Geez, they just keep getting stronger. Would love to see AMD as strong as that.

Pretty incredible results when you take into account that their core business market is shrinking. And that's exactly why Nvidia has been a runaway success compared to AMD: Diversification, platforming and support. Here is an example: http://nvidianews.nvidia.com/news/nvidia-to-train-100-000-developers-on-deep-learning-in-2017

GPU Technology Conference -- To meet surging demand for expertise in the field of AI, NVIDIA (NASDAQ: NVDA) today announced that it plans to train 100,000 developers this year -- a tenfold increase over 2016 -- through the NVIDIA Deep Learning Institute.

These are the data scientists that will be working in the major corporations in the coming years, and guess which hardware and software platform they will be familiar with and request? Nvidia. Nvidia's management deserves high praise for their foresight and their willingness to adapt. Their Tegra program had a good run until around Tegra 3, after which they pretty much had no clients for it. Now instead of what a lot of companies might have done and either closed it and/or sold off the pieces, they turned it around to be their Automotive and Console push, and have reaped the rewards in return instead of the losses. Their diversification into the datacenter also proved to be a runaway success. And while they might have some of the best hardware around, what really drives their success is their focus on the software platforms surrounding their products.

Nvidia practically has a monopoly where their Graphic Card prices go up to the Sky. Regarding graphics card market these years, AMD/ATI can't get even close to Nvidia and therefore we have the "Golden Years" of Nvidia all thanks to AMD only.

Nvidia practically has a monopoly where their Graphic Card prices go up to the Sky. Regarding graphics card market these years, AMD/ATI can't get even close to Nvidia and therefore we have the "Golden Years" of Nvidia all thanks to AMD only.

not only, they have near monopole: -In ProGPU (outside of some Hyped event AMD are always "aviable on demand" or "not aviable" for Pro). -In car computer (they are the only one now exept some startup...).

not only, they have near monopole: -In ProGPU (outside of some Hyped event AMD are always "aviable on demand" or "not aviable" for Pro). -In car computer (they are the only one now exept some startup...).

There are multiple companies involved in car computing. Intel being a fairly large one especially now with their purchase of mobileye. Tesla has it's own hardware/software division and while they are currently using Nvidia hardware, they aren't using Nvidia's driving package - they can switch at any point. In fact Elon said that AMD was considered:

This is a Tesla developed neural net. Although it’s somewhat hardware independent, it can be; we can actually run this on Nvidia, AMD or Intel. We did pick the Nvidia titan GPU as the main chip for the neural net. But it was pretty tight call between particularly between AMD and Nvidia. But ultimately we felt that Nvidia had the better hardware.

Yeah there is a reason they are Number 1 in the market, they must be doing something right! Ever since I got my Gtx 580 I never once looked back at amd radeon nor do I plan to.

Little off-topic; Volvo auto-braking system! https://www.youtube.com/watch?v=_47utWAoupo

There are multiple companies involved in car computing. Intel being a fairly large one especially now with their purchase of mobileye. Tesla has it's own hardware/software division and while they are currently using Nvidia hardware, they aren't using Nvidia's driving package - they can switch at any point. In fact Elon said that AMD was considered:

Intel will drive Mobileye to the ground, as prophesized. I don't believe that they can touch anything that NVIDIA will do there. The only danger for NVIDIA in the car space is Jim Keller being the head of Tesla's hardware department, but it will take at least 3-4 years before anything comes to fruition there. AMD is a current zero in that market, which is very weird considering that the traditional strengths of their architecture have been in compute performance and their experience with custom chips. I hope that Ryzen and Vega do well and that they manage to invest more in the software ecosystem which is the only place they seem to lack.

Tesla has it's own hardware/software division and while they are currently using Nvidia hardware, they aren't using Nvidia's driving package - they can switch at any point. In fact Elon said that AMD was considered:

Tesla Motors and NVIDIA have partnered since the early development of the revolutionary Model S. Today, all Tesla vehicles—Model S, Model X, and the upcoming Model 3—will be equipped with an NVIDIA-powered on-board “supercomputer” that can provide full self-driving capability. ... This in-vehicle supercomputer is powered by the NVIDIA DRIVE™ PX 2 AI computing platform. It’s an end-to-end AI computing system that uses groundbreaking approaches in deep learning to perceive and understand the car’s surroundings. - See more at: http://www.nvidia.com/object/tesla-and-nvidia.html#sthash.upDKsteY.dpuf

http://www.nvidia.com/object/tesla-and-nvidia.html

http://www.nvidia.com/object/tesla-and-nvidia.html

They have a PX2 for inferencing and a Titan X for training, but they don't use Nvidia's software.

This is a Tesla developed neural net. Although it’s somewhat hardware independent, it can be - we can actually run this on Nvidia, AMD or Intel. We did pick the Nvidia titan GPU as the main chip for the neural net. But it was pretty tight call between particularly between AMD and Nvidia. But ultimately we felt that Nvidia had the better hardware.

https://electrek.co/2016/10/21/all-new-teslas-are-equipped-with-nvidias-new-drive-px-2-ai-platform-for-self-driving/

Pretty incredible results when you take into account that their core business market is shrinking. And that's exactly why Nvidia has been a runaway success compared to AMD: Diversification, platforming and support.

Did you not read the article? How is their "core business" market shrinking when datacenters, mobile devices (including the Switch), and automotive are all getting more attention? If you think PC gamers are their core market, you're very mistaken. As with Intel, their main source of profits are not everyday at-home consumers. The only additional profitable markets Nvidia has over AMD are machine learning, mobile, and automotive. Meanwhile, AMD has access to x86 and game consoles besides the Switch. Claiming Nvidia has better diversification and platforming may be true but an overstatement and not a reason for their runaway success. Nvidia's success has more to do with more polished products, faster and more distinct evolution of their products, better marketing, and a fanbase loyal enough to fund stupidly high and arbitrarily priced Titans.

Nvidia's management deserves high praise for their foresight and their willingness to adapt. Their Tegra program had a good run until around Tegra 3, after which they pretty much had no clients for it. Now instead of what a lot of companies might have done and either closed it and/or sold off the pieces, they turned it around to be their Automotive and Console push, and have reaped the rewards in return instead of the losses. Their diversification into the datacenter also proved to be a runaway success.

Do you not realize you contradicted yourself within the first 2 sentences? How can someone have good foresight when their product loses clients? But, what makes you think a product is losing clients when one of the main contributors to their boost in revenue was the Switch (which uses a Tegra chip)? Even excluding the Switch, I would argue the Tegra program is fantastic and exciting. I strongly approve of Nvidia's work put into it. If they were priced better, I'd be buying into their Jetson platforms ASAP.

And while they might have some of the best hardware around, what really drives their success is their focus on the software platforms surrounding their products.

Nvidia's software is excellent and may be attractive to consumers, but, it's also their downfall. Nvidia is extremely greedy with their IP, and as a result it greatly hinders their products. Just look at what happened to hardware-accelerated PhysX. With Vulkan released, people's interest in CUDA is also greatly threatened.

The Switch prints money(But since it uses flash carts it can only print coins)!

Yeah there is a reason they are Number 1 in the market, they must be doing something right! Ever since I got my Gtx 580 I never once looked back at amd radeon nor do I plan to.

Intel sold very well even when they had crappy P4s to sell, it doesn't mean the products are better. Heck elections are won with good PR too even though the candidate(s) can be crap.

Good for them, good leadership and foresight (I wonder how much of this is due to Jensen Huang). And the fact that they also pretty much dominate in what made them big to begin with (GPUs), is another achievement.

Told ya AI is the sh1t. Why is everyone still rallying behind VR and autonomous driving is beyond me.

Did you not read the article? How is their "core business" market shrinking when datacenters, mobile devices (including the Switch), and automotive are all getting more attention?

PC graphics is the core of the core, and its shrinking in units shipped (OEMs, low end graphics), but in high-end and gaming is growing.

Why is everyone still rallying behind VR and autonomous driving is beyond me.

Because VR is fun, and autonomous driving could immensely improve the future of humanity in several different ways.

PC graphics is the core of the core, and its shrinking in units shipped (OEMs, low end graphics), but in high-end and gaming is growing.

But the "core of the core" is not "the core". Intel's i3/i5/i7 models are the "core of the core" to their Xeon series, of which is "their core" market. It doesn't matter if Nvidia's shipped units are shrinking; that doesn't correlate to their success.

Told ya AI is the sh1t. Why is everyone still rallying behind VR and autonomous driving is beyond me. PC graphics is the core of the core, and its shrinking in units shipped (OEMs, low end graphics), but in high-end and gaming is growing.

Because I want to not drive my car and while the car is moving I want to be in a VR world with mah friendsssss.

Because I want to not drive my car and while the car is moving I want to be in a VR world with mah friendsssss.

GitS is coming true.

Because VR is fun, and autonomous driving could immensely improve the future of humanity in several different ways.

Or lead to 40%+ unemployment rates for an already shrinking middle class, without replacement in any other industry since AIs are effectively slaves and create the same economic problems they created in the Roman Empire and the American South. I'm also not sure if I'm looking forward to a future that 80% of humanity will be eventually redundant except for genetic material. This is far away now, but that's where machine learning will take us. I'm not a Luddite, but I can't see us as a species handling this properly, and this time (unlike the Roman and Civil War times), there are nukes around and idiots like Trump who wield them. I love the tech, but I honestly believe we'll f*ck this up as a species. As for Tesla, the only reason they went for NVIDIA GPUs was because the Titan XP was the fastest single GPU they could get for that workload. With the advent of Vega, that will most likely change. It would be interesting to see what they would have chosen if both top end cards would have been available.

Because VR is fun, and autonomous driving could immensely improve the future of humanity in several different ways.

OK I wasn't clear enough. Both VR and autonomous driving are worthy goals to pursue. I meant business/investment wise; these two catalysts are somewhat murky. Admittedly autonomous driving is a disruptive technology, and its good to be at the forefront,but the competition will be fierce. Cheap and affordable will win (FPGA, asic), and I dont think Nvdia will be interested in selling sand. VR... I have no idea why investors think VR is a huge deal. OTOH computing driven AI, aka deep learning, is a much more general technology (autonomous driving being just one implementation), that will shape the future. There is no doubt about that.