VRChat - Easy Anti-Consumer

As many people may or may not know, I am incredibly passionate about virtual reality. It's a platform, a technology, and a concept that is so incredibly near and dear to my heart.

I've gone in-depth about why it's so special to me (and I highly recommend watching this before reading on), and the biggest reason I've fallen in love with it as hard as I have is because of a tiny piece of software, cobbled together in Unity with duct tape and a dream.

I am, predictably, talking about none other than... VRChat.

But over the past few days, a storm has been brewing within the game's community. A single update, one that almost feels entirely tone-deaf, that singlehandedly changed the course of VRC's life forever.

Before you read any further, I would like to preface this all with the fact that I wrote it at like, 5AM. Please pardon me if this all seems rambly or incoherent.

This post will outline my own frustrations with EAC, and how this negatively impacts the VRChat experience of myself and many others. But before we do that, here's a quick recap.

A Very Brief Introduction

Just for context, VRChat is a virtual reality application where users can interact with each other in a virtual space. They can upload avatars with all sorts of neat features, and they can create worlds where you can hang out, watch videos, play games, or sometimes just take in the sights. Although the fundamentals of VRChat are simple, I am not going to go any further in-depth about them despite the fact that they can, in fact, go much more in depth. The gist of it is that it's a social platform where you can make or do almost anything, and for most people, that simply involves chilling and hanging out with their friends. Got it? Cool.

So what's the drama?

On June 25, 2022, VRChat Team announced that there would be a Security Update releasing for VRChat. Although it did bring some welcome new features - for instance, an overhaul of the world invite system - it also announced that the game would officially require installation of Easy Anti-Cheat (EAC) in order to be played.

For many people, this is incredibly confusing. "How do you cheat in a social game?", asked countless people. The truth is, there isn't much cheating to be done. Instead, there are plenty of bad actors who abuse the VRChat system to wreak havoc on their victims. The VRChat Team claims that EAC will, in fact, negate these issues. However, what no one expected was just how much EAC would end up negatively impacting the end user instead.

For four years, VRC has been very popular among modders. Although client modifications were banned under the game's Terms of Service, it was widely agreed upon that you wouldn't catch any heat for using them as long as you approached them in good faith. This was even respected amongst VRChat Team themselves, who in 2021 sent out a banwave to mod developers, only to undo the bans an hour later and instead open conversation with the devs of these mods. Supposedly, these conversations were had to figure out why people modded, and to brainstorm ideas that people would have liked to see in the vanilla version of the client.

However, with the introduction of Easy Anti-Cheat to the game, "wholesome" modding has been rendered almost entirely impossible. Frustratingly, very few of the features that most people were modding for had been implemented into base game by this point, so the EAC announcement was rather blindsiding to everyone involved. Not only was VRC Team actively taking away mods that genuinely improved the experience, they gave no warning and no roadmap for when these features that were most frequently asked for would have a suitable replacement.

On top of preventing "wholesome" modding, EAC didn't seem to do much to actually negate the issues that people cared about the most. "Crashers", or avatars that have so much going on they crash the client of anyone who looks at them, were not affected by EAC, since the method for making a crasher avatar does not require modding of any kind (instead it involves exploiting Unity and the VRC SDK). Avatar ripping, where someone else steals an avatar and reuploads it for themselves, could still easily be done simply by extracting the local cache - once again, no mods required.

The backlash was immense. The most upvoted post of all time on the VRChat Feedback forums outlines many of the issues that EAC introduced to the game. The official Discord was completely bombarded by angry users, almost the entire VRChat subreddit turned against the VRC Team overnight, and VRChat was trending on Twitter for almost two days straight. Major gaming publications covered the story, even YouTubers like SomeOrdinaryGamers dedicating their time to it, and I even expect LTT's WAN Show to mention it too. Competitors like ChilloutVR saw an absolutely huge increase in users - CVR alone saw 10,000 people trying to sign up at once - and VRChat's player numbers and Steam reviews absolutely plummeted.

Needless to say, people simply were not happy about EAC. It caused significantly more problems than it solved - that was apparent to almost anyone. But why were people so passionate about these mods? What did we have to lose?

Why mod?

There were always plenty of legitimate reasons to mod your client. There were the obvious ones, like super-speed, flight, noclip, adding extra favorite slots for avatars and worlds. Some of them, like Lag-Free Screenshots, patched bugs that the VRC Team hadn't fixed yet. These were largely quality-of-life things that many people had asked for, but the VRC Team never delivered upon.

Some of these mods were performance-oriented. VRChat, due to its user-generated nature, is inherently unoptimized, so mods that helped increase FPS were very useful. The two most notable were the OpenVR FSR mod, which implemented AMD's FidelityFX Super Resolution into the game which significantly boosted FPS without a drastic decrease in visual quality. There was also CoreLimiter, which boosted performance on AMD processors by limiting the game to one CCX. And AvatarHider, which dynamically hid specific avatars that may impact the game's performance. Anecdotally, I've noticed my own framerate drop about 30-40 FPS since EAC was implemented and I've lost my mods.

Some mods were basic quality of life features that frankly should have been in the game in the first place. The ability to use the menu while laying down, displaying users' pronouns under their nameplates, auto-hiding the HUD when not in use - things that made basic things just slightly less infuriating.

But perhaps the worst losses were those that made the game significantly more playable to those who may not have been able to in the first place. Plenty of mods for hard-of-hearing/deaf players, or movement-impared players, were simply lost with no suitable replacements. A mod that automatically subtitled videos that were playing in a world, a mod that let you use TTS to communicate. Or, even, one that let you customize the voice falloff radius, so you could focus on conversations better. All of these mods - accessibility features! - rendered inoperable, and rendering their users in a significantly disadvantaged state.

These were all mods that people used, and they liked. There were plenty more - here's every mod I used with an explanation of each - but it just goes to show that there were so many things that people stood to lose if VRC Team went through with implementing EAC.

Other EAC drawbacks

Outside of blocking mods - the most obvious issue - there were a few others. Notably, Linux support was completely borked - the game, once working perfectly fine in Proton, seemed to not work through SteamVR, although this has supposedly maybe been patched. Virtual machines are now also unable to play the game, meaning macOS users playing through Parallels or Linux users playing through KVM are also unable to connect to VRChat.

Virtual Desktop has issues streaming the game, since it cannot properly hook into VRC without triggering EAC. Some integrations with OSC, a feature explicitly supported by VRC Team, were blocked by EAC. And, for some users, EAC also took a pretty noticable hit in performance overhead - a double whammy for those who depended on FPS-boosting mods just to play the game.

But with all of these issues, how did the VRChat team respond?

The response

After the initial announcement on Monday, almost everyone who publicly represented VRChat went radio silent. The post admitted that the team anticipated backlash for the decision - however, most believe they didn't anticipate THAT much backlash. As such, the negative feedback and frustration from the community kept rising, without any acknowledgement from VRC whatsoever about the responses they were receiving.

On Tuesday, they sent out an announcement with the patch notes, where they acknowledged the vast amount of feedback they had received. However, much to the dismay of almost everyone, one of the very first things said in that announcement was that they "do not have plans or intent to revert or roll it back". Naturally, this made even more people upset, as despite all of the negative press and feedback they had still decided to continue with their plans. They did, however, commit to improved transparency, and claimed to have rearranged their priorities to accomodate for the immense backlash. The next day, a blog post was released outlining some of the features that people wanted, and then then on Thursday, a developer update showcased some of the features that should be added soon.

These features include, but are not limited to, "Horizon Adjust" (a way to more comfortably play the game laying down without it messing with your tracking), noise gate and noise reduction, screenshot lag fixes, TTS, colorblind modes, and better portal handling. All of these features were, at one point, in the game via mods, but it does seem that the team listened to some feedback, as these were very frequently requested features.

The team also committed to providing frequent updates on how things were progressing moving forward - starting off daily for now, but slowing down over time.

My thoughts on what's next

I, personally, have always been an advocate for allowing mods. I used quite a few during my modding career, so to speak, and they were all incredibly useful to have. I appreciate the updated transparency from the VRC team on all of this, but I worry that it's too little to late on that front.

VRChat has always been about community. About the users, and the people that play it and create for it. A social platform is nothing without people, and for a very long time, it was easy to see VRChat Team as nothing but that - people. But, almost overnight, the perception shifted. VRChat was no longer a small game built by a community, for a community. No, no. VRChat was now a faceless corporation, operating in best interest for profits.

Cynicism ran wild with many, myself included. One of the most prevalent theories is that this was all timed with the release of the HBO documentary We Met In Virtual Reality, in an attempt to clean up the game's image before a huge influx of newer, inexperienced users flooded the servers. One other theory was that it was to protect future profits, as modders could simply add features that the VRC team could charge for, or alternatively block things like in-game ads, although this is much more speculative.

It does feel like a lot of trust has been lost, however. VRChat is trying to save face - very few people believe in the reasons they gave for adding in EAC, their reasoning sounding more like retroactive justification for wanting to eliminate modding. The crashers haven't gone away - in fact, they've increased! - and although avatar ripping is a little bit harder now, it's still fairly straightforward. The increased transparency is nice, but it really feels like this should have been there all along, instead of feeling like damage control after committing to a widely-disliked update.

One thing that struck out to me was that, according to Tupper, EAC was ready to go as early as December 2021. You'd think that would have been plenty of time to add basic QOL features like a pronouns tab or a menu that works while laying down, but I guess not. That part feels a little weird, but hopefully this means it will no longer take the VRC team three months to add a single button to the UI anymore. Seriously, their review process for new features was WAY too long.

I'm still inherently incredibly frustrated by the decisions made by VRChat. Their absolute tone-deaf response of "we're not rolling it back, sorry" feels completely disregarding of the community that made the game popular in the first place, and on top of this I generally dislike having to deal with invasive anti-cheat software - ESPECIALLY in a game that is not competitive. At all. I am happy that they did decide to improve their communication with players for the sake of transparency, but utlimately it's going to be incredibly hard to undo what's already been done.

Personally, I have cancelled my VRChat Plus subscription and instead reallocated the money to ChilloutVR's Patreon. CVR is a platform that is incredibly easy to port to from VRC, and it seems that it's the go-to backup plan for other VRChat players. At the time of writing, the servers are still on fire and inconsistent at best, so I've yet to truly experience it, but I still cannot wait to see it develop and mature even more, especially considering they've gained 500+ patrons in the past few days alone. As for me and VRChat...well, it's a bit more complicated.

I cannot commit to abandoning it yet. Ultimately, it's still a platform I want to love. I wouldn't be so upset if it wasn't, I suppose. I want to see the VRC Team actually listen to player feedback, to let the community grow and be weird. I don't like the way it's headed - what feels like an ultimately corporate future - as it's starting to stray from the early wild-west of the Internet that it currently emulates, but I do know that until then, I'll continue to watch what the devs do pretty closely. On top of that, I'm still keeping my eyes on ChilloutVR too, because I am all for competition in the market. Even if I end up sticking with VRChat as my go-to platform, I have no intentions on rescinding my pledge to CVR. Unfortunately, the network effect is a very real thing, which makes it incredibly difficult to completely abandon ship on a dime like that. All of my friends are still there, and ultimately, that's what I care about the most. That's what makes VR so special to me, after all.

Thanks for reading this ramble, if you made it all the way down. Sorry if it got long or messy. I'm very tired, and am getting ready to go to bed as soon as I finish writing this. I hope to see all of you soon, whether it's within VR, or over on Twitter at @KODA_VT.

Until next time.

Amber out.