Nvidia launches the Tegra X1, the first Teraflop mobile processor

Quote:

---

Maxwell architecture chip offers twice the performance of the Tegra K1.

---

Read more.

.... But can it play Crysis :-P

If coded for the chip, I suspect it probably could.

I was intending on getting an nvidia k1 tablet, I guess I'll be waiting now to see what they come out with next, also given the k1 had performance issues in places with valves games.

Update: 3 Days later I just ordered the Nvidia Shield 32Gb Tablet, I blame Hexus for putting the urge to buy it right back at the front of my mind ! Being more positive, at the very least it should be a decent upgrade over my HP Touchpad.

I do hate those comparisons to old supercomputers.....especially when talking about power usage....it throws the un-informed into a frothing state (which is why they do it).......while 2 chips may require 99%+ less power to run "than the first supercomputer", these days a difference of 20w can be a game changer for mobile devices.

I guess it wouldn't be so sensational if they compared against it's main competitor though....

Interesting. Gogoling nVidia tegra X1 got me here.

So this thing is for mobile devices and plays stuff in 4k at 60FPS and all that...?
Are they doing anything fancy for my desktop PC, though?

So they bought in A53 and A57 cores in a big.little configuration. What happened to their own ARM core, it should be coming along nicely by now...

Quote:

---

Originally Posted by DanceswithUnix

So they bought in A53 and A57 cores in a big.little configuration. What happened to their own ARM core, it should be coming along nicely by now...

---

It was so it could be brought to market quickly, apparently. I think their Denver core was their own architecture. It's a variation on the big.LITTLE configuration though. Annandtech states that:

Quote:

---

However, rather than a somewhat standard big.LITTLE configuration as one might expect, NVIDIA continues to use their own unique system. This includes a custom interconnect rather than ARM’s CCI-400, and cluster migration rather than global task scheduling which exposes all eight cores to userspace applications. It’s important to note that NVIDIA’s solution is cache coherent, so this system won't suffer from the power/performance penalties that one might expect given experience with previous SoCs that use cluster migration.

---

More here:

http://www.anandtech.com/show/8811/n...gra-x1-preview

Quote:

---

Originally Posted by phili_c51

.... But can it play Crysis :-P

---

The Atom Z3740 can - I suspect this can too.

https://www.youtube.com/watch?v=IeRXxmVxUX0

I don't get why they keep pumping these things out. I have never even seen a mainstream device that's using a K1 and now there's an X1, it just seems like its all about bragging rights. Is demand for their Tegra Chips so high that the industry needed this already? are the current ARM chips, Snapdragons, Exynos etc not keeping up? They should pump R&D dollars into producing better battery technology so these types of chips can be used for more than a couple of hours.

the nexus 9 uses the K1

Quote:

---

Originally Posted by Bloodbath

I don't get why they keep pumping these things out. I have never even seen a mainstream device that's using a K1 and now there's an X1, it just seems like its all about bragging rights. Is demand for their Tegra Chips so high that the industry needed this already? are the current ARM chips, Snapdragons, Exynos etc not keeping up? They should pump R&D dollars into producing better battery technology so these types of chips can be used for more than a couple of hours.

---

Missed in this article -

Quote:

---

Originally Posted by Ars Technica

Nvidia also announced the Nvidia Drive PX, an "auto-pilot car computer" powered by a pair of Tegra X1s. Used in conjunction with up to 12 separate HD cameras, a car with the Drive PX can build an "environment model" that it can use to "see" and "understand" its surroundings—it can supposedly detect other vehicles, pedestrians and cyclists, road signs, and other information. Using something Nvidia calls "deep learning," the system can be trained to detect partially obscured pedestrians, whether school buses or ambulances are flashing their lights, and other context-sensitive things that a real driver might see and respond to.

All of this data is then used to help the car automatically perform pathfinding and help drivers avoid collisions. Huang hopes that the PX can help even lower-end cars perform these kinds of tasks—most current driver assistance systems are confined to high-end vehicles.

---

IOW, it's not meant to be mainstream - at least not in any small consumer device type of mainstream.