AMD CEO talks of long-term turnaround

[Arachnotronic]

Ah, here we go again. The usual AMD doom and gloom rant, coupled with some armchair CEO advice.

Would you argue that things are going great for AMD?

 

[ShintaiDK]

Ah, here we go again. The usual AMD doom and gloom rant, coupled with some armchair CEO advice.

Given your past posting record I'm not sure how much faith to put in that advice. Seems like if anything you want AMD to go bankrupt as quickly as possible, so Intel can get its monopoly that you long for.

Yes thats why AMD financials tells how great it goes contra what we say. Because we are so blinded by hate to a company that we cant tell fiction and reliaty apart.

Q1 2012 1203M$ CPU revenue.
Q1 2013 751M$ CPU revenue.
Q1 2014 663M$ CPU revenue.
Q1 2015 532M$ CPU+GPU revenue.

AMD today worth 1.52B$.

Oh wait, can you please inform AMD and the stockholders as well? They must have missed your memo.

 

[mrmt]

Ah, here we go again. The usual AMD doom and gloom rant, coupled with some armchair CEO advice.

I didn't give any advice, just pointed out that AMD still has to provide ROI for a process node.

Btw, did you check the accounting rules to see that AMD will gift we shareholders with another write down?

 

[Fjodor2001]

My previous post was primarily in response to ShintaiDK's.

 

[Fjodor2001]

Would you argue that things are going great for AMD?

Where have I? And if you talk about the doom and gloom crowd, my point is that they are over the top. They've been predicting AMD's imminent demise for the last 10 years.

 

[ShintaiDK]

My previous post was primarily in response to ShintaiDK's.

Where have I? And if you talk about the doom and gloom crowd, my point is that they are over the top. They've been predicting AMD's imminent demise for the last 10 years.

Thats how you see it. Not like how others do despite what you try and make up.

I always said AMD is the new VIA. And they are.

 

[Arachnotronic]

Where have I? And if you talk about the doom and gloom crowd, my point is that they are over the top. They've been predicting AMD's imminent demise for the last 10 years.

AMD's competitive positioning in its key markets (PCs, servers, graphics) has deteriorated pretty heavily over the last 10 years.

I would say the "doom and gloom crowd" has been more correct than the perpetual optimists who buy into the idea that AMD's "next big thing" will turn the tables on Intel/NVIDIA and restore AMD to its future glory.

Or would you disagree with that?

 

[DrMrLordX]

I stated my position and stand by it. It still seems like a very bad value compromise to me to not double the performance for the price of one game. The point is really not whether an APU can play certain games, but whether or not something *much* better is available for a very small cost, and it clearly is. I just dont buy this "oh, I am going to get into PC gaming, but 50.00 more for double the performance is going to send me to the poorhouse." But each to his own I suppose. If someone wants to severely limit the utility of their system which will last several years, for 50 measly dollars, they certainly are free to.

Many people (such as the aforementioned f2p gamers) don't really know or understand the difference between a 240, a 250, a 250x, a 260x, or any number of other video cards that might be priced within ~$50 of one another. If I'm a relatively-ignorant f2per that just doesn't want to do a few hours worth of dGPU research, what I'm going to do is find a PC guy I know (if I know one) and ask him, "what's going to let me play DotA2?" And he's going to say, "uhhh all of those". If by some bizarre coincidence the subject of an AMD APU comes up (perhaps my ignorant f2p-ing butt is about ready to drop a few dollars on an 860k + 240, or an i3-4130 + 240, or whatever), my PC guy might say, "you know you can just get a 7650k and call it a day, it's about the same. I'll help you set it up".

There's almost zero cost benefit to me (the ignorant f2p-er) when buying a 250x or 260x when all I plan to do with the thing is play Dota2 anyway. You might say, "but what about future games!". Well, it's 2015, and people are still playing TF2 using a tired old engine. They might shock me and launch Dota3 with stringent hardware requirements, but really, their economic model depends on them not doing that. Forcing a bunch of their fans to buy new hardware takes money off the table that the fans could instead spend on cosmetics and such. F2p hardware requirements are probably not going to change a lot in the next 2-3 years. If there's anything that's bad about trying to sell hardware to this crowd, is that once they've got something that works "good enough", they'll probably stick with it for a long time.

All you have to do is look at APU sales and see that most people seem to agree with me however. I think APUs without high bandwidth memory are just an unsatisfactory solution, and charging 20 or 30 dollars less would not change that. And actually Kaveri is cheap now, and have sales taken off??

Kaveri would have been at its best in some cheap OEM machines, where AMD has been having problems getting traction with anything. And the A8-7600, at launch, was hailed as being the best chip of the lot. Problem was that you couldn't get one anywhere. People really wanted to buy one, in early 2014. Maybe not enough of them to make a difference, but the desire was there. AMD instead showed up with the $180 7850k and a, what, $160 7700k?

At this point, I think most of the people who were interested in such things (or could have been) have just moved on. OEMs do not seem to be flooding the market with cheap Kaveri, no. I don't think it's relevant that AMD is essentially trying to underprice Intel's 100mm2 die chips with ~200mm die chips, since for a large portion of Kaveri's lifespan, AMD didn't even bother to show up with the 7600. Yeah maybe Intel could have busted them on cheap-ness with some i3s + dGPUs or whatever, but they didn't have to. Even today, if you go to a big box store, it is damn near impossible to find a proper Kaveri quad in an OEM box as a step-up from a Jaguar chip or a Kaveri dual-core (and even those are not very common). You can find i3-based OEM machines all day long.

 

[Fjodor2001]

I always said AMD is the new VIA. And they are.

That would effectively give Intel its monopoly in the x86 desktop and server CPU segment too, which is what you desire regardless of how it happens.

But wait, you're saying that AMD already is the the new VIA!? They are not covering the same market segments at all. Does VIA make high performance desktop and server CPUs, GPUs and APUs?

 

[myocardia]

Does VIA make high performance desktop and server CPUs, and APUs?

No, but neither does AMD.

 

[ClockHound]

No, but neither does AMD.

Zing! That's gonna leave a mark.

However, in their defence AMD is trying, VIA, not so much. Many seem keen to kick the carcass to the curb.

High tech and primitive tribalism always make for interesting threads tho. Please carry on. Or carrion.

 

[jeff_rigby]

Come on jeff, think harder. If half of what you are writing were true why isn't AMD management touting is as the next big opportunity and doubling down their bet on embedded, but instead they are saying that they need to open up new markets far different from embedded?

AMD won't realize any benefits from the 3 network convergence and AR till after HEVC, Windows 10, Playready ND, Vidipath, 4K blu-ray, FCC DSTAC (2016) and they all start advertizing.

Marketing and consumer education is being delayed by EVERYONE. The Cable operators are ready to support Vidipath as required by the FCC but no clients are on the market. For the XB1, PS4 and PS3 it just requires a firmware update. For most ARM phones and tablets, it just requires Android L and an app. For Windows 10, just an app assuming the hardware supports TEE.

And there is the selling point for AMD; 4K and Windows 10 requires new hardware with TEE level DRM support. PCs will move to the living room starting with the the XB1 and PS4 game consoles which are 2013 designs with expensive workarounds to support 2017 features.

Chicken and egg issues are being overcome with "Game Consoles" as Sony did with the PS3, the blu-ray standard and DLNA. You can see the exact same roadmap in China;

1) (2006-2008) games and DLNA to create an ecosystem which when enough STBs are on the market, are used to support
2) (2010-2015) IPTV VOD with *advanced features* to be supported eventually by
3) (2015-2020) RF Vidipath which can support many more people on Cable than IPTV VOD which requires more bandwidth and a server for every client. A few IPTV VOD 4K channels
4) (2017 on)Till DOCSIS 3.1 and multi-stream where RF channels are phased out and everything is Vidipath IPTV served from a Cable modem. More 4K channels
5) (2020 on) 4K TV on Cable TV is mainstream via Vidipath IPTV from a cable modem. STBs downscale for 1080P TVs

Everything in the roadmap above except Vidipath Smart TVs will require a PC as STB or game console with hard disk. (NRT and DVR)

*Advanced features* = NRT, 1080P, S3D, XTV, Targeted commercials, Skype-ooVoo video chat, browser, NUI (Natural User Interface) voice and gesture with consumer blink rate and attention to areas of the screen used for targeted advertising and more, second screen, in home streaming of media and games and more.... Comcast has tweets able to take you to a TV channel and the same is likely from a browser, Email or Facebook in a tablet or the browser in the TV or STB. Miracast is a standard for Vidipath STBs.

Comcast already allows sharing home videos and pictures from a phone/camera directly to a X1 cable TV STB from anywhere to anywhere.

ALL AMD APUs have a trustzone processor for TEE, Xtensa stream processor and Gesture control middleware in the ARM block and it's low power. Xtensa processors can do audio, video processing and codecs and depending on the design/power even more.

Kaveri and Carrizo have full video and audio processing designs that don't need to use the GPU shaders for Codecs including HEVC. The GPU can be turned off with full screen video; same for XB1 and PS4. NUI and traditional controls do not need the GPU on with Xtensa processors which can also do UIs without waking the main GPU

 

[Fjodor2001]

No, but neither does AMD.

Really? So what market segment would you put them in? The same as VIA's according to ShintaiDK's flawed techical understanding? Get real.

Not making the fastest option in a segment does not mean that the products do not belong to that segment.

 

[ShintaiDK]

AMD is acting exactly like VIA. Its a historical rerun.

Why dont you elaborate us with the vision of yours? The financials doesnt lie, so what are we all missing?

 

[Fjodor2001]

You said AMD already *is* the new VIA. Care to clarify that? They are not even in the same market segments.

 

[ShintaiDK]

You said AMD already *is* the new VIA. Care to clarify that? They are not even in the same market segments.

Why dont you start with the vision of yours since we are all wrong and you are right. AMD is 1/3rd of itself in the PC segment over 4 years. And the downturn is accelerating. From the looks of it the PC division may be around 450M$ revenue in Q2. 475M$ with a bit of luck. Unless consoles tanked.

 

[Glo.]

Shintai, Im not arguing that AMD is in trouble right now, however, judging the future by one simple statement is... plain stupidity.

What Lisa Su said can mean EVERYTHING. With the inclusion of simple idea, that in first place AMD will develop for Workstation and pro markets, and they will derive GPUs from it.

It can mean that they will develop for Mobile in the first place, and from that they will derive GPUs.

It can mean absolutely everything.

And saying that AMD is new VIA is spreading FUD. Sorry. Only you think that AMD is stupid and doomed with that form of thinking. They only want to adapt to where the market is going, not adapt the market to them.

Edit. What I believe Lisa meant is that they don't want to develop for open PC's. It does not mean that they will stop developing GPUs and CPUs. The market is going in the direction of closed architectures. Consoles, laptops, smartphones, tablets, all-in-ones. You will not buy one part, but whole computer. That way of design gives every producer more revenue income every year, with designed obsolesce. Is it hard to see it?

 

[ShintaiDK]

Shintai, Im not arguing that AMD is in trouble right now, however, judging the future by one simple statement is... plain stupidity.

What Lisa Su said can mean EVERYTHING. With the inclusion of simple idea, that in first place AMD will develop for Workstation and pro markets, and they will derive GPUs from it.

It can mean that they will develop for Mobile in the first place, and from that they will derive GPUs.

It can mean absolutely everything.

And saying that AMD is new VIA is spreading FUD. Sorry. Only you think that AMD is stupid and doomed with that form of thinking. They only want to adapt to where the market is going, not adapt the market to them.

Edit. What I believe Lisa meant is that they don't want to develop for open PC's. It does not mean that they will stop developing GPUs and CPUs. The market is going in the direction of closed architectures. Consoles, laptops, smartphones, tablets, all-in-ones. You will not buy one part, but whole computer. That way of design gives every producer more revenue income every year, with designed obsolesce. Is it hard to see it?

Its not about 1 statement. Its about year after year after year with a PC division that keeps plummeting.

Let me remind you:
Q1 2012 1203M$ CPU revenue.
Q1 2013 751M$ CPU revenue.
Q1 2014 663M$ CPU revenue.
Q1 2015 532M$ CPU+GPU revenue. (This may mean less than 400M$ CPU revenue)
Q2 2015 ~450-475M$ CPU+GPU revenue?

Why is it saying doom and gloom that AMD is the new VIA? VIA still exist. Isnt the issue rather that you dont want AMD to be like VIA? 10 years ago VIA had something like 5% CPU share and 15% Graphics share.

 

[Glo.]

I absolutely know what you are referring to. R&D expenditures. But do you have ANYTHING for your statement apart from R&D expenditures, and what Lisa Su said?

No. Its completely too early to judge what will happen in few years from now for AMD.

 

[mrmt]

Edit. What I believe Lisa meant is that they don't want to develop for open PC's. It does not mean that they will stop developing GPUs and CPUs. The market is going in the direction of closed architectures. Consoles, laptops, smartphones, tablets, all-in-ones. You will not buy one part, but whole computer. That way of design gives every producer more revenue income every year, with designed obsolesce. Is it hard to see it?

No, this is not what Lisa Su said.

Check the FAD presentation. She basically said that consumer GPU and GPUs, the ones we use in smartphones, tablets, all in one and other short life cycle devices are an area that won't get much attention from AMD. Embedded, OTOH, and that means longer life cycles devices like consoles, well, these will get attention from AMD management.

She basically said at the FAD that she will try to transform AMD into a server/HPC company, with a small embedded business attached. Consumer will basically get the left overs.

No. Its completely too early to judge what will happen in few years from now for AMD.

No, it is not. It was way apparent for some people that Bulldozer was a disaster and that Rory wasn't doing the right things for the company, and here we are. Some things Lisa is doing are already giving hints on what kind of trouble she will face.

Lisa will have to make some very tough decisions, rather sooner than latter, on whether to sell assets to fund her strategy or if she's going to, well, streamline and optimize AMD a lot more.

 

[ShintaiDK]

I absolutely know what you are referring to. R&D expenditures. But do you have ANYTHING for your statement apart from R&D expenditures, and what Lisa Su said?

No. Its completely too early to judge what will happen in few years from now for AMD.

I am not even talking about R&D. That is another subject that AMD got huge issues with. No, I am talking revenue and marketshare.

 

[Glo.]

Rory is not anymore in AMD.

I will not speculate anymore about future of any company, because it can change immediately there is new factor added to market the company lives.

 

[MiddleOfTheRoad]

Where have I? And if you talk about the doom and gloom crowd, my point is that they are over the top. They've been predicting AMD's imminent demise for the last 10 years.

Actually, it was more like the past 25 years.

 

[MiddleOfTheRoad]

Why is it saying doom and gloom that AMD is the new VIA? VIA still exist. Isnt the issue rather that you dont want AMD to be like VIA? 10 years ago VIA had something like 5% CPU share and 15% Graphics share.

I actually agree with that sentiment. I do believe AMD is shrinking to become a smaller player like Via. I don't see AMD going anywhere unless someone buys them. They will probably just muddle along like they've done for most of their existence. VIA wasn't particularly profitable when they had all that marketshare a decade ago -- so they shrunk down their operations and focused on improving margins. I don't understand why anyone would view that as a bad thing. Better to be profitable and agile -- than bloated and bleeding.

 

[AtenRa]

AMD is acting exactly like VIA.

High performance ZEN CPUs for Server and Desktop.
High performance APUs for Mobile (Laptops) and Desktop
High performance GPUs for HPC, professionals and Desktop Gaming.

Nothing VIA related in those products.