Hacker News Reader

Comment by bane | Hacker News Reader

I'm starting to feel like Oculus is suffering from sequelitis. I have yet to use one, but by most accounts, it's in great shape for a v1.0 release. Launch early and they'll be sure to get good market advantage, and there's already a plethora of ready-to-run software.

But there's no release yet, specs and requirements keep getting bumped, and building out a metaverse of some kind seems to be eating up lots of resources.

It's not necessary to have super perfect ultra realistic scenes, people play minecraft and watch movies perfectly happily with their rifts after all.

Launch, then blow your install base away with a great 2.0 release.

View on HN

I think the opposite, they need to get this right the first time, or people will try it once, get a little dizzy or disconnected and decide "Oh VR makes me sick."

With the recent bumps in the specs, I think it shows that you need more power than most people have available to get that kind of performance.

This is exactly what Carmack said at Game Developers Conference. They don't want to "poison the well" by releasing before they're ready with games and experiences that make people ill.

I hear this argument a lot, but things changed a lot since the 90s. People's adaption to new tech has evolved.

Even if they fail now, when they make the product that is truly good, everyone will keep hearing that on the press and from friends.

Exactly this. Virtual Reality headsets for many will be an entirely new never before seen experience, if they get it wrong, people will definitely avoid trying it again and that will be bad not just for Oculus but the virtual reality industry as a whole. If Oculus are first to release, they have a lot of pressure on their shoulders to do it right because their v1 launch success will have a flow on effect to other devices.

The higher end specifications required to power Oculus will definitely be a stumbling block. Presumably the headset will cost more than the developer kits costs, then you need to throw in at least an additional $800 for a high-end graphics card to power the Oculus. This is the kind of money only hardcore gamers will warrant, not a small family with a few kids wanting to get them a Christmas present. I have always said Oculus need to partner with a graphics card company like AMD or Nvidia and give the headset away with a high-end graphics card purchase, or at least heavily subsidise it. They're killing two birds with one stone then.

I'm writing this from the standpoint of someone who did a small game project with the rift dev kit 2: It's good that they're taking their time.

When the DK2 was released, the drivers were a horrible mess. The hardware was kinda okay, if you didn't mind seeing the pixel grid. Blurring was better than with DK1, but still an issue. Lots of people still got motion sickness from just wearing the rift for a short amount time.

Now, don't get me wrong: I'm exited for VR. But I don't want a piece of so-so hardware that's just gonna lie around on my desk. I want something that's actually supported by software (games) I use (play) and fun to use.

It's useful to add that, generally speaking, when you get the motion sickness the effects can be intense. I found out the hard way that even if you've never suffered from motion sickness via regular vectors (car rides, boating), you're not immune.

I've only had one bad experience that resulted in motion sickness - during and following an extended play of Half Life 2. The feeling stuck with me for several hours, to the point where I couldn't even look at the TV screen from the couch.

It was a miserable experience; afterwards, I was careful to never ignore the signs, and stopped playing any demo that started to feel just a little 'off.'

Something like this could be a platform-killer and doom your project before it ever reaches critical mass.

Agreed. When I first got my DK2, I played for about 10 minutes despite feeling sick, and was really ill for like 48 hours. It was a horrible experience. Thereafter, I was very careful about giving up on demos that were glitchy rather than try to play through the glitches, and even more importantly, never ever strafing or moving in a way that is unnatural for the human body. With those two changes, I usually felt fine in VR.

I think the problem is that they have "blockers" that aren't quick to solve, and in the meantime other areas of the company are doing all of what you said.

Perceived blockers for a mainstream release:

* Wireless headset with feature parity of the current wired headset

* A large amount of VR content that doesn't require hacking configuration files or video drivers to work

* Minimal-to-no hardware upgrades for consumers after the purchase

I'm with you that I would rather see a release sooner than later, but I'm not a normal consumer. You aren't going to attract users if they need to spend more money or spend time changing their machines to make VR work.

I don't think there is any way around the hardware requirement if you want any kind of decent graphics experience. You need to calculate each scene twice (once for each eye), otherwise you're not going to get stereo vision.

And unless you want to view your minecraft world in 3D, that's going to result in hefty hardware requirements.

Definitely. The target specs[1] they announced are pretty beefy: NVIDIA GTX 970 / AMD 290 equivalent or greater, Intel i5-4590 equivalent or greater, and 8GB+ RAM. However, now that there is a target that means that tech will only continue to get cheaper/better from this point forward, so the longer a consumer waits to purchase the Rift the less likely it will be that they need to upgrade their hardware to use it.

1. https://www.oculus.com/blog/powering-the-rift/

> specs and requirements keep getting bumped

This is a (common) misperception based on hype. In fact, the release Rift hardware is more or less the Crescent Bay prototype they started demoing last September, and that was pretty close to Valve's "room demo" circa January of last year.

Oculus has been struggling with implementation problems and scaling, but the target spec for what counts as "consumer VR" hasn't changed much in years.

I'm also talking about the required specs of the machine that will be driving the goggles.

Sure, but there isn't much of a difference there. Again, in 2014 we knew we'd be looking at something like 2x1080p with 90hz vsync. Nobody would tell you you'd need something less than a high-end gaming PC to spit all those frames out.

Edit: The consumer confusion arose because Oculus released developer hardware with significantly lower requirements, and a lot of enthusiasts jumped on the bandwagon. But if you were developing for that hardware, you always knew it was insufficient for consumers, and that the release version would need a lot more muscle behind it.

They aren't just releasing a product though, they are quite literally inventing an industry. Care needs to be placed into the execution, this isn't an MVP type scenario.

> they are quite literally inventing an industry

But that's the thing, they aren't. VR in some form has been around for ages. It's already a multi-million dollar industry.

You're right about that - what I really mean is they have an opportunity larger than any other VR company there to put it in the hands of the consumer; in that sense they are creating a channel to an industry that was previously only for the hobbyist or geek.

They learned from history. How many people do you know refuse to go see a 3D movie? Mostly because they went to an older-generation 3D movie and got motion-sickness or headaches?

I know plenty - and I'm one of them. To this day I still won't give 3D movies a chance. If you want mass adoption of VR, it needs to be near perfect the first time around.

1) They recently announced a target launch and specs.

2) I bought a DK2 just to mess around with. It was definitely _NOT_ in great shape for a v1.0 release.

Granted, I'm not a gamer, and I was attempting to run things on a 15" rMBP with discrete graphics in Windows 7 via Bootcamp, but the software was super glitchy. It took a ton of time just getting things set up correctly to even run, and then, whether due to software bugs or inadequate hardware, I faced plenty of nausea inducing glitchiness. They are wise to take the time to smooth out the software as much as possible (albeit at a serious risk of letting numerous competitors catch up).

Reply on news.ycombinator.com