Open Beta

Make a UIText (TextMeshPRO works fine) and set the text in graph with nothing in the string and it will crash in the open beta client. This does not happen in the current live client.

Tested with Syn AckToon shader, with unity versions 2022.3.6f1 and 2022.3.22f1. Avatars uploaded with Unity version 2019 and below are unaffected. Steps to reproduce: Upload an avatar using aforementioned shader Open VRChat Observe how avatar's shading looks nothing like it does in unity.

The system default mic in the QM is blank.

Background Audio is a feature of the Quest OS that let's you play any kind of music in the background, while keeping all the game's sounds still intact. It's not a widely known feature but for those that do know how to enable Background Audio this bug is a pretty bummer. Not being able to listen to my own music (or podcast, or audio book) severely affects my VRChat experience. Please fix! Device: Quest 3 with Quest OS PTC v65. Steps to reproduce: Start VRChat (Quest Standalone) -> Press the Meta button on your right touch controller to open the Quest system menu -> Open Meta web browser -> go to Spotify, play some music -> Go to your system notifications, you should be able to see a notification there from the browser with the name of your song -> click on the three dots, press Playback -> Switch back to VRChat Expected Result: The music should continue to play in the background while playing VRChat with all the game sounds still intact. This works on the LIVE version of VRChat (build 1436). Actual Result: The music will fade out and stop playing entirely. It doesn't work anymore in Build 1444. Game sounds still continue to play.

Open Camera Disable "Trigger Takes Photos" in the Behavior tab. Press Trigger to take photo. Photo is taken despite setting being disabled.

join this world https://vrchat.com/home/world/wrld_f3fecd46-5697-4e68-8b6f-e938fd8f95ca Notice that the ambient occlusion is visible through walls (see screenshot) Switch back to the live version Issue disappears I contacted the creator of that world, it seems like the creator used the Mochie Standard shader.

Issue: Invite Photos are now also being taken with the normal camera, which is great! There are a few issues with how this is currently implemented though. The Camera does not tell you anywhere that you are taking an Invite Photo You can easily swap out of taking an invite Photo without ever noticing it Suggestion: The Camera should be out into some sort of "Invite Mode" when taking pictures for an Invite / Invite Request This mode should do 2 things: Tell the user that they are currently taking an Invite / Invite Request Photo with some sort of banner Prevent the user from changing the Camera Mode There should probably also be some sort of button to Exit this mode, so that if the user decides that they don't want to take an Invite Photo any more and just want to use the camera normally again they can. Attachments: Here is a video showcasing the issue: https://streamable.com/9ker8l

Issue: With the recent addition of Animated Emojis the Camera Mode sub menu also got a little bit of a rework. However this rework made it quite a bit more confusing in my opinion and could even be considered a bit overwhelming, especially for a new user. There are 2 main issues that I have with this redesigned menu: Inconsistency & Unpredictability: The sub menu appears to have 4 buttons in it, but they all work differently. - Photo & Emoji both open sub menus, while Multi Layer and Stream do not Photo & Emoji both have sub menus, but they are also different sub menus - Photo opens a sub menu for where you want a picture to be saved - Emoji opens a sub menu for how many frames an Emoji should have Every button behaves differently, even though from the start they all appear to look the same. Some open sub menus, while others don't. This results in the user not being able to predict what is going to happen when they click on any of these buttons. This is confusing, even for me as a long time user. Complexity: The Camera Mode sub menu has quite a few Layers to it compared to other Camera sub menus. Other Camera sub menus basically go 2 Layers deep, you first have the selection of the category and then the settings it self. (e.g. Mask > list of toggles ) The Camera sub menu however goes 3 Layers deep. As normally, you select the category and are then being provided with options, however a few of these options now also have sub menus in them. (e.g. Camera Mode > Photo > Save Location) Conclusion: Knowing all of that, the main issue I can see with the Camera Mode sub menu is that it just tires to fit too many options into one menu. This sub menu technically hides like 8 different buttons inside of it, 10 when you also count unfolding the Photo and Emoji sub menu as buttons. This is especially noticeable as some options are for what kind of image you want to take, while others are for where those are supposed to go / be saved to. Suggestion: Instead of trying to squeeze all of this into one sub menu, maybe consider separating them a bit? Move things into different sub menus and group them more together by what kind of setting they are. To prevent users form selecting incompatible options you could for example also grey them out / block them from being selected at the same time. Attachments: Here is a video showcasing the issue: https://streamable.com/ks5fsr

Since build 1441, which came with the Unity version 2022.3.22f1 upgrade and a mention of a potential "slight performance uplift", I have noticed performance to be slightly worse. When switching between live and open beta (1444), I consistently notice up to 3-10% lower FPS in more intensive worlds where my FPS doesn't get capped. For example, 75 fps vs. 67 fps when comparing in the world "Goodbye Is Not the End". Here are some more sample worlds to try: https://vrchat.com/home/world/wrld_cdb7147e-2bc2-40e1-a410-cbbecf0cc1d4 https://vrchat.com/home/world/wrld_8441122c-dc69-4243-97a7-8c73d06c48f8 https://vrchat.com/home/world/wrld_251da81b-4c64-48dd-903b-9f23df42dae7 I actually started testing this because I was crashing out of SteamVR very frequently starting from build 1441, when opening the photo or stream camera. I can't reproduce these crashes consistently right now, but I can reproduce the FPS drop very consistently.

When a texture is compressed using crunch compression, the new estimated memory usage field in the avatar stats shows the crunched size, not the actual size in memory after crunch decompression. This is not ideal, as it will encourage users to juice these numbers by crunch-compressing textures, when this doesn't actually help memory usage, and makes load hitching worse. As a concrete example, this avatar shows a texture memory of 10.67 MB, but an estimates memory usage of 0.48 MB. The avatar contains a single 4k all-white texture that has been crunch-compressed. https://vrchat.com/home/avatar/avtr_ef3447a0-5fea-4a7d-aa5d-d4d9a75aeb74