invested into then unneeded dynamic branching and the ring bus). Such stuff must end rather sooner than later in favour of more cost-efficiant designs.
 
Oh and Geo with that 500 employees comment: I suppose you have heard the term restucturing before:)

Every time I heard that word in any tech industry it clearly 100% sure ended with heavily reducing the staff, no other option. We've been going through it a few times in the last ten years alone here at DaimlerChrysler (which coincidentally could soon become MercedesBenz again; oh - what does this tell us?)
 
Talk is cheap.

That's exactly right. That they're mostly talking rather than doing says they don't see the need to be doing. They didn't pick these guys off the street corner this morning to run AMD.

Sure, they could be wrong. Absolute ZOMG disaster end of the company may be just around the corner. They wouldn't be the first exec team in the history of forever to catastrophically crash and burn the whole shooting match.

But they clearly don't perceive that's anywhere near or they wouldn't be still in the "talk is cheap; so let's talk" mode.
 
AMD's execs may be hoping on getting some kind of big financing deal, as well as stock sales.

If that's not enough, there's still a stake in Spansion they can sell off, though right now Spansion's not looking its best either.

If that's not enough then, they might consider using outside fabs for part of their production.

If they need more, they might start selling off parts of ATI.

If that's not enough, they might sell everything from ATI but the assets they'll use for Fusion.

If that fails, they'll likely reposition AMD as an also-ran chip provider, much as it was before Athlon.

If that fails, they can just give up on CPU manufacturing.
After that, they might consider selling a fab.
 
I am tempted to copy/paste "all the reasons why AMD can't go fabless and compete with Intel" from the other thread...
 
That's exactly right. That they're mostly talking rather than doing says they don't see the need to be doing. They didn't pick these guys off the street corner this morning to run AMD.

Sure, they could be wrong. Absolute ZOMG disaster end of the company may be just around the corner. They wouldn't be the first exec team in the history of forever to catastrophically crash and burn the whole shooting match.

But they clearly don't perceive that's anywhere near or they wouldn't be still in the "talk is cheap; so let's talk" mode.

The company isn't likely going away. But that doesn't mean a bankruptcy reorg is out of the question if something drastic doesn't happen. Because who wants to invest serious money in a company with that much debt that is generating those kind of losses with no relief in sight? Or at best only moderate relief. Intel surely isn't likely to loosen the screws and even if Barcelona is a performer it will take a while to ramp it. Meanwhile cash is being eaten.

Just because they are talking doesn't mean they aren't doing. But lets face it, if you are an AMD manager none of the options are exactly pleasant. Also, if you are an AMD manager with just a little bit of vision, you might have seen a glimmer of this coming considering Conroe performance, and the speed at which Intel is capable of ramping products. Not to mention prior Intel price war strategy. So you might have taken a serious look at making more meaningful cost cuts BEFORE the situation got so dire. And there was a good opportunity to do that with the ATI acquisition. Instead, lots of cost savings numbers were thrown about, but it was pretty ambiguous where these savings were going to come from. Seems like the big thing was more microprocessor sales, but given Conroe that seemed like B.S. to appease investors. The layoffs after the merger seemed really token at best compared to the overall size of the combined company.
 
If AMD's statements are to be believed, the conference call mentioned that a large portion of capex savings will be in the delay in transitioning Fab 30 to 300mm wafers.

Their plans seem to include long-term work on both fabs, and there is no mention of selling them.

There is the intent to sell 200 mm fab equipment, as well as the possibility of Spansion stock and land sales.

The asset-lite strategy seems to be variable, and apparently will be modified for each target segment.
They do not seem to be going asset light for Barcelona's market segment, and there is a possibility that the mobile product will have fab usage in AMD, since that is of higher margin than most desktop parts.

Early models of Fusion, if anything, would be the best bet for a foundry, since it will likely target a market that is not too picky on performance anyway.

What happens to desktop, I wonder? I don't think AMD will give up there, but it's not the focus of the company's CPU business anymore.
 
Maybe AMD should sell it to Intel, since the most direct beneficiary of a process customized for x86 chips is an x86 manufacturer.
Intel builds its fabs according a copy-and-paste strategy: they first develop and tune a process in the initial fab and then make an exact copy in the other fabs. Same machines, same settings, no changes allowed.

Buying an AMD fab would be a radical departure from that system or would require replacement of a lot of equipment.

The process is not very good for any analog or mixed-signal product. It's too hot and leaky for much of that.

You can use the same kitchen for different recipes. It's no different for fabs: changing processes is like changing recipes. It's common for foundries to run different processes in the same fab.
 
Bolding is mine...
For the desktop, Agena or Kuma is not likely to beat Penryn, and unlikely to beat Conroe. Barcelona's big changes deal with FP and SSE processing and quad-core optimizations, but the integer side is not radically different from K8.

Integer single-threaded performance is likely better than K8, but there is little indication and no definitive statements from AMD that it will be better than Conroe, and a number of hints and ass-covering vague statements to indicate that it will not be.

AMD's all about the platform and scalability, none of which will affect the desktop for quite some time.
Bingo! Some people will perhaps disagree with that, and AMD certainly might not like it either, but IMO there really is no way to repeat that enough. The hardware enthusiast and financial communities keep blabbering about how good Barcelona might (and most likely will be) in some market segments, but in the end, that doesn't even really matter.

Barcelona is not going to double AMD's server market share compared to the one they had in H2 2006. In an absolute best-case scenario, they'll increase it slightly and keep good margins and ASPs there. That's good, but as I said, it doesn't really matter because servers currently aren't more than 40% of AMD's CPU gross profit. So, even if that increases by even as much as 50%, but desktop/laptop revenue halves, they'd still only have 90% of today's gross profit. And given the operating loss they've got right now, that'd be rather catastrophic to say the least.

In the end, what AMD needs is a very strong line-up in all market segments (at least within 5% of Intel at decent margins, and ideally better than that), and I see no evidence whatsoever that Barcelona can give them that. Perhaps I am wrong. Heck, I hope I'm wrong for AMD's sake and for the industry's sake. But annoyingly enough, I very much doubt I'm wrong at this point in time.

Here's a simple way to think about it: what does a dual-core Barcelona have that Penryn, or even Conroe, doesn't have architecture-wise? Besides for minuscule things that are really implementation-dependent like TLBs and most likely balanced properly in Conroe (and, if they weren't, you'd expect them to be in Penryn!) there really isn't much that's better than what Conroe proposes.

The IMC is good, but sadly I think it's pretty much the only advantage AMD will have against Conroe for desktops/laptops. On the plus side of things, and unlike with K8, they won't have as many disadvantages. Also, FP performance might be good, but it matters less than in servers, and it shouldn't be overestimated.

In the end, if Intel decides to release an unlocked high-end dual-core Penryn, it could be clocked at 3.43GHz+, and there's no real way for AMD to keep the desktop performance crown then. I'm also not very bullish about their 2008 laptop platform (Griffin) because it'll be 65nm-based and that Penryn looks to be a formidable competitor in that area.

Either way, I hope that AMD's management realizes that their problems are deeper than they might first appear. It's one thing to say that you understand your problems, but it's another to actually understand *all* of them rather than just the most obvious and/or problematic ones in the short-term.

Also, I sure as hell hope that their 2009 roadmap is more than a souped up Barcelona derivative with an on-chip GPU, because that'd a recipe for disaster against Nehalem. I believe it will also be interesting to see if Intel releases a Penryn derivative with a smaller die size. Looking at the die shot, it would be rather straightforward to halve the L2 cache and create a new ~80mm2 chip for the value segment to replace Conroe-L.

Anyway I guess it's fairly obvious that I'm not too optimistic (read: I'm extremely pessimistic!) about AMD's 2007/2008 CPU roadmap. They might do OK in the server market, but that's only one small part of the picture, and it is a massively insufficient revenue source for the company as a whole. I think it's about time everyone, and especially the financial community, started to realize this.

P.S.: And obviously this opinion is too opiniated and perhaps subjective for me to ever write something about it in an article or a news post, which is why I decided to just do a rather long forum post instead... :)
 
I think that AMD managment knows how deep in trouble they are, but who would take them seriously if they kept saying - we don't know if we will be around next year :)

Sometimes, even if you see all the problems, you just cant respond to them as competition is in much better shape, you have some weights from past and so on....

Zvekan
 
I think that AMD managment knows how deep in trouble they are, but who would take them seriously if they kept saying - we don't know if we will be around next year :)

Let's stipulate for the record that what they might say in public and what they recognize internally could be different. But you're still missing the point. Even if they were just putting on the happy talk for public consumption, there is no evidence of crisis-like *actions* being taken. No announcements of 20% layoffs, selling off the CE business, nothing like that.
 
Well, it could either mean the company is not really in trouble, or that the management doesn't have the foresight to act decisively to save it.

BTW, I remember when ATI deal was announced, some, myself included, raised the question of a potential capital crunch AMD was putting itself in. We were told not to worry because "Hector clearly knows something you don't; Clearly Hector knows how much cash is coming via Dell; Hector knows what he is doing". Pardon my French, but either Hector doesn't know sh*t or he is one of the best actors alive.
 
I think they could still very well pull it off, but they can't afford any more mistakes like what happened in the last 2-3 years (ATI specifically, AMD-CPU less so)
 
I think they could still very well pull it off, but they can't afford any more mistakes like what happened in the last 2-3 years (ATI specifically, AMD-CPU less so)
Referring to the other thread about 'real men have their own fab': this is probably a textbook case about the dangers of owning a fab. They may be better for margins when times are great, but they are a disaster when times are not so good: it's hard to reduce carrying costs of a fab without selling it, whereas you can just reduce orders in a fabless model. (As 3D pointed out, they don't apparently have a choice because of contractual reasons...)

So where can they find $600M ? After spending way too much time on their (awful) website, I finally found out they have 16000 employees. They obviously can't do drastic cuts in R&D without jeopardizing the future, but if Arun's insight is true and upcoming CPUs won't allow them to increase prices, then I really wonder what they can do?

500 employees @$150K/year is just $75M...

That said, they lost $1225M in 2002, so they have been there before...
 
(As 3D pointed out, they don't apparently have a choice because of contractual reasons...)

Clever forward-thinking move on Intel's part when drawing up that settlement, do you think? Restricts AMDs room for manoeuvre and pretty much forces them in to a financial capital expenditure slugging match as fabs get exponentially more expensive. A slugging match which Intel was always in the better position to win in the long-term.
 
but if Arun's insight is true and upcoming CPUs won't allow them to increase prices, then I really wonder what they can do?
Well I doubt it can be much worse than the current line-up relative to Conroe, sorry if I gave that impression. However, only being slightly more competitive won't enough to stop the bleeding, at least in terms of CPU operating loss.

As for how to cut costs... Well, it's fairly hard to judge how significant of a restructuring they are thinking of. However, consider this:
http://www.iht.com/articles/2001/09/26/techbrief_ed3__154.php
And then this, from: http://seekingalpha.com/article/32901
Hector Ruiz said:
While I expect this task force to be temporary in nature, lasting no more than a year, I expect this transformation to be bigger and more dramatic in impact than the one we undertook in 2002. As I mentioned earlier, it will be the transformation needed that allows us to fully complete the transformation of our industry, to put the customer back in charge, to put competition back in our industry, and to put sustainable value creation back in the hands of real innovators.
Bolding is mine. Depending on what you want to read into that, you might presume he might hinting at a restructuring that will fire even more than 15% of their employees. Ouch.
Or, he could just be thinking in terms of the impact it will have on the industry (you know, like Barcelona... *cough*)
 
Back
Top