Hello,
we are an ISV from Germany and our main product is a product configuration solution that is widely used in different commerce scenarios. Now we have written an Omniverse-connector that syncs the 3D-representation of the configured product to an Omniverse-stage.
In the next step we would like to implement a Kit-app that runs on the server and renders the scene, so that we could transport the live rendering to the client application. But - to be honest - at the moment we do not have a clue where to start. I think “kit-remote-render-server.bat” could ba a good starting point… but again… no clue how to start with this.
Can anyone provide some info about how to use that sample?
Thank you very much
Carl
This is when I start kit-remote-render-server.bat
This is what I get when I open the mentioned url:
Same when I open the url mentioned here ( Livestream WebSocket — Omniverse Create documentation (nvidia.com))
Then I tried to load a scene from command line like this…
…but as you can see, that does not work… can anyone help?
Hi Carl,
Thank you for reaching out and great to hear that you have a connector built and set up for your workflow!
The remote-render-server.bat is along the right lines but is something that is used and spawn internally by kit and create to do some deferred rendering but for a very specific use case so would probably not work for what you are trying to do but you are on the right path.
There is a rendering micro service embedded in the extension manager that you can enable and that you can drive via http requests.
If you enable that within the extension manager and then go to: http://localhost:8011/docs you should see the various endpoints enabled that you can then use to submit and render with from a remote server.
How familiar are you and the team with http apis/micro services?
I can send through some snippets on how to work with that API from python or similar?
Thanks,
Jozef
Hi Jozef,
thank you very much for your reply.
Just to be clear… we want realtime streaming (like the live-streamining extenion) that we can use in our html-based frontend. Does the extension you mentioned provide that, or does it only provide image rendering?
To be honest, I dont have any clue where to start, so yes some “snippets” would help. I think what we need is the following:
- A kit app, that is as small as possible (viewport, streaming, nothing more)
- A kit agent-setup that enabled out backend to start an instance of the “rendering-app” when a configuration session starts
Rely nice would be a sample how we could extend the functionality of the rendering-app, because late we need to give it some more features (for example hit-testing, and reporting the “clicked” object back to the configuration engine etc.)
Thanks you very much for your help
Carl
Hi Carl,
I see, I thought you were looking for offline rendering but it is more the streaming side of things.
We have a websocket streaming extension available as well.
You should be able to embed that into any app as a renderer and it should be available in the extension manager as well.
That extension comes with some simple javascript to get the stream and display it in a web web browser that could be adapted to fit into other applications as well.
Thanks,
Jozef
