"AMD Moves Away From PCs Amid Steep Losses"

[Olikan]

Edit: Car analogy. Most 5 or 6 year old used cars still get you from point A to point B adequately. Yet many people still upgrade every few years because they want something faster, quiter, with new features, etc. And a new computer costs only a few hundred dollars compared to many thousands for a new car.

yes, but that is the problem...it's bad for companies to live by upgrade....

Intel is this big today, because nobody had a shinny computer in the 90...well,intel did
and nobody had a shinny laptop in the 2000...well, intel did
right now, nobody have a shinny tablet...apple does, intel don't

oh...and GM wanna talk to you 😛

 

[blastingcap]

Cars are also sexy and sleek and people like to look at them. How many desktops are sold for the same reason? What about repairs? At least Intel got the hint and tried that with Ultrabooks, but they weren't willing to compete on price and now they're paying for it with sluggish sales.

This is a very techie enthusiast board that loses sight of how the rest of the world views desktop PCs. For most people, a PC is sort of like an appliance. Most people just want to buy it and not replace it until it gets unbearably slow or something, sort of like an appliance (fridge, washer/dryer, etc.). A desktop PC you can just turn on a few hours a week and it'll last a decade. To this day, Win XP is still wildly popular (until recently it was still the no. 1 OS in desktops) because people treat their desktops like appliances, and Win XP with a decent hard drive is good enough to run MS Office and a web browser, so there's no good reason to upgrade. http://www.webpronews.com/windows-7-finally-dethrones-xp-as-most-popular-operating-system-2012-09

Even the quintessential enthusiast upgrade, the SSD, is overkill for a lot of people. My gf was spectacularly unimpressed when I upgraded her ancient Seagate Barracuda 7200.9 to a Samsung 830 256GB. Yeah it boots a little faster and programs pop up faster, but she didn't really care. If it can run MS Office and Firefox without crashing, she's satisfied.

Also, you aren't seen in public with a desktop PC like you are with a car or even a smartphone or tablet or laptop. Pelov is right about this.

Cars though, that's Americana and publicly viewable and something of a status symbol. (I bet every billionaire in the world drives a nice car (or boat or helicopter) or hires someone who does. Many of them probably also have nice smartphones and tablets. But I bet many of them either don't have desktops or else have desktops that would be put to shame by a large fraction of the forumers on here.) Guys like nice cars (or jets, or boats, or bikes, or anything mechanical). Girls can appreciate nice cars even if they never pop the hood. Everyone can appreciate a smoother ride, luxury amenities, etc., and furthermore a large % of the population drives a LOT (commute and otherwise) so cars mechanically wear out over time faster than PCs and must be maintained and repaired throughout their lives.

 

[BallaTheFeared]

My gf was spectacularly unimpressed

Sounds like you need a new girlfriend 😛

The difference between us and them is insane, we appreciate minor differences and they don't care about the biggest. It's like two different worlds 😀

 

[Remobz]

Sounds like you need a new girlfriend 😛

The difference between us and them is insane, we appreciate minor differences and they don't care about the biggest. It's like two different worlds 😀

Haha!

My gf is 44 and she could care less as well. She just checks email, facebook, cooking recipes and gossip sites😀

As long as the computer works and does not crash she is happy.


At least she respects my hobby and gives me my space.

 

[blastingcap]

Sounds like you need a new girlfriend 😛

The difference between us and them is insane, we appreciate minor differences and they don't care about the biggest. It's like two different worlds 😀

She did appreciate the extra speed to a limited extent... but she basically thought it wasn't worth the price, which at the time was around $180-200. That's like half an iPad, or a bunch of nice clothes, or half a bag of groceries from Whole Foods--things more likely to appeal to women than a slightly faster desktop.

I can see her point though; I recently downgraded my laptop by removing the SSD inside (Intel X25-M G2 80GB) and going back to a HDD because I'm limited by bandwidth, not OS drive speed. I sold the SSD.

 

[BallaTheFeared]

I'm not using the i3 system because it's faster in benchmarks, I know exactly where you're coming from 🙂

 

[xmonger]

I bought some AMD stock today at the close.

I am either going to lose every single cent or make a killing if they get it together in the next 3 to 6 months.

I have always had/will have a soft spot for AMD. If they go away it will not be good news for any of us.

 

[ViRGE]

At this point any number of players could acquire AMD outright, or at least a controlling stake. Or they could wait a bit longer and pick the choice pieces from the carcass once AMD enter bankruptcy.

I'm not sure AMD could go bankrupt. Intel needs AMD to stay alive, otherwise they risk the FTC coming in and doing something about the resulting x86 monopoly.

 

[v8envy]

I bought some AMD stock today at the close.

I am either going to lose every single cent or make a killing if they get it together in the next 3 to 6 months.

I have always had/will have a soft spot for AMD. If they go away it will not be good news for any of us.

I am not a broker, and this is not sound financial advice in any way shape or form.

That said, next time consider options. With a stock purchase you're in 100%. If AMD declares chapter 11, you've lost your whole investment. Using options you could invest a much smaller % of the total you are willing to risk today, while still participating fully in any potential meteoric rise.

And make no mistake, you're simply gambling on AMD making a huge move, eventually. I looked at the Jan 2014 calls briefly and then decided it would be less risky and more fun to just fly to Vegas and play roulette.

 

[ShintaiDK]

I'm not sure AMD could go bankrupt. Intel needs AMD to stay alive, otherwise they risk the FTC coming in and doing something about the resulting x86 monopoly.

FTC would basicly kill innovation and development then.

AMD is too small to compete in any way really. All about the volume so to say. Intel wont be big enough if you look down at 7 or 10nm without expanding further. And at 5nm and below, we need something close to world monopoly.

I dont even dare to think what a 5nm fab will cost..40billion? And a 20billion? chip design to follow it.

But again, considering MS, Google and Apple. FTC wouldnt touch Intel or nVidia for that matter.

 

[Vesku]

They haven't replaced anything with Jaguar, Jaguar isn't out yet.

They replaced it on the roadmap, true, but roadmaps are not sellable products.



Make a list of everything you might foresee to be a problem if AMD cancelled Jaguar right now at this exact point in time.

Then write across the top of that list "Things that became a problem when AMD cancelled the 28nm bobcat shrink".

It really is one of those "all of the above" snafu type situations of snatching defeat from the jaws of victory.

But AMD recognizes this, hence the inclusion of the bullet point "significant execution problems" on their slide. No one can claim AMD has no process gap and no execution problems and no design complexity problems when AMD themselves acknowledge as much.

But this still doesn't answer the question "what does AMD think 'agile and flexible SoC methodology' means?"

I think "agile and flexible" means becoming an x86 SoC integrator for other companies. They've been advertising their willingness to design with custom IP blocks. But is there much demand for that? There is the rumor that there will be full x86 SoCs in the next Xbox and PlayStation but thanks to Android ARM has made significant inroads in direct consumer use SoC presence.

Focusing on perf/W and HSA makes a lot of sense. AMD certainly has not been able to keep up with Intel in pure performance since Core 2.

"Focus on disruptive play in data center" is just an extension of perf/W and HSA. Taking those chips and packaging them in a manner that will be attractive to certain data center applications. At least this aspect of the strategy has seen some concrete action, buying SeaMicro.

"Adopt industry partnership approach" is what AMD has needed to do for at least a decade, they just haven't demonstrated much success in that department even after Intel had to tone down it's aggressive market capture practices.

If their Jaguar powered products in 2013 aren't fully HSA, as in shared memory with GCN or GCN+ GPU units then I'll be extremely doubtful of their ability to execute the above strategic plan.

 

[Keysplayr]

I'm not sure AMD could go bankrupt. Intel needs AMD to stay alive, otherwise they risk the FTC coming in and doing something about the resulting x86 monopoly.

Doesn't Intel already have over 80% of the x86CPU market at this time?

 

[Olikan]

FTC would basicly kill innovation and development then.

AMD is too small to compete in any way really. All about the volume so to say. Intel wont be big enough if you look down at 7 or 10nm without expanding further. And at 5nm and below, we need something close to world monopoly.

I dont even dare to think what a 5nm fab will cost..40billion? And a 20billion? chip design to follow it.

But again, considering MS, Google and Apple. FTC wouldnt touch Intel or nVidia for that matter.

FTC won't punish intel with ARM just in the corner....

 

[ShintaiDK]

FTC won't punish intel with ARM just in the corner....

That too. Plus people keep forgetting that a monopoly is not an antitrust case in itself. And you can have antitrust cases without being a monopoly.

And lastly, that competition sometimes in itself can be more damaging than good due to the lockdown in the will to take risks and commit to changes. Usually we use government funded subsidize to move on in cases where that happens.

 

[Olikan]

And lastly, that competition sometimes in itself can be more damaging than good due to the lockdown in the will to take risks and commit to changes. Usually we use government funded subsidize to move on in cases where that happens.

lately, intel is beeing more reactionary than trying something new...
i mean, haswell with it's igp and avx2... is more a reaction to AMD that never-seen falcon chip back in 2009...and OoO atom an answer to ARM...

intel is just putting money and improving things faster, to things that theyr competition just announced

 

[ViRGE]

Doesn't Intel already have over 80% of the x86CPU market at this time?

Sure. And they've had roughly 80% of the x86 market for the last 5 years. They always somehow seem to leave the least profitable 20% for AMD. Funny how that works... 😛

Anyhow, I do seriously believe that Intel only keeps AMD around these days to keep the FTC at bay. The FTC, the EU, and various other anti-trust regulators have already smacked Intel around a couple of times for various offenses. We may consider ARM an Intel competitor, but based on their history with the FTC I believe the FTC wouldn't hesitate to split up Intel if Intel went from 80% of the x86 market to 100%. Or if not the FTC, then the EU would find some way to raise hell.

So in that sense I don't believe it would be possible for AMD to go bankrupt; AMD could give up all x86 R&D and they'd somehow still capture 20% of the x86 market. Intel will always keep their token competitor right where they need them to keep the FTC/EU off of their back.

 

[anongineer]

Competition issues aside, if government IT infrastructure makes heavy use of contemporary x86-64 processors, those countries may want to bail out AMD just to have a second source.

 

[2is]

lately, intel is beeing more reactionary than trying something new...
i mean, haswell with it's igp and avx2... is more a reaction to AMD that never-seen falcon chip back in 2009...and OoO atom an answer to ARM...

intel is just putting money and improving things faster, to things that theyr competition just announced

Intel doesn't need Haswell to react to AMD. Heck, they don't even need Ivy Bridge.

 

[Red Hawk]

I dunno, integrating GCN into the next APU could provide AMD with its Fusion holy grail AMD has been looking for. Its compute performance could finally make things comparable to Intel.

 

[ShintaiDK]

lately, intel is beeing more reactionary than trying something new...
i mean, haswell with it's igp and avx2... is more a reaction to AMD that never-seen falcon chip back in 2009...and OoO atom an answer to ARM...

intel is just putting money and improving things faster, to things that theyr competition just announced

You focus on a way too narrow area. You forget platform integration, energy efficiency etc. What OEMs and majority of customers as for. Without losing performance.

Haswell applies ondie VRM, SOIx states, extra issue ports that will also benefit HT, 256bit wide engine and busses. AVX2 alone can be 50% relatively easy gained performance.

What did the competition announce in those areas? Nothing!

 

[ShintaiDK]

I dunno, integrating GCN into the next APU could provide AMD with its Fusion holy grail AMD has been looking for. Its compute performance could finally make things comparable to Intel.

Problem is the limitations of compute applications. And that will never change.

Plus Intels iGPU already does DX compute for example.

And still, AMDs GCN will just be bolted on via a PCIe style bus like the previous. And any discrete card will counter it and outperform it.

Remove the PR and marketing and look at whats left. And we are back to CUDA style.

 

[Olikan]

Intel doesn't need Haswell to react to AMD. Heck, they don't even need Ivy Bridge.

amd anounced falcon in 2007, what happend to intel igp's since that day?
oh yes, massive performance boosts...

Falcon got canceled and then Llano delayed, sandy bridge got so much igp performance that when llano arrived, it was unimpressive

 

[Olikan]

Haswell applies ondie VRM, SOIx states, extra issue ports that will also benefit HT, 256bit wide engine and busses. AVX2 alone can be 50% relatively easy gained performance.
What did the competition announce in those areas? Nothing!

i can only see ondie VRM as revolutionary, the rest is just evolution...
arguably, ondie VRM can be said to reactionary, because intel competes with M-board makers...that love putting more VRM's for overclock than really needed...anyway,

SOIx states? evolution to core parking
AVX2? answer to Gpgpu
...oh yes a better igp and a better performer HT ...meh

 

[xmonger]

I am not a broker, and this is not sound financial advice in any way shape or form.

That said, next time consider options. With a stock purchase you're in 100%. If AMD declares chapter 11, you've lost your whole investment. Using options you could invest a much smaller % of the total you are willing to risk today, while still participating fully in any potential meteoric rise.

And make no mistake, you're simply gambling on AMD making a huge move, eventually. I looked at the Jan 2014 calls briefly and then decided it would be less risky and more fun to just fly to Vegas and play roulette.

Thanks for the reply. I decided to make an investment in AMD and bought the stock as a value play. I think the stock is extremely oversold at this point and represents a great opportunity to get in at a rock bottom price. The GPU part of the business is estimated to be worth $2.25/share on its own while the whole company is trading at $2.18. The CPU part of their business has significant value. Right now everyone is heading for the doors, but when things settle down there will be a rush back in. Mark this post and come back to it in 6 months.

Further, when you buy/sell options it is a short term trading strategy not an investment. You have to be right on the timing and direction to make money. This is much more difficult, plus it's a leveraged derivative so Vegas is a better bet.

 

[Idontcare]

Thanks for the reply. I decided to make an investment in AMD and bought the stock as a value play. I think the stock is extremely oversold at this point and represents a great opportunity to get in at a rock bottom price. The GPU part of the business is estimated to be worth $2.25/share on its own while the whole company is trading at $2.18. The CPU part of their business has significant value. Right now everyone is heading for the doors, but when things settle down there will be a rush back in. Mark this post and come back to it in 6 months.

Further, when you buy/sell options it is a short term trading strategy not an investment. You have to be right on the timing and direction to make money. This is much more difficult, plus it's a leveraged derivative so Vegas is a better bet.

Value Trap

Definition of 'Value Trap'

A stock that appears to be cheap because the stock has been trading at low multiples of earnings, cash flow or book value for an extended time period. Stock traps attract investors who are looking for a bargain because these stocks are inexpensive. The trap springs when investors buy into the company at low prices and the stock never improves. Trading that occurs at low multiples of earnings, cash flow or book value for long periods of time might indicate that the company or the entire sector is in trouble, and that stock prices may not move higher.

Investopedia explains 'Value Trap'

Companies, and even sectors, can be doomed, because of situations such as the inability to survive competition, the inability to generate substantial and consistent profits, the lack of new products or earnings growth, or ineffective management. Often, a value trap appears to be such a good deal that investors become confused when the stock fails to perform. As with any investment decision, thorough research and evaluation is recommended before investing in any company that appears cheap when reviewing its relevant performance metrics.