Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

(Meta issue) Improve sample load times #1202

Open
SaschaWillems opened this issue Oct 21, 2024 · 3 comments
Open

(Meta issue) Improve sample load times #1202

SaschaWillems opened this issue Oct 21, 2024 · 3 comments
Assignees
Labels
framework This is relevant to the framework

Comments

@SaschaWillems
Copy link
Collaborator

Doing a full batch run in debug mode can take up to 20 minutes. This is especially cumbersome for maintainers, as doing a batch run is now a requirement when doing changes to the framework. One culprit are assets that store textures in ASTC format. This is only supported on mobile and on desktop (where samples are actually developed) those samples need to transcode those textures. This can take a long time, and for a batch run this is done for each and every sample that uses those assets.

We should improve load times of the samples, esp. the ones that use complex scenes.

Related:

@SaschaWillems SaschaWillems added the framework This is relevant to the framework label Oct 21, 2024
@SaschaWillems
Copy link
Collaborator Author

@tomadamatkinson
Copy link
Collaborator

@SaschaWillems step one would be to add non compressed versions of our current assets to the asset submodule

I think @JoseEmilio-ARM may be able to help with the scenes provided by ARM. The only other option would be to decompress the ASTC (lossy compressed, so would not advise this)

In terms of compressing assets per platform we could use a python script that compressed images (added to the assets repo) and then add an image resolver to the gltf loader that looks for compressed alternatives if they exist.

There may be better options but this seems inline with the shader work where we do as much offline as possible

@JoseEmilio-ARM JoseEmilio-ARM self-assigned this Dec 2, 2024
@JoseEmilio-ARM
Copy link
Collaborator

We have an internal solution to cache the decompressed textures locally, will share a PR for review

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
framework This is relevant to the framework
Projects
None yet
Development

No branches or pull requests

3 participants