Gltf load time difference

I tested the same model in a different computer environment and found the following results.
Why is 3060 faster than 8000?
Can you tell me?

WebGL2 Windows
RTX 3060 3~4 sec
RTX 1060 ti 6~7 sec
Quadro RTX 8000 10 sec

It totally depends on the model but the bigger the gpu the faster it uploads textures compile shaders uploads geometry and such :slight_smile:

1 Like

Thank you for your reply hahaha.
Then isn’t it normal if RTX8000 is faster than RTX3060?
But RTX 8000 10 seconds RTX 3-4 seconds. What’s the reason?

the 8000 is turing based and focus on compute and not gaming I guess and also not mainstream so I wonder how browsers would have optimized for it.

the 3060 on the other hand is a full consumer segments ampere arch with probably nice support from the browsers.

It might then not be unreasonable to believe the 3060 could do better in web. Also there could be bugs or unsupported features ?

1 Like