CPU AMD apparently being sniffed by Private Equity firm

http://www.fudzilla.com/news/processors/38694-private-equity-to-get-20-percent-of-amd

I'm still hoping for the split and spin-off though for reasons.

Ignore the 'it wouldn't cost very much' referencing current stock price. I have read alternative opinions elsewhere about current large investors/owners will want their $ back. If you think in terms of its IP alone, AMD is undervalued.

 

If I had the money I'd be very tempted to buy AMD stock myself, like you I think AMD is undervalued and I'd be willing to take a gamble that what's in the pipeline is going to see a change in fortunes for AMD over the long term.

ZEN Core, all signs point to it being good or better than good.

Arctic Island GPU's with HBM2 should be much better what with their previous experience of HBM, DX12/Vulkan, smaller node size (16/14nm FinFET).

 

For HPC it should be good, not sure yet about desktop. Although even if AMD bring 8+ cores to DT they'll win regardless due to Intel sandbagging.

AMD would yield more value as a split company though: semi-custom + CPU(HPC/server) and GPU + licensing would allow them to sell more to more people, with less conflict of interest.
> Exit low margin PC business (or make token effort) and focus on server/HPC, offering semi-custom designs to big buyers. Not what I want as a PC user but on the business side - channel, manuf relationships, Intel MDF etc this makes sense.
> Keep in GPU business but focus on highend + HPC accelerators. License GPU tech to CPU business as required
> License GPU tech to ARM SoC vendors as some are desperate for differentiation

And the news today:"The company will be reforming a singular, monolithic graphics group – the Radeon Technologies Group – to oversee every aspect of AMD’s graphics efforts. The Radeon group will in turn be led by long-time ATI and AMD graphics guru, Raja Koduri."

This is a stage in that process.

 

I wonder if AMD will try and delist themselves. It would remove a lot of pressure if they did. The big question is why they are doing this. Are they lacking the money to survive till Zen launch if they don't. Do they need the cash for a down payment on there next set of chips?

There's also the question of how they will do this. AMD shares are not owned by one or two people buying them back would be extremely difficult. They could simply issue 20% more and dilute everyone else but that requires shareholder permission and unless they are going bankrupt without the cash then not sure why they would agree.

They could buy the shares on the open market but that does not help AMD as they would gain nothing. Not to mention if they did go to 20% they would have to state there ideas as they would trigger certain buy out clauses.

 

I've read some large% shareholders will want their $ back. I don't know how authentic the financial analyst was (some are full of ****), but they guessed it would require $10-up/share to buy them out. That at least means they won't delist voluntarily. There's no point in issuing more because they are already unpopular as the immediate future is not good until possibly ~mid 2016 and it would dilute the current stock's worth even further.
GloFo ****ed them out of 20nm causing them $14M write-down AND product delays. 20/22nm APUs should have happened already and would have clawed back at least some marketshare: more cores or lower power - and to top it off, GloFo are only NOW making an FD-SOI 22nm node that should be super low leakage, but it's too late for AMD this year. It costs multi-millions to do tapeouts and testing. AMD's Bulldozer arch is bad, sure, but GloFo's persistent lack of execution just loads on the shitcake.

According to another financial report - the one that stated bankruptcy by 2020, they are, however, in serious financial stress. I can only trust they did their homework on the public accounts in that front. Apparently they've literally no more assets to sell, and hence why they've canned a lot of projects like x86/ARM and it's mostly rebrands for this gen of GPUs because again tapeouts/design/validation of each chip is $$$. What's left has gone behind Zen success.

If Zen isn't fantastic to claw back marketshare it will be nothing short of catastrophic for the entire PC industry. Remember Intel needs AMD to exist too, so a break-up and sell-off would cause extreme ripples and legal problems. The x86 license isn't transferrable - yet - but compared to the othersight of anti-monopoly divisions in US/EU(/China?), there's several suiters sniffing - none of which are interested in the PC market. Almost everyone is focusing on both ends: either HPC and semi-custom where there are real margins, or mobile where there's differentiation from ARM/Imgation/MALI and bigger volume.

The problem is compounded by the fact that, compared to Intel, AMD CPU team doesn't give nearly enough technical support or validation to its partners because they simply don't have the manpower and budget. (Example the Bulldozer platform was still unstable AT STOCK SPEEDS at launch day! That's totally crazy). That - combined with limited (read:nothing to make anyone get out of bed) marketing support and a sales channel that looks at recent history of AMD sales and goes 'hmmm not good, I need to hit my targets in a declining market so I'll focus on what I know sells' means motherboard manufacturers are reluctant to spend much on R&D as ROI is lower and risk is much greater. As much as a level playing field is way more preferred from a negotiating standpoint, given the total income potential of Intel hardware across all platforms to TWese business', the relationship to Intel will always be more important than AMD's - more so than ever now it's graphics market is also sub 20%. Sure, they have a loyal fan base, but the numbers are diminishing month on month and business' look at figures not online commentary. Those business commitments are long decided by the time reviews hit.

Even if Zen is fantastic, AMD has a monumental mountain to climb.

 

I have wondered even if Zen came out faster than every Intel Chip if it would be enough to save the company. By the time it launches AMD will be in a bad way cash wise and even if its a huge success and sells out. That might not be enough to save them from the Bankrupt courts.

The Enthusaist Fan sector is not enough to get AMD out of the Red long term or they would not be in this mess. They need Apple / Dell / Lenova to want to put there cpus in products and pay them the $billions they pay Intel each quater.

 

AMD's strengths are its server/HPC and semi-custom channel. Even IBM's OpenPower is seeing returns here this year, despite Power arch's decline. HPC is real growth and if AMD can work with Xilinx on FPGA's (which is why Intel bought Altera) it'll be a HUGE win. It's extremely difficult to create server channel, which is why Qualcomm/Sam might be thinking server but difficult to build those relationships, and despite the fact AMD server% is dwindling they still have those relationships. The margins are much better, and even though the validation work is more intensive there's overall less to do because spec will be tighter than the super-broad consumer PC. There's also no need for marketing budget, just fulfillment and execution.

In that respect I expect Zen server to launch before PC parts. Get some income ASAP.

 

If rumors are anything to go by we may not see ZEN until the end of 2016.

 

So another 15 months of hanging onto 28nm? Jeebus..

 

Someone like Samsung could be a good match for AMD perhaps, at least for its chips.

The problem any buyer will have is x86 licensing, the license in place with Intel has a clause IIRC that means Intel get to veto any new buyer from using it.

 

By the sound of things AMD would do well to work at positioning themselves in those other markets and just withdraw from public processor market. Even if Zen is fairly successful then what is the probability AMD still announce withdrawal around 2017?
The problem is even if someone does take up the x86 licence why would they even decide to compete with Intel? Surely pushing ARM at this point and forcing it into laptops and desktops would be more successful than fighting Intel. Think Apple and Samsung here, even Nvidia getting into the desktop space. While Enterprise would keep x86 alive, it wouldnt take long for things to start shifting as ARM processors continue to improve.

 

That would require getting Microsoft onboard really, as for it to have any chance it'd need Windows support realistically. Apple might well be looking to move to ARM for MacOS at some point in the future as iOS and MacOS are liable to merge. Doubtful Samsung would want to enter the desktop market, laptop market maybe.

 

Good luck getting Intel to agree to Samsung or Apple getting x86 licence. It's not going to happen both have the buying power to push Intel hard.

There's a reason most rumours about Samsung or Apple buying AMD only mention gpu patents. Most assume they would kill off the rest of the company.

Nobody wants to enter a desktop market where profit margins are tiny and there's little to no way to gain higher profitability. Laptop market is all high margin stuff but why would Apple jepodize long term Intel contracts with a buyout of AMD. Apple gets priority on all new mobile chips Intel launches most industrial insiders say it's a 3 month lead.

Intel even boosted the gpu to help out Apple.

3rd party like what is mensioned is the way to go. External investment kill off the weak departments and focus on what they are good at.

 

Apparently not only are Silver Lake showing interest in AMD but rumors are whispering that Microsoft are also interested in acquiring AMD, sauce.

 

Interesting.

As the PS4 and Xbox One both use AMD chips/GPUs, that could become an interesting development if MS own AMD, as they could leverage that to drop the Xbox One price with Sony being unable to match them without huge loss, depending also on the contract terms for supplying to Sony, MS could also pull a double whammy and refuse to provide more chips to Sony, though doubtful they would.

Any company that does buy AMD will more than likely do what a lot of such takeovers end up being, that being to asset strip for patents as thats where the true value lies.

 

MS mightn't have a say if chrome books continue to rise. I still can't see linux or chrome dethroning Windows but there is potential for something to come along and slowly eat into MS's domination. SteamOS has the potential to do damage, androidTV, even Apple could in theory take an increasing share. Part of me thinks that is why MS want's windows on everything, ability to adopt to hardware changes.
MS buying AMD would be a massive to protect their business as a whole. From consoles, desktop, enterprise, surface line and possibly eventually mobile. Having Apple beat surface to the latest Intel processors while Intel remaining the only viable option is a bad position to be in. Samsung snapping up AMD could be just as bad if they strip the company and kill part of it off. Like I said before tho I can't see anyone wanting to compete against Intel in any shape or form. If MS was to buy it would be for specific products ie. xbox, surface or servers and not for public use

 

All very valid points, although outside of the iPad/iPhone Apple sales volumes are tiny compared to Intel or AMD cpus powering Windows. I remain unconvinced about SteamOS at present, as still too few games are available for Linux, which will be the big hurdle for that.

I still see Apple eventually moving its laptops/desktops to ARM to be able to control the hardware even more like they do with their A series chips in the iPhone/iPad ranges, with MacOS/iOS merging effectively like MS are doing with Windows 10 in a sense (Xbox One/Mobile/Desktop&laptop will all get it). Apple also don't sell much in the way of desktop parts for Intel anyway, as all the iMac and Mac Mini range use laptop cpus etc, only the Mac Pro using Xeons, but the time between refreshes for that might likely see it killed off before long. Though Apple are also diverging into areas Jobs never would have allowed so theres the risk of them making a pigs ear of some future releases if going on past experiences of them doing that.

As I say though, all very interesting, but likely all going to turn out to be untrue in the end.

 

Yep this would be a massive win for Microsoft in terms of Xbox exclusivity. From a demographic, brand and services PoV Xbox is MS' ace card, so they'd be willing to spend money on 'winning' next gen. I mean decision can't be worse than buying Nokia...
PS and Nintendo would be ****ed out of GPU deals. Nvidia/Sony burnt that bridge and Nintendo would probably go full ARM/Mali. By next gen could be argued ARM might be able to provide a suitably powerful CPU core though. (side note: It's a shame Ninty went AMD this gen, I was hoping for an ARM/ATI mix)

But ultimately: "If MS was to buy it would be for specific products ie. xbox, surface or servers and not for public use" - this, yes. Same with Samsung and Qualcom. No one is interested in the PC retail market.

bionicgeekgrrl - 42% of new Dell PCs are shipping with a Linux OS in China now. Mostly Gov/Mil/some local-only business but it could end up being consumer too, which will leak into software, middleware and games eventually. SteamOS won't be the breakthrough.

 

IIRC the licensing agreement between AMD & Intel is automatically canceled if either company changes ownership, although I'm guessing that wouldn't include spinning off the GPU division.

 

hmmm been reading a bit about securities stepping in if that was to happen, that or Intel would sign an agreement and quickly with whoever buys the; again monopolies department would take a close look (via dotn make X86 cpu`s anymore)