Adds limitations on installation for DGX Spark #3602
Conversation
mpgussert
requested review from
Mayankm96,
jtigue-bdai and
kellyguo11
as code owners
github-actions
bot
added
the
documentation
| DGX Spark: details and limitations | ||
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ | ||
|
|
||
| The DGX spark is a stand-alone machine learning device with unique requirements and specifications due to it's specialized SBC nature. These details are... |
There was a problem hiding this comment.
Missing information. Please elaborate.
There was a problem hiding this comment.
I left this blank purposefully because I was told that we just needed the section, and not the information yet. I can write a general section for DGX spark, but I can't flesh out the details and limitations without either having some source information, or a working virtualization of the spark.
I'm sure we have a section for stull related to this somewhere! I just need to be pointed to it :)
|
|
||
| There are many ways to "tune" the performance of the simulation, but the way you choose largely defends on what you are trying to simulate. In general, the first place | ||
| you will want to look for performance gains is with the `physics engine <https://docs.omniverse.nvidia.com/kit/docs/omni_physics/107.3/dev_guide/guides.html>`_. This is because | ||
| specializing the engine to task will reduce the cost of something that must be called every newly rendered frame, so specializing the engine usually leads to large performance gains (in frame rate). |
There was a problem hiding this comment.
Complex weird sentence phrasing. Can you please simplify this? What does updating physics engine have to be rendered frame? Do you just want to say stepping the simulator in general?
There was a problem hiding this comment.
yes! I can rephrase this. What I want to express is that the physics engine is something that is invoked frequently within the sim, by definition. the best perf gains should come from optimizing the largest compute needs first, so the physics engine is a natural choice
Mayankm96
changed the title
mpgussert
requested review from
Toni-SM,
hhansen-bdai,
masoudmoghani,
ooctipus and
pascal-roth
as code owners
2 participants
No description provided.