0

I have quite impressed with this deployment kit. Instead of buying a new CUDA card, which might require new main board and etc, this card seems provide all in one.

At it's specs it says it has CUDA compute capability 3.2. AFAIK dynamic parallelism and more comes with cm_35, cuda compute capability 3.5. Does this card support Dynamic Parallelism and HyperQ features of Kepler architecture?

phoad
  • 1,801
  • 2
  • 20
  • 31
  • http://www.anandtech.com/show/7905/nvidia-announces-jetson-tk1-dev-board-adds-erista-to-tegra-roadmap - From a programming standpoint, K1 (and Jetson as a result) are Compute Capability 3.2, which is essentially the GK110 feature set minus the dynamic parallelism functionality, and including all of the integer, register, and thread improvements that differentiate GK110 from GK10x. On that note, it’s worth mentioning that while Jetson/K1 has a unified memory pool – phoad Jul 20 '14 at 07:21
  • Have you tried the `simpleHyperQ` and the `cdpSimplePrint` samples that come with the CUDA installation? What do they say? – Vitality Jul 20 '14 at 07:28
  • I was trying to decide either to buy this toolkit or not. So I could not be able to test them yet. Dynamic Parallelism was a must to have, because I would use it for scientific reasons to implement some algorithms depending on these features. – phoad Jul 20 '14 at 18:02

1 Answers1

3

Does this card support Dynamic Parallelism and HyperQ features of Kepler architecture?

No.

Jetson has compute capability 3.2. Dynamic parallelism requires compute capability 3.5 or higher. From the documentation:

Dynamic Parallelism is only supported by devices of compute capability 3.5 and higher.

Hyper-Q also requires cc 3.5 or greater. We can deduce this from careful study of the simpleHyperQ sample code, excerpted:

// HyperQ is available in devices of Compute Capability 3.5 and higher
if (deviceProp.major < 3 || (deviceProp.major == 3 && deviceProp.minor < 5))
Robert Crovella
  • 143,785
  • 11
  • 213
  • 257