Featured Articles

AMD SVP John Byrne named turnaround exec of the year

AMD SVP John Byrne named turnaround exec of the year

Director of AMD’s PR Chris Hook has tweeted and confirmed later in a conversation with Fudzilla that John Byrne, Senior Vice…

More...
Shield Tablet 8 launching on Tuesday July 22nd

Shield Tablet 8 launching on Tuesday July 22nd

We knew the date for a while but as of right now we can confirm that Nvidia’s new Shield Tablet 8,…

More...
AMD confirms 20nm in 2015

AMD confirms 20nm in 2015

Lisa Su, Senior Vice President and Chief Operating Officer, AMD, has confirmed what we told you back in May 2014 – …

More...
AMD reports loss, shares tumble

AMD reports loss, shares tumble

AMD’s debt load is causing huge problems for the chipmaker -- this quarter it had another substantial loss. The tame Apple Press…

More...
AMD A8-7600 Kaveri APU reviewed

AMD A8-7600 Kaveri APU reviewed

Today we'll take a closer look at AMD's A8-7600 APU Kaveri APU, more specifically we'll examine the GPU performance you can…

More...
Frontpage Slideshow | Copyright © 2006-2010 orks, a business unit of Nuevvo Webware Ltd.
Wednesday, 20 March 2013 14:35

Nvidia details Kayla, dev platform for CUDA on ARM

Written by Slobodan Simic



To get developers ready for Logan


In addition to GPU and Tegra roadmap updates Nvidia also showed and detailed its newest Kayla developer platform for CUDA on ARM, which should get developers ready for Logan SoCs scheduled for next year.

 

Built on a mini-ITX-like motherboard, the Kayla developement platform is based on a Tegra 3 CPU and connected via PCI-Express to a yet to be detailed dedicated GPU based on the Kepler design. The GPU has 2 SMXs for a total of 384 CUDA cores, according to a post over at PCGameshardware. The developer platform is solely made in order to give developers something to work with in case they want to see how well CUDA works on ARM, since Logan, with a Kepler-based GPU, will be the first CUDA-capable ARM SoC.

nvidia kayla 1

The Tegra 3 CPU part, paired up with 2GB of RAM, was most likely picked due to its PCI-Exprsss bus capability and should not reflect the actual performance of the Logan SoC, both performance and power consumption wise, considering the fact that the current Kalya board sucks around 10W. The "unnamed" GPU, paired up with 1GB of its own RAM, will be available in both MXM and PCI-Express form depedning on developers choice. It will bring all those neat CUDA features including OpenGL 4.3 features like tesselation, geometry shaders and compute shaders, all to be available on Logan as well since it is also based on the Kepler GPU.

The Kayla developer platform is certainly different from what we will see with Logan, at least on the specification and performance level, but should offer similar feature list, at least when CUDA is concerned.

nvidia tegraroadmap 1

Nvidia did not shed any light on specific details of Kayla, as it keeps all the juicy details for developers interested in it. It should be available in next few months and it would be quite interesting to see what some developers can do with it as it will certainly be a good indication on what to expect in Logan parts.

Thanks to the fact that Nvidia is currently quite keen on giving superhero names to its Tegra related products and Jen-Hsun note that Kayla is "Logan's girlfried" it was just a matter of time before it can be pin-pointed to Kayla Silverfox that appeared in X-Men Origins: Wolverine and the original X-Men Animated Series.


Last modified on Wednesday, 20 March 2013 16:17
blog comments powered by Disqus

 

Facebook activity

Latest Commented Articles

Recent Comments