NVIDIA 16nm Pascal Surfaces - Drive PX 2 Mobile Supercomputer

Published by

Click here to post a comment for NVIDIA 16nm Pascal Surfaces - Drive PX 2 Mobile Supercomputer on our message forum
#5214834
https://forums.guru3d.com/data/avatars/m/232/232130.jpg
Nvidia drivers installed in your car? It's just asking for disaster. :banana: Car gonna drive itself to nearest milk farm...
#5214837
https://forums.guru3d.com/data/avatars/m/16/16662.jpg
Administrator
Waiting on the first G-Force pun here 😉
#5214839
https://forums.guru3d.com/data/avatars/m/244/244064.jpg
Car gonna drive itself to nearest milk farm...
You won the internet for today!
#5214853
https://forums.guru3d.com/data/avatars/m/243/243702.jpg
It is nice performance, but as always, nV intentionally blows it out of proportion for marketing purpose. It may be stronger than mentioned Titan X for given purpose. But at end of the day this is ASIC made for cars, it does not mean that this thing delivers Titan X rendering performance or even universal compute performance. But as HH mentions jokes: One horse can navigate and get you to many places. PX 2 eats 1/3 of horsepower (hp) and still gets you to crash.
#5214858
https://forums.guru3d.com/data/avatars/m/230/230424.jpg
Nvidia drivers installed in your car? It's just asking for disaster. :banana: Car gonna drive itself to nearest milk farm...
That is beautiful.
#5214861
https://forums.guru3d.com/data/avatars/m/134/134194.jpg
wonder if you can use them in sli for extra GeForce "sorry Hilbert couldn't resist"
#5214909
https://forums.guru3d.com/data/avatars/m/243/243702.jpg
wonder if you can use them in sli for extra GeForce "sorry Hilbert couldn't resist"
Sure you can, And it allows you to crash twice as fast. And before you do, you can enjoy stuttery speed control.
#5214923
https://forums.guru3d.com/data/avatars/m/80/80129.jpg
The tech they showed off yesterday with it is actually really cool, especially when they visualized it with the map in 3D.
#5214925
https://forums.guru3d.com/data/avatars/m/90/90667.jpg
you won the internet for today!
+100000
#5214946
https://forums.guru3d.com/data/avatars/m/243/243702.jpg
The tech they showed off yesterday with it is actually really cool, especially when they visualized it with the map in 3D.
It was not bad, but in their presentation system completely ignored "Zebra" on street as another warning sign. And big question is how that visual system reacts to vehicle which is painted with hardwood camo, or person in similar camouflage? BTW, we have this in place for few years.
#5214947
https://forums.guru3d.com/data/avatars/m/156/156133.jpg
Moderator
Why does it feel like Nvidia is going to become skynet...
#5214988
data/avatar/default/avatar30.webp
PhysX destruction just got real
#5214993
https://forums.guru3d.com/data/avatars/m/243/243702.jpg
Why does it feel like Nvidia is going to become skynet...
Do not worry nVidia is at back seat in skynet show:
Six years ago, IBM and our university partners embarked on a quest—to build a brain-inspired machine—that at the time appeared impossible. Today, in an article published in Science, we deliver on the DARPA SyNAPSE metric of a one million neuron brain-inspired processor. The chip consumes merely 70 milliwatts, and is capable of 46 billion synaptic operations per second, per watt–literally a synaptic supercomputer in your palm.
The architecture can solve a wide class of problems from vision, audition, and multi-sensory fusion, and has the potential to revolutionize the computer industry by integrating brain-like capability into devices where computation is constrained by power and speed. These systems can efficiently process high-dimensional, noisy sensory data in real time, while consuming orders of magnitude less power than conventional computer architectures.
#5215001
https://forums.guru3d.com/data/avatars/m/216/216490.jpg
Why does it feel like Nvidia is going to become skynet...
I'm pretty sure if that's the case, Google/Microsoft/Apple will be the first ones to it. One day... They will take over.. You all shall see. Probably we won't be alive by then but... One day... That's for sure. It just simple evolution logic. Many humans they would(many also have tried too) take over the world if it was possible and without or at least very few consequences. Maybe in 50 or 100 or 200+ years time. But it will happen. That or either I have watch iRobot/Terminator too many times.. 😀
#5215006
https://forums.guru3d.com/data/avatars/m/239/239175.jpg
lol, **** self-driving cars. I want 60FPS minimum on 4K at "ultra" settings with a single GPU :-P
#5215093
https://forums.guru3d.com/data/avatars/m/263/263710.jpg
I'm bothering about MOORE's LAW... Yesterday it was FinFet 14nm and here 16nm....
#5215101
data/avatar/default/avatar38.webp
Nvidia drivers installed in your car? It's just asking for disaster. :banana: Car gonna drive itself to nearest milk farm...
:D LOL, not bad
#5215123
data/avatar/default/avatar32.webp
The tech they showed off yesterday with it is actually really cool, especially when they visualized it with the map in 3D.
No one demo yesterday have been run on the PX2 as far as we know, it was all running on a TitanX .. http://www.anandtech.com/show/9903/nvidia-announces-drive-px-2-pascal-power-for-selfdriving-cars
On that note, while DRIVE PX 2 was the focus of NVIDIA’s presentation, it was GTX Titan X that was actually driving all of the real-time presentations. As far as I know we did not actually see any demos being powered by PX 2, and it’s unclear whether PX 2 is even ready for controlled demonstration at this time. NVIDIA mentions in their press release that the PX 2 will be available to early access partners in Q2, with general availability not occurring until Q4.
Nvidia will always surprise us.. now a new you metric : DL TOPS .
Curiously, NVIDIA also used the event to introduce a new unit of measurement – the Deep Learning Tera-Op, or DL TOPS – which at 24 is an unusual 3x higher than PX 2’s FP32 performance. Based on everything disclosed by NVIDIA about Pascal so far, we don’t have any reason to believe FP16 performance is more than 2x Pascal’s FP32 performance. So where the extra performance comes from is a mystery at the moment. NVIDIA quoted this and not FP16 FLOPS, so it may include a special case operation (ala the Fused Multiply-Add), or even including the performance of the Denver CPU cores
#5215129
https://forums.guru3d.com/data/avatars/m/80/80129.jpg
No one demo yesterday have been run on the PX2 as far as we know, it was all running on a TitanX .. http://www.anandtech.com/show/9903/nvidia-announces-drive-px-2-pascal-power-for-selfdriving-cars Nvidia will always surprise us.. now a new you metric : DL TOPS .
Well yeah I was speaking more to the comment about Nvidia driver crashing and DriveWorks in general. I don't expect any PASCAL based unit to be ready until mid 2016. If you look closely you can probably spot woodscrews in the PX2 demo unit. And honestly it wouldn't be an Nvidia conference without Nvidia making up it's own performance unit. They are the masters of deceptive marketing.
#5215132
https://forums.guru3d.com/data/avatars/m/216/216490.jpg
I like the comments but lets not forget that this will be specific product and with drivers coded/configured for specific purpose/usage. It wont need to be compatible with hundreds of different system configurations/components like current consumer products.

Next Page