Live data from Hacker News

Nvidia Unveils First Mobile Supercomputer for Embedded Systems

nvidianews.nvidia.com

11–20 of 40 posts

Re: Nvidia Unveils First Mobile Supercomputer for Embedded Systems

#14

I am surprised that no hackernews users has responded to this news yet. It sounds an exciting platform for all kinds of innovations.

The people who were interested probably bought the AMD-powered Gizmo Sphere already? (www.gizmosphere.org)

Well, this one has about ~6x the computing power (if their numbers are to be believed), and it has an ARM core rather than x86. Some differences at the least.

Re: Nvidia Unveils First Mobile Supercomputer for Embedded Systems

#15
post #9

Anybody know if the FPGA's branching model has been improved?

What FPGA? Do you mean GPU? If so, then - it's improved a bit over the years, but it's still much different than CPU programming, and branching remains relatively expensive.

Re: Nvidia Unveils First Mobile Supercomputer for Embedded Systems

#16
this is very cool. It operates in the "sub-10 watt space", had pretty decent io options for a device of that size, no apparent moving parts, decent graphics capabilities, and a reasonable price point. Seems like a great little device for a media machine or a graphics installation; somewhere between a raspberry pi and a mac mini.

Re: Nvidia Unveils First Mobile Supercomputer for Embedded Systems

#17
post #9

Anybody know if the FPGA's branching model has been improved?

What FPGA? Do you mean GPU? If so, then - it's improved a bit over the years, but it's still much different than CPU programming, and branching remains relatively expensive.

I have my terms mixed up. I thought GPUs used FPGAs, but I was wrong. Thanks for clearing that up.

Re: Nvidia Unveils First Mobile Supercomputer for Embedded Systems

#18

Could be a great platform for WPA2 key cracking and encryption bruteforcing when paired with tools like Pyrit or HashCat

I wonder how many pmks it could get through per second in Pyrit. I imagine it's a bit more power efficient than a bunch of GPUs going through pmks.
Post reply on HN