Richland & Kabini rumours

[mrmt]

Of course it's an excuse! What would you do if your rival - a monopoly - decided to dump half a billion worth of inventory? It's not just an excuse it's the actual fact for why AMD couldn't meet their target. It's why they have scaled down because intel will be dumping even MORE inventory for the forseeable future.

Intel barely touched the sub-60% margin territory, probably because of the charge they took with the pre-production Haswell wafer that will be reversed in the next quarter, If you see Intel close to 50% gross margin, then you'll have your worst case scenario for AMD.

1) What would you have done with Kabini, because I'm 100% certain it would have been what AMD did.

Instead of quick rehash like Brazos 2.0, I would have pressed ahead to Kabini ready for primetime at 28nm TSMC whenever they had the available volumes, Q112, Q312, Q412, you name it, but it would be ASAP. Sooner or later volumes would be ok, wafer prices ok too, it was a very straight decision. I would also canned Bulldozer and gone for a Phenom die shrink.

Those two things wouldn't have solved my fundamental problems, but would have bought me some time (and cash).

2) What should they do to mitigate their monopoly competitor from dumping cpu's and eating into their market share in a market that is rapidly capitulating? I'm 100% certain the answer to that is - "they should scale back their production, even if the have to pay to do it because that is better than being left with a mountain of unsold inventory".

IIRC I didn't say they had any choices regarding the WSA, once they signed the spin off deal the soul is now the devil's property. What I said is that the WSA was a very bad deal, that GLF is acting like a loan-shark (IDC, I'll pay royalties in beer some other time) and GLF is bleeding AMD dry and will kill it as soon as the foundry is on its feet.

I did bash for them missing every sales forecast since Rory Read arrived.

 

[SiliconWars]

Based on what we now know, was scaling back GF production the correct or wrong decision? It really comes down to that simple question.

 

[mrmt]

What can they do vs that? You tell me because it seems clear that they are doing the only sensible thing by scaling right back.

AMD isn't scaling back because of a new business strategy, AMD is scaling back because AMD will have to scale back or burn cash at a far greater rate and will be bankrupt rather sooner than later. It is not a decision that management is consciously taking, but it is being rather forced on them.

What they could do that would inspire investors confidence? First accurate financial projections, no matter how bad they were. At least that would show that they are on top of the situation. Second a credible roadmap with reasonable performance expectations, but this is too much for AMD, even a Senior Marketing Director went to tech forums to hype products. Third, what about not promoting an interim CFO that tried to hide information from investors in a Q&A?

 

[mrmt]

Based on what we now know, was scaling back GF production the correct or wrong decision? It really comes down to that simple question.

Yes, that was correct. The previous WSA was unbearable for AMD current structure.

 

[pablo87]

old vs new mgmt is an excuse, it's the same BoD and they approve the deal with Albatross Foundries and nothing has changed in their dealings either - new amendment to WSA every 6 months aka more kicking the can down the road.

For example, the most recent amendment was necessary because AMD could not meet the $1.5B take or pay for 2013. so at the q3 earnings, the situation - that AMD is not going to meet it having excess inventory and still $500MM to go - is mentioned very casually by saying "negotiations are going well". Then later in the 4th quarter, the other shoe drops (what was it $300MM in total incl the promissory note? I've stopped counting)

Moreover, by q2 earnings or midway in q3, did they know they would be nowhere near the $1.5B take or pay and would have to take a charge? Of course they did! The inventory was already up more than 50% in 6 months - one way or another, it was going to be a house of financial pain. btw, sp on July 22? Close to $5.

 

[Idontcare]

What options does he really have? He can continue on with a disasterously large cpu that would simply not sell in the current climate, or he can scale back production. AMD are at the mercy of what Intel does every quarter, always have been, and they can only react to what they think Intel is going to do. Half a billion of inventory is more than 1/3rd of AMD's total revenues. What can they do vs that? You tell me because it seems clear that they are doing the only sensible thing by scaling right back.

If all Rory can come up with is the same thing that you and I can surmise he ought to be doing while we are sitting in our homes talking things out on a forum then he is either ridiculously over-compensated or you and I are ridiculously under-valued.

Humility binds us to assume Rory ought to be able to do something beyond that which we can imagine ourselves doing in his shoes, I am not about to assume I am better than Rory when it come to business management (or much else for that matter)...and yet here we are basically using our uncompensated collective intelligence to make excuses for his actions/inactions.

Doesn't that strike you as a little imbalanced?

I completely agree AMD's goose is cooked and for reasons that were baked into the cake before the current management took the helm. But if AMD is bound to such an extent that Rory has little choice then AMD doesn't really need Rory, they can't take advantage of his skillset (but are paying for it) because Rory's hands are tied behind his back.

If the CEO position at AMD is such a toothless tiger then surely they could have found someone for the job who wasn't going to expect millions in compensation just to sit on his/her hands. :|

 

[Ajay]

Chapter 11 would most likely involve wiping out the entire equity stake that shareholders own (meaning stock value would go to zero).

The executive decision makers would stand to lose millions upon millions of dollars in stock options if AMD were to take such a path. They are highly incented to do anything and everything they can to avoid making those stock option worthless.

Even if the stock is trading at only $1, that is a heck of a lot better than zero.

As much as "AMD" the business entity is in the business of staying in business, the decision makers that are driving AMD are not making those decisions for the betterment of AMD, they are making those decisions to better themselves.

Sometimes this works to the benefit of AMD the business entity, sometimes it becomes a conflict of interest between what is best for the decision maker versus what is best for the business (see HP and any one of their past 3 or 4 CEOs, or Apple and the difference between Sculley's tenure and that of Jobs').

Well, if their options are under-water, then there is no incentive to avoiding Chapter 11, except that they'll likely lose their jobs (unless they are delusional and think some 'white knight' is going to step in and save them). I am becoming more convinced as time goes on that Chapter 11 is the only path forward that would preserve AMD as a business. What that business will look like, is beyond me.

 

[Idontcare]

Well, if their options are under-water, then there is no incentive to avoiding Chapter 11, except that they'll likely lose their jobs (unless they are delusional and think some 'white knight' is going to step in and save them). I am becoming more convinced as time goes on that Chapter 11 is the only path forward that would preserve AMD as a business. What that business will look like, is beyond me.

Stock options aren't issued much nowadays, companies prefer to offer restricted stock instead. (this change happened over the past 5yrs)

Motivation Through Restricted Stock

Issuing restricted stock is a better motivating tool than granting stock options for two reasons. First, many employees don't understand stock options. They don't know that they have to take action to realize any gain. It is far easier for them to understand a vesting period on restricted stock. Second, restricted stock can't become worthless like stock options. Even if the stock price falls, restricted stock retains some intrinsic value.

A stock option grant with a strike price of $10 has no value when the stock trades at $8. Restricted stock awarded when trading at $10 is still worth $8. A stock option has lost 100% of its value. The restricted stock has only lost 20%.

 

[SiliconWars]

If all Rory can come up with is the same thing that you and I can surmise he ought to be doing while we are sitting in our homes talking things out on a forum then he is either ridiculously over-compensated or you and I are ridiculously under-valued.

Humility binds us to assume Rory ought to be able to do something beyond that which we can imagine ourselves doing in his shoes, I am not about to assume I am better than Rory when it come to business management (or much else for that matter)...and yet here we are basically using our uncompensated collective intelligence to make excuses for his actions/inactions.

Doesn't that strike you as a little imbalanced?

I completely agree AMD's goose is cooked and for reasons that were baked into the cake before the current management took the helm. But if AMD is bound to such an extent that Rory has little choice then AMD doesn't really need Rory, they can't take advantage of his skillset (but are paying for it) because Rory's hands are tied behind his back.

If the CEO position at AMD is such a toothless tiger then surely they could have found someone for the job who wasn't going to expect millions in compensation just to sit on his/her hands. :|

Well is it really inaction? It seems to me like Rory is at least trying to do something different. When I think back to Dirk, I'm almost certain his reaction would be intel-like, ie he'd decide to scale production up instead and take on intel head-on. While the AMD fanboys might like to think that was a great idea, it would almost certainly be suicide.

Scaling down is the way ahead for the company just now. That means in everything they do. They need to base themselves around smaller chips like Kabini, which they are clearly heading towards. They need less people, less real-estate (which won't be a problem because they are selling it all off ), less big dies...less focus on beating intel at the high end.

The high end is gone, intel wrapped up servers years ago. Bulldozer should never have existed. AMD's future is all about being small and nimble like Nvidia or the ARM crowd. I believe that is the direction that the current management is taking the company, and I've believed it is the direction they should have been taking the company ever since Brazos was released.

 

[piesquared]

Speaking of humility, I guess there might be a few people about to get a healthy dose. Chapter 11 isn't even a concern to those who know, it seems. People can't underestimate the importance of and pent up demand for next gen consoles. Since the console's primary purpose is to play games, it follows that the provider of that hardware stands to benefit. And given the booming gaming market, being #1 in graphics that isn't a bad place to be in. Add to that the huge success of Gaming Evolved and it's starts to look like game developers are working closely with AMD. Intel is a newcomer to the gaming market and have no record to stand on other than dismal performance and drivers. IOW, AMD will be the platform of choice for developers including APU's and HSA. AMD have a lot of IP to leverage as a design house for an open platform that suits customers.
And that's just consoles, there's supposedly a slew of new products coming with Richland, Kabini and Temash, and Sea Island after that VI and PI.
I'm not sure how even the most skeptic among us could envision a chapter 11 from that. AMD also released a new Open Compute 3 board and has Sea Micro under it's belt.

But I'm willing to eat crow if I'm wrong, anyone else?

 

[Idontcare]

Speaking of humility, I guess there might be a few people about to get a healthy dose. Chapter 11 isn't even a concern to those who know, it seems. People can't underestimate the importance of and pent up demand for next gen consoles. Since the console's primary purpose is to play games, it follows that the provider of that hardware stands to benefit. And given the booming gaming market, being #1 in graphics that isn't a bad place to be in. Add to that the huge success of Gaming Evolved and it's starts to look like game developers are working closely with AMD. Intel is a newcomer to the gaming market and have no record to stand on other than dismal performance and drivers. IOW, AMD will be the platform of choice for developers including APU's and HSA. AMD have a lot of IP to leverage as a design house for an open platform that suits customers.
And that's just consoles, there's supposedly a slew of new products coming with Richland, Kabini and Temash, and Sea Island after that VI and PI.
I'm not sure how even the most skeptic among us could envision a chapter 11 from that. AMD also released a new Open Compute 3 board and has Sea Micro under it's belt.

But I'm willing to eat crow if I'm wrong, anyone else?

I'll second that motion that you will eat crow

Seriously though, AMD doesn't need chapter11 for economic reasons, not anytime within the next 2yrs at least.

I think all the discussion surrounding chapter 11 is based on the notion of taking advantage of the re-organization aspects that chapter11 legally affords and in doing so AMD could wrestle itself free from the existing protracted 2024 WSA + Take-or-Pay contract with GloFo.

Emerging from such a bankruptcy re-org, free to actually choose their foundry as a fabless entity, would be priceless at this point in time. GloFo is bleeding off what precious little resources AMD has...sucking away their future by starving their R&D pipeline today.

Where would rather see AMD's $360m Q4 write-off go? AMD R&D or GloFo as a cash transfer? At least if it disappeared into the vacuum of R&D there might be something to show for it in 3-4 yrs time.

 

[jpiniero]

The Problem is that if AMD does go bankrupt Intel could argue that would violate the treaty. Maybe AMD can get to a point where they don't need x86, but they don't have that long.

 

[piesquared]

I'll second that motion that you will eat crow

Seriously though, AMD doesn't need chapter11 for economic reasons, not anytime within the next 2yrs at least.

I think all the discussion surrounding chapter 11 is based on the notion of taking advantage of the re-organization aspects that chapter11 legally affords and in doing so AMD could wrestle itself free from the existing protracted 2024 WSA + Take-or-Pay contract with GloFo.

Regarding AMD's restructuring of the business and the WSA, i'm of the opinion that AMD is responding mainly to market conditions and adjusting the portfolio for market trends. Trying to steer consumers seems to be a lot like herding cats. They have spoken and are demanding low power mobile devices. AMD has reacted to that reality and is condensing around their major projects it seems. The reduced wafer purchases could be a result of the small core strategy.

Emerging from such a bankruptcy re-org, free to actually choose their foundry as a fabless entity, would be priceless at this point in time. GloFo is bleeding off what precious little resources AMD has...sucking away their future by starving their R&D pipeline today.

Perhaps AMD honors their commitments and would rather not burn bridges.

At least if it disappeared into the vacuum of R&D there might be something to show for it in 3-4 yrs time.

Maybe R&D started 3-4 years ago is upon us now. With strong current and upcoming products, reduced debt and 1.3B cash on hand there should be plenty of revenue for AMD and shareholders alike. TMO
products products products If AMD becomes the gaming platform of choice, what's left for intel in the enthusiast market? If games are optimized for the AMD platform on both PC and console enthusiasts aren't going to buy bloated cores just to run some meaningless benchmarks. Neither are SMB, tablet makers and other innovative IHV's.

 

[Vesku]

I'd love to hear Read's reasoning for not actually using the capacity they are paying for from GloFo. Maybe this is my naivety speaking but I'd try to find something to use up the 32nm wafers when GloFo's 28nm fell flat. What happens in the agreement if GloFo can't meet their side of the wafer equation?

 

[Idontcare]

The Problem is that if AMD does go bankrupt Intel could argue that would violate the treaty. Maybe AMD can get to a point where they don't need x86, but they don't have that long.

"The treaty" LOL, so ironic because it is true

I'm sure it would up-end the existing cross-licensing agreement but that is something Intel would presumably be rather keen to renegotiate for comparable terms. They depend as much on IP access to x86-64 as AMD depends on IP access to x86.

 

[SiliconWars]

I'd love to hear Read's reasoning for not actually using the capacity they are paying for from GloFo. Maybe this is my naivety speaking but I'd try to find something to use up the 32nm wafers when GloFo's 28nm fell flat.

They don't have anything that is small enough, and they can't do anything when Intel dumps chips like they have been doing.

If AMD uses that GF capacity they will be forced to reduce prices in order to compete with intel dumping their inventory. Margins aren't wonderful already on these parts, but the situation is a lot more complex than that. OEM's, mobo makers all have to be considered in the equation. Taking the charge is the sensible way ahead. It's easier to improve your position when sales are good even if expenses are bad, and I believe AMD is doing ok with what they have now.

What happens in the agreement if GloFo can't meet their side of the wafer equation?

That was last years wafer agreement, which benefitted AMD a lot. It's easy to forget about that but when GF underperformed, AMD got the benefit of a pay per good die agreement.

A lot of the problem is that GF have a lot of money and AMD doesn't, so we think that GF should let AMD off with this. Sadly that's not how business works, although I do believe that it's not in GF's interest to drag AMD closer to bankruptcy, and I'm sure they know it too. Relations must be strained to a point.

 

[erunion]

they can't do anything when Intel dumps chips like they have been doing.

Why do you keep saying that? That's so completely untrue. Intel's ASP actually grew in Q4, and their revenue was only down 3%.

Intel idled plants to reduce inventory, plus 22nm is running at only 50% of capacity.
Its expenses that killed them.

 

[SiliconWars]

Why do you keep saying that? That's so completely untrue. Intel's ASP actually grew in Q4, and their revenue was only down 3%.

Intel idled plants to reduce inventory, plus 22nm is running at only 50% of capacity.
Its expenses that killed them.

Well that's even worse. Take your pick of what is more likely, it's likely a combination of the two.

I'm sure it was mentioned in the CC or afterwards that older inventory was sold off?

http://seekingalpha.com/article/111...arnings-call-transcript?page=4&p=qanda&l=last

Paul Otellini - President, Chief Executive Officer, DirectorYes, I will take that. We think that there was an inventory drain and a worldwide supply chain for PCs in the fourth quarter. Our channel checks would suggest that a lot of older generation Windows 7 PCs were burned off in the quarter. When we look overall at inventory levels across all the downstream inventory, we think it’s a healthy level of inventory. It looks appropriate for how we see demand and then in terms of our own inventory levels, obviously we reduced them significantly in the fourth quarter.

 

[Idontcare]

Welcome to the forums erunion :thumbsup:

I'd love to hear Read's reasoning for not actually using the capacity they are paying for from GloFo. Maybe this is my naivety speaking but I'd try to find something to use up the 32nm wafers when GloFo's 28nm fell flat. What happens in the agreement if GloFo can't meet their side of the wafer equation?

My wife works at a company that just took a $2m hit from a take-or-pay contract in which her company decided to just pay and not receive the goods.

It is odd and it basically comes down to these companies are throwing away other people's money, and it takes real effort on their part to avoid having the money go to waste. $2m of lost shareholder wealth is a shame, but no one is about to make their weekends hell while scrambling around trying to figure out what to do with $2m worth of product xyz that the company doesn't want.

It really is just laziness and lack of accountability. My wife's company could take delivery of $2m worth of product xyz and turn around and sell it at pennies on the dollar and at least stand to recover say $0.5m or $1m of that $2m outlay...but no one wants to step up and volunteer to be the one who is going to spend their evenings doing that.

So management just decided it was easier to take the charge, easier for them personally that is, and tell shareholders "oops, sorry about that, we'll do better next time".

Could AMD have done something with $300m worth of 300mm wafers? Absolutely no question, at the very least they could have sub-contracted out to universities and research consortia and either gave away the wafers (tax deductible) or charged pennies on the dollar for them.

But that would have mean someone, somewhere, inside AMD would have had to step up and champion a proposal that would have entailed expanding their own workload so the project got off the ground...and in the end it was just a lot easier for Rory to stand up in front of a bunch of analysts and say "oops, sorry about that, we'll do better next time".

OPM - Other People's Money

OPM makes it difficult to give a damn when OPM is disappearing and you've got some killer weekend plans with the family coming up :|

 

[erunion]

Thanks, IDC.

I've been lurking for a bit. But there's only so many times that I can stand to watch the same incorrect statement repeated.

 

[SiliconWars]

Thanks, IDC.

I've been lurking for a bit. But there's only so many times that I can stand to watch the same incorrect statement repeated.

And now you're going to go back to lurking after I've proven your post to be inaccurate?

 

[erunion]

Well that's even worse. Take your pick of what is more likely, it's likely a combination of the two.

I'm sure it was mentioned in the CC or afterwards that older inventory was sold off?

http://seekingalpha.com/article/111...arnings-call-transcript?page=4&p=qanda&l=last

If you are referring to OEMs dumping PCs, that has nothing to do with Intel's top or bottom lines. You could make that claim even if Intel had record profits. (eg OEMs are dumping old PCs to make room for all the new chips they just bought)

Intel idled plants, that's the opposite of dumping. And the equivalent of what AMD did when choosing to Pay rather than Take.


Also from the conference call

The drop from the third quarter was driven primarily by excess capacity charges as a result of the aggressive tactical actions we took to reduce inventory levels and to redirect space and equipment to 14 nanometer.

 

[SiliconWars]

If you are referring to OEMs dumping PCs, that has nothing to do with Intel's top or bottom lines. You could make that claim even if Intel had record profits. (eg OEMs are dumping old PCs to make room for all the new chips they just bought)

Intel idled plants, that's the opposite of dumping. And the equivalent of what AMD did when choosing to Pay rather than Take.


Also from the conference call

Also...

On PCs, what we saw was the strength in the core product line principally going into Ultrabooks and laptops and little more weakness than we would have first thought in the bottom of the PC market in our Celeron and Pentium product lines.

You wonder why ASP is up? Guess how the inventory mix changed from SB ULV's and quad cores to Celeron and Pentium's and you have your answer.

Come on, who do you think is footing the bill for these Windows 7 PC selloffs?

 

[erunion]

Also..

You wonder why ASP is up? Guess how the inventory mix changed from ULV's and quad cores to Celeron and Pentium's and you have your answer.

I don't follow.

A higher ASP means that OEMs are buying more Core processors than Celeron. It reasons then that they are selling more PCs equipped with Core, than ones with Celerons.

 

[SiliconWars]

I don't follow.

A higher ASP means that OEMs are buying more Core processors than Celeron. It reasons then that they are selling more PCs equipped with Core, than ones with Celerons.

Yes that's exactly it. If you offer a PC builder a Core cpu at a price that is much closer to Pentium prices, what do you think they'll do? Yes they'll buy the Core and ignore the Pentiums and Celerons.

Drop prices on the Cores...and suddenly Pentiums and Celerons show weakness in sales? Is this a surprise to you? I really hope not. So now intel's inventory is made up of much lesser cpu's than it was, but ASP last quarter was still up because of the "deal" they got on Cores.

AND THE REALLY GOOD THING ABOUT THIS IS - AMD Trinity sales suffer badly because they cannot compete vs Core when Intel lowers prices in this kind of selloff.

Easy.