musehq / spacesvr Goto Github PK

using the regress api from drei, would be cool to add functionality to automatically increase or screen reduce resolution based on the current fps

you would have to add a bit of dampening so it doesn't rapidly change, and definitely heir on the lower side before moving your way up

would look into this: https://github.com/pmndrs/drei#adaptivedpr

One thing it mentions is that directly setting dpr might be costly compared to using a pixellation postprocessing effect which would have the same effect but be faster to switch. might require a new rendering pipeline though, so maybe not the best idea

would be nice, i think it should work with wired controllers to desktops, and maybe game console browsers? not too sure tbh

https://developer.mozilla.org/en-US/docs/Web/API/Gamepad_API/Using_the_Gamepad_API

Add a component to livestream video. I see this going one of two ways

1. We enable livestreaming only per peer party, allowing us to use WebRTC but meaning that you can't broadcast to multiple parties on the same site

2. We use a service like https://mux.com/ to enable livestreaming for everyone

Create a camera tool to

• take pictures
• record video

then

• upload to social media
• save locally

biggest consideration is how it will work on mobile as it all needs to be controlled with swipes and taps

Hey Muse,

Getting this error when installing spacevr into the React-Three-Next starter

any idea why this is happening? please help

pretty straightforward. ~25 updates / second seems reasonable in the way of getting the position correct, but some smoothing would be nice lol.

recommend using
https://www.npmjs.com/package/@geckos.io/snapshot-interpolation

Testing spacesvr.io with Quest 2.

After right hand SnapTurn, left hand movment don't work correct?

For some reason yarn link doesn't work, probably has something to do with the build settings. Needs to be fixed for faster development.

The sandbox is broken due to a security issue relating to a change in 17 react-Dom to 18. How do i get it to work?

what would be the best to import the FacePlayer component? I am getting this error when importing as a named export

Collidable component isn't loading automatically, looks like you need to hard refresh it to get collisions loaded.

more specifically, we should not update the video texture every frame instead of pausing it since we might still want the audio.

the frustrum check can be done by calculating each vertex of the video plane and doing some fun math
https://stackoverflow.com/questions/41163030/determine-if-point-is-in-frustum

the stopping of the texture updates seems more sketch, we miiight need to hijack the video texture's update method

Been trying extensively to load a texture into a (ground type) model, but continue to get this error.

I am able to see the render on a latest three js version environment, but for some reason I continue to get this error within spacevr

would really appreciate any help!!

It looks like the new Tool Belt UI on mobile is to sensitive and opens when you tap on the screen.

If the world has a collider coming from a model, the user will often fall through the collider before it finishes loading.

The solution is to wait until the collider is loaded before enabling physics.

Pretty self explanatory

Will have to account for nipple and gyro controls

Add controls to lower or increase volume and mute / unmute.

Each of the actions should have a story that goes along with it, so mute/unmute should play an animation of your character putting on earplugs and the volume up and down .... something

think the proper way to implement this is through some settings page where users can enable/disable it for themselves rather than enabling it by default ..

It's already made, just need to add it on. come on now

Hey!

Been looking for ways to better present my component, but I can't seem to render any Drei 7.0.6 component inside the spacevr StandardEnvironment. This is the error I get:

Happy any code or environment if needed, but you should be able to recreate using the spacevr next js starter with:

"three": "^0.127.0", "spacesvr": "latest", "react": "^17.0.2", "@react-spring/web": "^9.2.1", "@react-three/cannon": "^4.9.0", "@react-three/drei": "7.0.6", "@react-three/fiber": "6.0.14",

I see this being implemented by getting relative distance covered via gyro motion and, when it hits a threshold, easing into the gyro controls

Everything is pretty out of date and causing some issues ... this all needs to be done in one pass, so, yeah

benefits include

• new 3js features
• probably some bug fixes
• the huge vr fps fix in one of the recent 3.js versions
• no more annoying package pinning

Have the onscreen 2d cursor complete some animation whenever the user looks at an object that has an onClick function

allow the user to pass in a color for the frame of the image, if framed

Allow the developer to specify how far away they want the player to be before they cant click their Interactable component

Started this branch a while back and never got to finishing it. Want to see where to take it.

https://github.com/spacesvr/spacesvr/tree/ui-ux
https://spacesvr-gjudb1xki-metaplug.vercel.app/

This would be a pretty big shift in how things work, so i want to consider pros and cons before even starting.

it's got

• a 3d rendered cursor (that can change in any way to present information such on hovering)
• no 2d pause menu, which is muuuch cleaner imo
• allows for click and drag, or double click for pointer lock. i actually think people might understand pointer lock better than click/drag since the latter requires one more action before moving the mouse for each movement, so this may not actually be great

unsolved problems are

• doesn't automatically handle first click for playing audio
• kinda ugly cursor at the moment, although i like how it kinda works like a compass lol
• extra keystroke for the user to learn to use our sites (T)
• mobile swiping to show/hide pause menu hasn't been figured out yet, although once it is i know it will be so cool

Allow for png images to be shown without a background on the image component

At first glance it looks like this will allow us to make a Collidable modifier for any mesh to allow for physics while also running much faster than the current Trimesh solution.

Not too sure whether this will replace @react-three/cannon or supplement it

Project link: https://github.com/gkjohnson/three-mesh-bvh

Check in with Steven, running iPhone X iOS 13.1.3, happens on both safari/chrome

.ktx2 images do not show up on mobile but they work ok on desktop. I've spent a decent amount of time trying to figure out why this happens. I talked with Don McCurdy about the issue here
https://discourse.threejs.org/t/ktx-texture-loading-with-useloader-r3f/26515/6
I'm not sure what else to link with regards to this problem b/c I have tried a lot of things with no success
This issue can be seen on the starter page on branch ia-ktx

There should be as little friction as possible on the end user, so auto-detecting keyboard layout would be ideal