Could we possibly get a video of the new face tracking using a webcam instead of a headset
Someone motivated might be able to run a test build of a world without EAC.
When I run a test build of my world VRChat doesn’t run EAC, and I only have access to the world I’m testing and avatars I’ve uploaded. Well, also the test avatars.
At some point I need to figure out how to record a video on the steam frame. I got used to recording short little videos on my quest 3s and throwing them into discord
Apparently offline mode does indeed work on the Frame but that doesn’t mean PC VRChat running natively on the Frame is a good idea, you’re still running on a mobile chip so it won’t take much for the entire thing to fall over.
I’d test it myself but well [“This is Where I’d Put My Trophy, If I Had One” meme goes here]
The face tracking update is really exciting, honestly. Honestly can’t wait to play with it when it comes out. Don’t worry about this other post, by the way. I’m glad we’ll be able to free up some parameters on our avatars.
Fun fact: EAC absolutely has an ARM version (surely we didn’t forget the Nintendo Switch or the Macbook exist) so really its down to whether they feel like adding suppor for it.
VRChat has previously stated they do not officially support the Steam Deck, and then someone quietly changed things around to make it work anyway, even despite its comically low install base. It still works to this day. It even has the exact same problems that were outlined here as reasons not to port the PC version to Frame.
Even if that’s one less issue (assuming there isn’t some VRChat specific technical problem that stops them from enabling it) it doesn’t invalidate the other reasons VRChat didn’t make a bespoke version.
The Steam Deck isn’t ARM but a full blown x86 APU, it also has a much larger install base* at somewhere around 4 million units: Three years later, the Steam Deck has dominated handheld PC gaming shipments | The Verge, so no I’d argue the Frame and Deck are not a fair comparison in the slightest.
*And yes I know it wasn’t this large at launch but the Deck wasn’t facing even close to the supply issues the Frame is now.
I’ve honestly been in favor of a dedicated blink and eye parameters for a while
isn’t it down to the Steam recommendation from their own docs that if people were to port the PC versions to use the android build or something?
Please I beg of you to reconsider your senseless blocking of the Steam Frame. You have effectively blocked me out of VRC because to you only running at 50 fps is not pleasant. Well I’d far prefer a slightly worse experience than no experience at all. The Android version of VRC is an unacceptable compromise.
On top of that, blocking a perfectly working game claiming that it’s because of the anti-cheat incompatibility despite the anti cheat working well on the deck or the machine needs more explanation because it sounds disingenuous.
So basically I have to stream VRC from my laptop with far worse performance because the Frame doesn’t have “enough performance”..?
absolutely awful decision on the steam frame using the android version
at least give us the option, I was so excited to be able to play the full version standalone even if i had to stomach slightly degraded performance
I think you are vastly overestimating the Frame’s capabilities.
All good reasons to find alternative ways to play social VR, at least ones that would enable the users to use whatever platform and create whatever they want. ![]()
I think we’ve gone so far down the deep end with this platform because now it’s all about “surviving as a business”. There is no way VRChat would fail at this point given how eager people are to develop for it.
We have SO MANY people, in droves, literally thousands (not even considering lurkers) wanting to create some of the most immersive and ground-breaking technology possible. Yet, so many of them can’t because in order to get special access to features or APIs, you need to be a trustworthy organization that can bring VRChat some tangible value that can be recorded to boost metrics.
The only reason that EAC matters in the first place is because VRChat’s core architecture requires a secure client. I won’t poke at the motivations for having a secure client instead of simply securing the server and letting the clients do whatever they want, but I digress.
I’m afraid that VRChat as a business may not be aligned with what the community actually needs. There is clear desire and VRChat could very, very easily justify it financially, but they just… don’t for some reason. People begging for actually useful features, and we do get them sometimes, but only when they directly align with some other business goal or revenue stream.
Could VRChat on the Steam Frame enable new developments or experiences, or even opportunities for people that they may not have gotten otherwise? Welp, sorry, that doesn’t align with staying in the black.
Also, we need to have authority over the client so… no custom clients for you, sorry. ![]()
Whether or not something is a good idea is subjective. Someone was playing Skyrim VR relatively fine on the Frame and it was absolutely playable. Plus there are a lot of people out there who are totally fine with dealing with issues or poor performance if it means being able to use it at all. ![]()
I think a lot of people actually underestimate how well VRChat can run, even if you are just in a simple world with optimized avatars. There’s a reason it can run at all on Quest while still maintaining a lot of core features.
I know we’ve had our disagreements before, but excluding that, why are you defending VRChat so hard? I know that you were selected by VRChat themselves to work on things like the Wiki, but you seem to be very… against the possibility entirely. You even falsely claimed that EAC doesn’t exist for ARM when it was a simple Google search.
We should not be defending a company for not implementing what its customers want because it can not be financially justified. If it weren’t for experimentation, tinkering, and silly smart people doing silly fun things with their hobbies, VRChat wouldn’t even exist in the first place.
VRChat received $96m in total funding with an $80m series D. Lets not forget that. They are not a small company by any means. But maybe that’s part of the problem. This company is no longer about the overall social VR experience, is it?
Why is it that plenty of other software, written by smaller organizations, or even just individual people, has builds for basically every piece of hardware on the planet, but VRChat can’t manage a single ARM build for Steam Frame, even if it permanently exists as a separate listing/branch so normal users can’t stumble upon it and complain about it breaking? They could even just have it exist as a separate build pipeline that they only trigger when they do official stable releases.
What even is this platform anymore that simply having a build for the latest VR device is such a heated topic? Why are we a supporting a platform that enables these conversations in the first place?
What happens if an avatar isn’t using some or all of those keys? Are the values for unused keys filled with zeros or something like that?
The reason I ask is that I was concerned that, if there were any unauthorized clients, other users’ actual facial expression data might be unintentionally read by them.
Of course, VRChat has EAC, but I don’t think we can technically say with certainty that there are absolutely no such malicious clients, so this point caught my attention.
Only thing missing in the FAQ and the current bindings documentation would be the Steam Frame controls/bindings.
I’m not really clear on how many steam frames are out in the wild, but two things of note
I don’t thing OSC works on the steam frame stand alone. Also only h.264 appears to work. I was hoping that av1 videos would play, or maybe vp9. I also tried h265 but I’m not going to ask for that
If your local avatar doesn’t support native face tracking, then UE face tracking keys are not sent across the network.
Thanks to the feedback here, each viseme will now be usable as a tracking key for Parameter Behaviors. This also means that the ability of a Parameter Behavior to forward its result to the animator enables per-viseme floats in the animator as well.
Oh yeah, just a silly question while this is still open:
What is this?
It appears it is no longer loading on the login page, at least for me, but it would be great to hear from VRChat on what exactly this was and why it was there.
And here’s the URL just incase, from presumably VRChat’s official CDN:
https://dtuitjyhwcl5y.cloudfront.net/4538ae084d847650-1672w.png
With a wayback machine backup:
https://web.archive.org/save/https://dtuitjyhwcl5y.cloudfront.net/4538ae084d847650-1672w.png
A painting that has a VRChat mic icon embedded into the art itself?


