Rivals Intel and AMD Team Up on PC Chips to Battle Nvidia
wsj.com
wsj.com
Am I the only one who smell this as very "Apple" wanted?
I dont think AMD will be giving up any GFX secret, more likely this is AMD shipping Intel a Mobile Gfx Die to be integrated within the same CPU package.
But in any case, Why not just have Intel ship a Mobile CPU without iGPU and a Separate GPU.
And AMD, why now? When Zen is doing great, has great roadmap and potential, along with much better GFx then Intel. Why?
Edit: OK, I didn't read carefully, while this is WSJ, it is still a rumor, nothing has confirmed.... yet.
Edit2: It is confirmed now.
Edit3: Yes AMD will be shipping die to Intel, and it is EMIB at work. https://www.pcworld.com/article/3235934/components-processor...
Power, manufacturing costs, integration and testing costs, design size...
> Am I the only one who smell this as very "Apple" wanted?
Why would Apple care about NVIDIA? They're already beating them in the kind of performance that matters on products Intel will be competing for.
AMD hardware at a specific price point is more powerful than nVidia hardware at the same price point. However, nVidia has superior drivers that eliminates the difference.
And since Apple prefers to use their own drivers, nVidia loses their main point of differentiation.
But of course the "Apple" drivers for video cards are basically vendor drivers with Apple doing QA & release management. But the driver is nVidia's secret sauce, they're not going to show it to Apple, since Apple is now a very competitive GPU manufacturer. At some point Apple will probably put their own GPU's inside Macs, so nVidia doesn't want to give them a head start.
AMD cares less about giving Apple a head start because they care more about the short term than the long and competing against nVidia.
As for the performance per buck thing it’s not accurate. For gaming and 3D applications AMD has a huge bottleneck in thier geometry pipeline which causes the to underperform in games especially pre VEGA. NVIDIA also used tile rasterization and shader output caching on die to increase performance this is something that AMD adopted only with VEGA.
Pure flops don’t mean much even for compute.
https://newsroom.intel.com/editorials/new-intel-core-process...
https://hardforum.com/threads/from-ati-to-amd-back-to-ati-a-...
Well, last time AMD had an upper hand on the CPU's technical features, they tried taking Intel into a fight and couldn't handle it. Why would the same thing have a different result now?
It is much smarter to not have a full-on direct fight.
I agree with your conclusion that fighting head on right now would not be smart when they have nvidia on one side and arm on the other. But not with your assertion that "AMD failed to defeat Intel products in the market" the last time.
I'd love AMD to bloody Intel's nose (Intel deserves it after their payola anti-competitive behavior against AMD in the 2000's), but it's clear that Intel is too big to fail.
So AMD can win by helping Intel win.
It is not good in the long run for consumers, but it's probably the right move for AMD today.
I also hope this is the start of a long journey where AMD chips away bit by bit at Intel (near-)monopoly in the CPU market.
Why not now? Why not generate additional income from the sale of these Intel/AMD CPUs instead of relying on just Zen?
This way, AMD can make money both when a Zen and a Core CPU is sold.
These chips from Intel are going to be the super-high top end mobile chips in laptops that cost serious money, and have a rumoured 45w TDP.
Raven Ridge on the other hand is a traditional AMD APU, albeit with a decent CPU component - it's not going after the same market segment at all, it's going after the segment where you want to maybe do some light e-sports gaming.
This way, AMD gets both bites of the cherry - they get to sell their APUs with Zen, and they also get to target the super top end.
credit to trynumber9: https://news.ycombinator.com/item?id=15635771
1) AMD must be preventing Intel from in future building similar integrated GPUs using anything like AMDs patent portfolio in GFX.
2) Intel will be able to 100% push out any vendors from moving to Ryzen for presumably faster integrated graphics.
3) Nvidia doesn't have an x86 CPU and last time I looked - all PCs and Mac laptops were using x86.
As some have speculated maybe this is Apple telling their suppliers to jump and Intel and AMD said how high...
I'm going to be interested to see if we ever get a jump to Apple using ARM and internally designed GPUs, I have a feeling that Jonny Ive must be wetting himself over making a reasonably powerful laptop that thin.
1) AMD can sell chips in high end systems where it can't compete with Intel on CPUs or Nvidia on GPUs.
2) Intel can ship a high end graphics experience without dGPU which gets them a level of graphics in a form factor they couldn't otherwise achieve.
3) Ryzen is a low/mid range chip. Even if it could match Intel's performance it could never match Intel's brand, and Intel doesn't want to match AMDs price so they will stay in different market segments. Ryzen sales will not be hurt.
4) AMD gets valuable brand recognition by getting Radeon into more premium devices which could actually boost sales for cheaper Ryzen/Radeon devices down the line.
5) Shafts Nvidia which is a win for both sides.
6) Opens lines of communication for possible future merger or fab deal, which is not so much an issue from an anti-trust standpoint when you look at the total competitive landscape of ARM, Nvidia, Apple, Qualcomm, etc and the shrinking relevance of x86 in the big picture.
If there is a merger, I hope this will make a lot of AMD dream projects like the HPC APU take off, but I also fear that it might lead to a more stagnant intel+amd in a few years time.
Having said that, it MIGHT open up the potential for AMD running (some, all?) chips through Intel's fabs and actually enabling a nearly competitive playing field.
The best thing that the (US) government could do for the market would actually be to force Intel to split in to a fabrication company and a chip design company. That would enable the military to contract fabrication of higher-end validated CPUs on US soil from whichever sources they wanted...
(i.e. argued that way this would be more like "setting up for failure" on AMDs part.)
For the average consumer, every high end system they have seen in recent memory is Intel, and every ad they have seen is Intel, and if they have seen AMD at all it has always been positioned as a budget product.
That is going to take a long time for AMD to turn around assuming they can sustain performance parity with Intel.
That is also why AMD is focused on servers and semi-custom where they are selling to technical people who are evaluating based on price/performance and not based on brand perception.
XBox One's processor is from a pre-Ryzen generation, so this should not be a concern. :-)
Imho parent was talking about the PC sector and in that regard, AMD has been sadly trailing way behind Intel/Nvidia for quite a while now. AMD has struggled to oppose Intel's i5/i7 dominance. Similarly, they still have no real competition for the high-end NVidia GPU's, like the 1080's.
If money is not a limiting factor, then you will be hard-pressed to come up with a build that does not include an Intel CPU and at least one Nvidia GPU.
If money is not a limiting factor, you probably look at AMD EPYC now.
Ryzen helped some with that but afaik Intel's chips are still way ahead in terms of single-thread performance, which is something many desktop users (like gamers) are looking for.
On one hand, it's cool to have i5's/i7's last so long, on the other hand, it's never a good situation when there's no real competition. Too bad Intel didn't pull through with Larrabee, a third player in the dedicated GPU market would have made this whole situation way more interesting/dynamic.
Tho could just as well have resulted in AMD going belly up, trying to compete against Intel and Nvidia vs the current situation sounds like it would been a worse deal for AMD.
AMD are on the brink of offering products that for the first time since forever at least has potential to build up their brand vs Intel, to actually have a significant differentiator that they can market. It seems really harsh to kill that before it even begins. But yeah, it could actually be a sign that AMD knows it couldn't go the whole nine yards in this game anyway.
I've had friends go out to buy a computer and insist on an i7 because "it is the best."
Same people who buy the newest Samsung Galaxy every year.
But i9 is the best... ;-)
When I see "Intel Inside", I immediately think "Intel Inside - Idiot Outside". ;-)
CPUs aren't the bottleneck though, disk I/O and I/O in general are where most "slowness" issues are promulgated from.
But this is server compute stuff which this news doesn’t seem to be about.
AMD can be Aldi, Intel can be Whole Foods.
That sounds a bit far fetched IMO. They aren't that dominant in the PC market (esp. compared to the Big Three (Lenovo, HP, Dell)) for Intel and AMD to warrant such an "out there" cooperation just at the behest of Apple...
a) Sure, Apple doesn't have a low end and as such higher overall margins, but: The Big Three all have a sizable mid-/highend/ultrabook/business range as well with just the same i5/i7/core m CPUs (only updated more often). Apple surely is a good customer, but I severely doubt that the overall margin difference for Intel between them and Apple is even in the remote vicinity of an order of magnitude. What is an o.o.m. higher though is the market share of those three (~60% vs ~6%).
b) I just don't see a bargaining chip. What's the "or else" from Apple's side? Full Ryzen? Would actually be cool, but Ryzen is still completely unproven in the mobile world, and this would be all the more reason for AMD not to enter the deal. ARM? Yeah...I wouldn't hold my breath.
You have to consider that this isn't a simple internal SKU customizing, but a significant and delicate licensing deal with THE direct competitor in x86 world. (The tech is actually the simpler part, seeing that it's not an integrated solution but a MCM.) Apple would have to have severe leverage that they would be the main driver for Intel for such a move. Or they are paying a real buttload for this tech - but then you can be sure they want exclusivity. And the statement really doesn't sound like it.
[1]https://www.anandtech.com/show/12003/intel-to-create-new-8th...
I am probably out of my depth guessing why companies are doing this with NVMe. Maybe its just that the market does not exist for it maybe its just the cost...
I think Intel have a better chance of catching up with Nvidia at this stage. They've been on an acquisition spree and have picked up a huge amount of DL-related IP. They have immense R&D and fab resources at their disposal.
https://wccftech.com/nvidia-volta-tesla-v100-gpu-compute-ben...
AMD's 16-bit packed performance with Vega is more than respectable vs NVidia's 16-bit packed performance in Pascal.
In the future, all AMD needs to catch up to Volta's Tensor cores is to build Tensor cores themselves. That doesn't seem like a major technical hurdle. I'm fairly certain that Google would be the primary patent holder on Tensor-cores.
I haven't kept up to date on what AMD has done with that work, but i believe they use it as part of their compatibility story.
They had to be beaten to finally start proving a bytecode format similar to PTX, for multiple languages, while accepting that most researchers want to use C++ or migrate their Fortran code into GPUs.
Meanwhile on the CUDA C++ side,
"Designing (New) C++ Hardware” - https://www.youtube.com/watch?v=86seb-iZCnI
So practically speaking, OpenCL 2.0 is the best you can get, unless you want to run on Intel's iGPUs or Intel's AVX 512 on their CPUs.
AMD does have support of C++ in OpenCL 1.2 as an optional extension, but their support of C++ in OpenCL 1.2 doesn't work when you enable OpenCL 2.0 for some reason. Also, the CodeXL debugger only works on OpenCL 1.2 at the moment...
As far as I can tell, AMD's implementation of OpenCL 2.0 is still early stage. Its fine if you're cool with debugging with "printf" statements.
this is what OpenCL is, afaik.
Drivers are only part of the story.
Even Google preferred to create their own Renderscript dialect than supporting OpenCL.
Nvidia will be hard to beat; they're going to be the next Intel.
Why? It’s really a shit api, hard to program for and hard to make efficient. I think AMDs problem is that they didn’t make their own tooling pipeline (and then hopefully make an open standard out of it). Instead they stuck with a crappy open standard with poor tooling because, well, it was the standard.
I don't know about that... Sure they've had to liquidate some things but that made them survive long for Ryzen and Vega to come out which has enabled them to claw their way out of the red and into the black for the first time in forever.
PS: Am I right to assume that by "bad decisions" you mean things like the "Bulldozer" architecture with that wonky resources-shared-between-cores-thing?
This is a quote:
"Are we afraid of our competitors? No, we're completely unafraid of our competitors," said Taylor. "For the most part, because—in the case of Nvidia—they don't appear to care that much about VR. And in the case of the dollars spent on R&D, they seem to be very happy doing stuff in the car industry, and long may that continue—good luck to them. We're spending our dollars in the areas we're focused on."
"Car stuff" being self-driving cars, while "the areas we're focused on" is VR. From http://arstechnica.co.uk/gadgets/2016/04/amd-focusing-on-vr-...
AMD has made numerous press releases about supporting deep learning, sometimes via OpenCL, sometimes via cross compiling CUDA or sometimes something else.
I used to get excited about it.
Now, I have a rule: don't get excited about AMD (or Intel, or any new hardware) until they are winning at training neural networks on an absolute speed basis. (Note: vendors will frequently release benchmarks showing how they beat Nvidia. Almost inevitably these are for inference, and often on a speed per watt or speed per dollar, or you can't actually buy the hardware.)
I wonder how much of this is a fix for GPU reliability rather than GPU performance.
My experience with Intel iGPUs on skylake and kaby lake is quite different. They work very well on Linux.
Kaby Lake's GPU is not a change from Skylake, so I expect the same problems, just in Win10 instead of 7.
I don't think a security fix is going to address your GPU issues. You should update to a supported operating system.
Also, we build systems which run our application suite controlled by a custom control panel, and we have yet to manage to fully tame Win10's intrusiveness into the operator's workflow. Even if we do fully tame it, maybe the next version will change the number of things needing tamed. I'd use Windows Server instead, but it isn't supported by ASUS on desktop motherboards. We're in a dead end and seriously thinking about Linux at this point.
https://newsroom.intel.com/news-releases/intel-announces-new...
Since both Nvidia and Intel are monopolies in their respective fields.. that was far more egregious towards AMD than this is towards Nvidia.
Then there's the option of Intel buying just the Radeon parts of the company. However seeing as the GPU market seems to be growing fast, it would have to be quite the offer for shareholders to agree.
[0] Played a cowboy game on one of these in a Shenzhen VR house, most fun I've had in years! Just don't look at the video of yourself playing afterwards!
[1] Cheap realism.
Somehow I doubt, AMD want to give away their APU competitive advantage to Intel.
AMD GPU Die + Intel CPU die on one package.
And as for game consoles, both Xbox One and PS4 uses an AMD. Only the Switch uses NVidia.
Nvidia should get an automatic x86 license.
Actually, I have a proactive anti monopoly idea. Any company that is the predominant player in a market cannot use patents to limit the ability of a competitor to make a compatible product.
This would mean anyone could make x86 chips w/o a license.
My best guess is Apple -- they were already using Intel CPUs with AMD GPUs, and they probably opened up their checkbook to make this collaboration happen. It should give them performance and power benefits, and Intel and AMD sell the same number of CPUs and GPUs, respectively, as before.
Not sure why you think this is so anti-competitive. Who knows if NVIDIA was offered the same deal and declined? They are much less prone to making "semi-custom" parts for third parties, while AMD already has a track record of doing this for Sony with the PS4 and Microsoft for the Xbox One.
Source: https://newsroom.intel.com/news-releases/intel-announces-new...
Good luck with that argument.
Why would AMD agree to this? This would massively eat into Raven Ridge laptop sales.
They could've also replaced all of their dual-core laptops with quad-core Ryzen APUs for about 2-2.5x increase in both CPU multi-thread performance and GPU performance. And I don't think it would've cost them more, or not significantly more at least. AMD seems to price their cores at 50-60% of Intel's cores.
Laptops have always been the most profound demonstration of Intels monopoly because they are tightly integrated products that end users never get to customize deeply. So its really easy for Intel to just "persuade" notebook vendors to only put AMD chips in garbage models, if at all.
The keyboard is crap. It has a dedicated key for a 'Look! This is my current system load. You can even connect a mobile to see it on your phone!' ROG key where a numlock would be. It doesn't have a End key, which infuriates me. There's a power button in the top right of the keyboard which I WILL press in the future. It's just a failure about to happen. The keyboard is utterly flawed.
Well.. I at least have a Ryzen 7 with 8 cores, right? Yeaaaaah... Right now the braindead EFI interface (you can switch between 'ez' (sic) and advanced mode) doesn't seem to expose a method to enable AMD-v. So ... Yeah. I have 8(16) cores to .. I don't know. Watch YouTube I guess.
AMD and Asus really dropped the ball here. The keyboard is unacceptable. The AMD-v issue is .. low. So fucking low.
Don't buy this laptop?
AMD licenses x86 from Intel, and Intel licenses x64 from AMD (because Itanium failed to win the 64 bit market).
Didnt Intel play dirty with AMD by forcing OEMS to use Intel chips instead of AMD?
P.S AMD K series were named after kryptonite as Intel's pentium series were considered to be superman.
Intel and AMD seems to have seen the light of Linux. I wonder who's next - Microsoft perhaps ditching Windows for Linux and building a super GUI on top of Linux to make it an OS X killer?
What this deal will do is allow Intel to become the de-facto leader in the growing "laptop gaming" market, displacing both Nvidia and AMD from that market. AMD had a chance to dominate that market now with multi-core Ryzen CPUs and its dedicated GPUs, but it seems they've just decided to hand that market over to Intel on a silver platter.
Such a huge mistake from AMD. This is why I would rather AMD would be acquired by someone like Samsung or Broadcom to give AMD the money it needs than do stupid deals like this one with Intel because it's so strapped for cash.
The only competitive part from Intel is Iris Pro 580 with the Crystal Well cache, and those parts are capital-E expensive.
https://www.youtube.com/watch?v=h0MeI0sQfy4
https://www.notebookcheck.net/Mobile-Graphics-Cards-Benchmar...
Intel's highest end can't meet Nvidia's lowest end embedded, which can be had at a very affordable price.
Intel wants/needs better GPUs than their own ones for their laptop offers, a fairly impressive admission of incompetence, and (as others noted) AMD lacks laptop CPUs and wants to sell more laptop graphics solutions, which are supposed to be their specialty. Good products can be expected, but I suspect Intel might strike a similar deal with Nvidia: whatever thin laptop the public buys, it's going to have a Intel CPU.
If it were easy to build a CPU competitive with Intel, AMD would do it more consistently.
CPUs need to make legacy code up to 40 years old run faster so there is a ton of complexity in the hardware. They are chasing people who want modest speed bumps without large changes.
Kinda like how apple was able to pull off a performant phone/tablet after Microsoft failed a bunch... Because they got people to rewrite apps (or create) for their platform instead of shoehorning windows apps into a different form factor. Much bloat was cut, usability was redone. It's an analogy so don't go silly over the differences.
As an aside: I know it's not the done thing to complain about down votes, but I do find an honest, polite question - not even a statement - being modded to oblivion somewhat unusual.
Did they at least revise their licensing deal where Intel basically adds a requirement that AMD can't be sold to other companies? If not, then AMD's leadership must be clueless. They should've revised that clause the first chance they got to make another deal (like this one!) with Intel.
This just helps them make inroads with Intel who already locked up much of the laptop market. I'm pretty convinced there was no way AMD could enter that market for a few years in any meaningful way regardless.