it is now possible to build a state of the art gaming notebook with the most leading edge GeForce processors, and we want to deliver gaming experiences many times that of a console in 4K and have that be in a laptop that is 18mm thin. we plan our roadmap about five years out. It takes about three years to build a new generation, and we build multiple GPUs at the same time, and on top of that, there are some 5,000 engineers ... Nvidia's focus on a single architecture stands in stark contrast to Intel's approach of offering five (coincidence?) different solutions, such as CPUs, Xeon Phi, FPGAs, ASICs, and now GPUs, for parallel workloads.http://www.tomshardware.com/news/nvidia-intel-amd-gpu-koduri... I suspect many those hard core gamers use consoles because they have to, but would rather have a laptop.And if you can fit at kind of power in a laptop it becomes lots easier for a car too.Fro the above if a competitor just copies Nvidia, they will ben 3 to 5 years behind, and Nvideia will be on a whole new platform. Commoditization seems to be years away.
At some point revenue growth is going to slow down for NVDA as well as for ANET and SHOP et al. But that could be next quarter or it could be many years from now. Look at SHOP, $10 billion in revs and still growing at 25% yoy. Ali Baba, jeepers! Amazon, ditto jeepers!The whole key is finding companies with enough CAP and SAM to make it so.Intel is not going to take down NVDA in GPUs. NVDA has no plans to have 100% marketshare. NVDA is shooting for continuing dominance in the high end most profitable markets. Thus, that leaves room for Intel or AMD to try to move in to the other market tiers. So news of Intel and AMD will always be around. The only way Intel does anything material to NVDA is if Intel comes up with something disruptive. Intel can use its monopoly in CPUs to take the integrated GPU market that AMD already has marketshare in to begin with and try to move it upscale from there. Shall see how it goes.But discrete GPUs are still and will be where high end graphics will be, and as Mauser states, Nvidia's products today will not be the products of tomorrow. Always a moving target. AMD's newest GPUs are still behind Nvidia's last year GPUs in both performance and power consumption. AMD cannot keep up with each new cycle. Intel has proven they cannot produce a competitive GPU. Nvidia appears to have plenty of CAP and enormous runway on SAM that continues to grow.I am still anxious in regard to autonomous driving, but that will start to play out in the next two to three years. Data center is just starting a new upgrade cycle, and will be bigger than currently projected, just as the PC and data center markets are far larger than anticipated back in the 1980s and 1990s.Tinker
Isn't there a double desire for gamers ... absolute power, especially rendering, of course, but also display? I thought serious gamers wanted 3 or 4 20+" monitors as well ... hard to deliver in a laptop. Yes, you can connect a laptop to such displays, but if you have the displays, then a desktop is not a burden.
what's the next growth market for NVDA beyond GPUs if they have a 3-5 year roadmap (which seems to be the case)?
The AMD-Intel laptop chip is not a worry. It only has 24 cores, while the full Vega has 64. It's also clocked slower than the full Vega. So we are looking at somewhere around 25% performance of the full Vega (which is slower than the Nvidia 1080 Ti). This basically excludes it from most AAA games. It will have a niche in thin-and-light. But most hard-core gamers will desire performance over form factor.Going forward, Nvidia is launching their next gen Volta in early 2018. AMD has Navi on the schedule for 2H 2018 on a high-risk, high-reward 7nm process. Anything Raja helps built at Intel is 2-5 years away.
https://www.thestreet.com/story/14387976/1/nvidia-stock-upgr...Relatively bad news. Another analyst hold out has come on board the Nvidia train some 200%+ later. Underestimating data center demand. It takes a very ignorant person, particularly when it is one's job to analyze this stuff full time, to be that far off in regard to data center demand for AI.Nevertheless, I like to see still existing negative people out there because it is a pile of future investors who will change their tune and climb the wall of worry.This said, the guy still has no vision of just how large data center AI will grow to be. I am not sure how fast it will grow to get there, how many fits and stops there will be along the way, but it is not difficult to see that data center AI is going to be much larger than we currently envision and it is just starting.Tinker
Best Of |
Favorites & Replies |
Start a New Board |
My Fool |