Future of gaming?

If you lookup input lag in the dictionary there is a picture of the kinect.

edit: Plus no real game application use with body tracking if you still need to watch the whole thing on a standard montior/tv.

 

I've been watching a few AI videos in this yt channel... and I'm not going to pretend I understand what's going on, but it looks very impressive. Gaming might see some improvements in terms of physics, fluid, character behavior, character animation that might really take a leap towards more realism and believable gameplays.

A few vids:

Spoiler

 

We have on the roadmap to try machine learning to estimate the waist of the VR player. Could be a fun POC

 

I really doubt they will waste their money like that. They mostly are not doing much with WMR anymore, and are focusing their efforts on enterprise AR via HoloLens.

Having a massive company like ID focus on a tiny userbase like VR is just a silly waste of an AAA studio, and thats why you are not going to see any proper VR games come from them or anyone else. Alyx was the nail in this coffin, if it was going to revitalise the indistry it would have, instead you saw a small bump. A bump on the graph, but a tiny miniscule one in relation to everything.

I get it, you guys have spent years on this thing and the tech is just not going anywhere despite what you keep telling yourselves, but sometimes its more sensible to actually deal with reality than to keep looking at biased graphs and saying "THIS YEAR, THIS WILL BE THE YEAR".

Everything you have said about the index and quest etc is just enforcing the fact that its not going anywhere.

If you have to buy a £900 headset and run that on a powerful PC to experience "true" VR, then its not going to go anywhere beyond a niche specialist thing for people who are super into it. Which is exactly where it is, and where it is going if you look at a variety of data instead of things coming from VR fansites.

VR is going to go the same as every other fad before it, and will once again re-emerge in another 10-20 years when the tech again has something new to offer. If it was going to make waves or be the future, its had 8 years to do that. 8 years...come on you dont really tell yourselves that is not enough time to get somewhere further than it has do you?

Now enterprise VR, such as AEC sims etc, thats another matter. Thats what you should focus on if you want to get the tech anywhere.

 

Bullshit, the numbers are climbing exponential since 2016.

edit: 8 years, you have no clue on the VR market I hear, the first commercial HMD was released in 2016 (And even Vive and Rift first gen were hardly consumer ready products).

edit: We just had EA (Triple A much?) release a VR game last month so thats bullshit too. And everyone is in consensus that the VR part of Star Wars Squadron is the way coolest part of that game

 

We have all been over this, its not expontential unless you are reading really biased graphs that show different things, and using that to make a wild assumption.

Look at the actual numbers, are you not reading them right or something? Even with the "growth" since 2016, if you compare to ANYTHING else, its doing very poorly.

The thing is though it doesnt matter what the numbers say because your team has spent the last couple years avoiding or disregarding anything that doesnt agree with your close minded bubble view of the industry - which is biased due to you making a VR game and ofcourse wanting to succeed.

We want you and the hardware to suceed too, but we also live in reality. Its niche, and you can cater to niche markets and make a profit, but you only will if you accept its niche and change your tactic to meet that. Building for a future that will never come and is based on the niche market becoming a major staple market (and you said it yourself, look at the numbers since 2016, the numbers not the line on the graph because if you measure the "growth" in real numbers its TINY, like laughably tiny for something that will become the future) is not the way to make a success in a niche market.

Again none of this matters ofcourse because you will say something about the index is the only way to experience VR, or some crap about the quest, or point to some unrelated graph showing very different things and assume trends based on that, or cite data from pro-VR media

 

it is interesting that you mention ID, which seems to be ID software?
Because John Carmack quit ID in 2014.
https://www.eurogamer.net/articles/2014-02-05-why-john-carmack-quit-id-software

And he was actually working on Oculus Rift.
The other interesting thing is that in 2019 he stepped down from his position of developin oculus in order to .... "work on AI".
https://www.theverge.com/2019/11/13/20963899/john-carmack-stepping-down-cto-of-oculus-work-on-ai

Makes one wonder.

I mean... he's human, but this guy is responsible for some technologies that had major impact on game industry.

 

Look to any other new hardware and it was slow too. Look at GPU introduction for example. My first 3D accelerated GPU was a card from S3 a year or two before 3DFX.. Even 3DFX was a niche product. It was first with the all in one GPUs that we saw mass adoption. This is nothing new. And VR is such a ground breaking change, much more than hardware accelerated GPUs was

 

Speaking of VR.... some time ago I found a mention of DIY headset on thingiverse. Not a "VR case", or "smartphone holder" or anything like.

But an actual headset. Then there are attempts like this:

Spoiler

Makes one wonder why is there not much of a DIY VR scene.

I mean, we live in interesting times, because you should be able to get any sort of part, order lenses, even have custom hardware made elsewhere. Then there's Raspberry PI, and FPGA boards. So it should be possible to jury rig a headset that will be "good enough".

Well, at least some people tried.
https://www.reddit.com/r/cyberpunkdiy/comments/fmpszx/diy_vrar_headset_pi_goggles/

If there was an open standard, there could be an evolving (opensource?) hardware which would eventually be mass-produced by somebody.

 

Back in the day before consumer HMDs there were alot going on over at mtbs3d.com both carmack and palmer lucky were hanging on that forum back then. It have died out a bit since consumer HMDs surfaced. HMDs are complex things, the optics more so. Its hard to compete with consumer products.

 

Despite what some people say, this isn't the case.

From what I've heard the Quest 2 is one of the best headsets going, and it requires no external hardware, and it's cheaper than a game console.

The enthusiast market who demands realistic graphics and has a beefy PC and wads of cash and so on is limited, and may well remain that way. But accessible entry level gear that's actually good is only just arriving, and that's where the action will be... assuming there is any. That will require something coming along, or multiple things, that provide clear value to non-enthusiasts aside from the novelty of using VR. It'll also require much better input systems than are current. Nice controllers won't do it, it has to be no (physical) controllers.

I'm sceptical, unconvinced either way. But I was the same when touch screen phones hit the shelves from a computer company at exorbitant prices, and look where that went.

 

I agree and hey as I said before I want it to suceed, but I am unconvinced (I say this as someone who has and does own multiple VR headsets and works on enterprise VR mostly to pay the bills).

That said, the quest 2 looks awesome - BUT the fact facebook is forcing people to use a facebook account and the issues that come from that are greatly holding back one of VRs best (if not the best) chances of going mainstream. The quest 1 really took off in relation to other headsets, and if they want to keep the momentum they should reconsider the move about facebook accounts.

 

It is partially true.

Quest can work in standalone mode, but you'll get, well, reduced visual quality compared to PC.
You can also use it on PC, via a USB 3 cable, or 3rd party software over wifi.

The biggest problem with it that in order to use Quest 2 you have to sell your soul to Satan, I mean Facebook. Basically, they REQUIRE Facebook account, and can randomly ban you for no reason, which will turn your headset into a paperweight. There's ton of horror stories about that on reddit. In this scenario most likely the problem won't be resolved, as the company apparenlty insists that their mysterious AI algorithms that issue bans are infallible, and therefore getting banned is always user's fault.

As somebody said - great hardware, horrible company behind it.

Hopefully someone will jailbreak it.

This is actually already a thing. You can operate quest without controllers, using your hands. I'm not a fan of the interface, because it wants me to point at things using my ARM and not my index finger, but home menus work without controllers, and there are games that use this hands withotu controllers. (Elixir is a nice demo of the concept)

Also, from my experience of showing quest to a relative, this device has potential to be popular, but not as a gaming hardware. As "experience" hardware. Because said relative was wowed by 360/180 movies of beaches and oceans, and reacted very weakly to, say, Gun Club VR.

So, basically, it has potential to be a replacement of in-house TV. Or something similar to it.

 

Hand tracking without an actual device to hold is pretty useless in most titles. You want something to grip. For example in our game the pistol grip of a firearm. We dont even use the finger tracking data in our game. That is a bit of a gimmick, it would destroy our careful handcrafted hand grips. But it could be used when hands are not gripping anything just for a visual gimmick nothing more.,

 

No, you don't, and that's why I want a glove controller.

I would prefer to use my hands and fingers instead of dealing with controllers. Because humans normally have 10 fingers, and controllers turn you into a pirate that has hook for each hand. And your hands have more degrees of freedom than controllers.

Now, it is true that many games it makes sense to have a "prop" to hold, but that prop does not have to be a controller.

Gripping something for a long time can be annoying or even tiring, by the way.

Also, try typing or playing piano while still in the helmet.

 

You can let go of the Index controllers if you want

 

The best would be gloves with motors that simulate the item you hold. But it would be clumsy with the tech we have today.

The index is the best we have on the consumer market today in this regard

 

Index is not what they are talking about I think, more like proper haptic gloves, which as far as I am aware are still a ways off. I do agree index controller is closest to this so far though

 

Yeah I know. But index is best right now in this regard, since you can let go of controller etc.

edit: The motor function to be able to resist your grip to simualte something in hand must be strong. will be bulky

edit: yupp. bulky

 

Also you need joystick to move about in the game until they have solved proper locomotion, which gloves does not have

 

ArmSwinger locomotion. Also see Creed.

 

Its imerssion breaking plus works bad in Shooters and similar type of games. Joystick locomotion is the best we have for many type of VR games.

 

It needs to get to the point where both the tech and standard UI designs are intuitive enough for a newbie to use, and without discomfort. I know it's coming along, but I don't think it's there yet.

In the land of phones I think that Apple getting early market dominance and insisting that people design their software with zero buttons* was a significant factor. At the moment vendors are competing, in part, on how good their controllers are, and I think that's a stop-gap at best. They'll probably always be relevant for specialist tasks, but they need to become the exception rather than the rule.

Yeah, I'm of two minds about this, too. But as for them wanting to keep momentum up, I'm sure they want exactly that, but I suspect it's primarily because that momentum will help tie people to their platform.

* Their devices have buttons, but they're for OS functionality.

 

Buttons are great, I hate for example the trend in cars with touch screens that's just dumb, sure most crucial systems can be controlled from the steering wheel but the times your need to use the touch controls you get mad.

Or universal remotes with touch screen, were you need to take your eyes off the movie to change something, stupid.

 

Unqualified statements aren't.

You're right, buttons are indeed super useful in many use cases. In other use cases they're not, and I was talking about a specific use case.

The above aside, I think this is deliberate. As the driver you're only meant to care about the driving part, so that's what's available on the wheel. The rest is a distraction, so optimising that stuff for ease of use while in motion would be make the car less safe, and thus be silly.

 

Writing these posts on my One plus 8 Pro makes me miss my old trusty HTZ Desire Z, best phone I ever had, loved that QWERTY

 

This assumes that people are actually going to pull over to operate these things. I wonder how many people just fumble around and take their eyes off the road instead.

 

That's my point. Making it easier to use would encourage them to interact with it more while driving, so they don't do that.

The ones I've seen have a full screen message at boot to the effect of "Drivers should not interact with this device while the car is in motion, passengers only. Otherwise, pull over to use it."

 

I don't have any statistics or anything, so I can just speculate, but I suspect that this probably doesn't really work, and that people are using those controls just as much regardless of if they are buttons or touch.

I'm pretty sure the reason why those controls are touch sensitive is because Touch Controls Are The Future, and not any safety reasons.

I'm not aware of anything else that was considered dangerous that was made safer by making it harder to use.

 

Oh, I don't mean I think they deliberately made it harder. I just think they would have made a deliberate choice to not optimise that part of the interface for the driver. Especially since the stuff that matters is already optimised by being on the wheel.

For passengers touch controls probably are better, because the software can re-configure the on-screen buttons based on context, which is awkward at best with physical buttons.

 

Joysticks have simlar problem. In VR shooters ideally you want a scheme where you move relative to orientation of your controller, because that allows you to strafe and look around freely.

This makes using two-handed (ranged) weapons problematic. And usually you'll end up one-handing your gun, even it is an anti-materiel rifle. Which feels incredibly weird, and definitely unconvincing.

IN a situation when only one hand ends up busy with a weapon anyway, using hand gestures to navigate won't be much of a difference. And in case of gloves, nothing stops you from having buttons on them.

 

Yes, it is bulky, but that's because it's an old model. Here is one from just two years ago which is much more inline with the actual gloves on the market.

https://vrscout.com/news/plexus-unveils-vr-ar-gloves/

 

.... a game that requires me to do all this to walk in direction I want, goes into trash can. Stock, stabilizing controllers against each other, and the rest.

 

I saw this one, its problem is that the price is "contact us".

I think that data glove should keep fingertips open so you could type while wearing it. You only need to track finger movement up to 2nd phalanx, because humans do not normally independently move 2nd and 3rd phalanx independently.

The glove you linked to, however, aims to produce tactile sensation, I think. Hence the covered fingertips.

 

How much force can it withstand? Can it simulate object in hand at all?

 

You don't need to todo it, you can hold your off hand free in air if you want.

 

Requiring certain controllers for specialist applications is one thing, and already limits things to enthusiasts. Macguyvering stuff together to make an application usable is quite another, and paints you into a very niche corner.

Fair enough if your audience is cool with it. But if you want things to go mainstream? Not happening.

 

Not specifically, though it was prompted by your suggestion to "find ways to stabilize your controllers against each other".

The point is that just because something works for enthusiasts doesn't mean that it's a viable mainstream solution. Even if it works well most people just aren't even going to bother to try it. They want stuff that they turn on and it just works.

 

You just need to find a way that works best for you, I found the wand a bit easier to stabalize against each other and then the main hand shaft can be used a a stock. Making it pretty much as stable as a stock accessory. It works ever which way, you can free float the off hand if you want and if it feels good for you.

But the inital argument that pad/stick movement doesn't work becasue you can't use your movement hand is invalid as shown above. Though some games do have it badly implemented so the movement vector is all over the place when you use the hand

 

No, he doesn't. When user has to experiment for basic movement,something is wrong. They should pick up the game and go.

Additionally.

... I am not a "guy above", I have a name, and my statement was in your response that "arm swinger is immersion breaking", then providing stick locomotion as an alternative and insisting that it is superior, especially "when coupled with 3rd party attachments or specific way of holding controllers", which is pretty much opposite of being more intuitive.

The reason why I even mentioned ArmSwinger, is because there are people who SWEAR by it, claiming that it is the most natural way of movement in VR. Not sure about that, but, for example dashing in Blade and Sorcery is done by swinging your arm's in running patterns, and it works great. Along with their jump detection. As you can jump there with your legs.

The problem with left stick movement is that you have to be aware which direction your hand is facing at all times, meaning it is more distracting than traditional gamepad/mouse movement on non-vr titles. Additionally, like I said, this control type heavily encourages movement where only one arm is occupied, because you need to control direction of your movement arm, or you'll be standing still. At this point replacing left stick with a hand gesture (data gloves) wouldn't make much of a differences.

And regarding immersion, as I said, movement via left stick does not mesh well with two handed ranged weapons, because then you'll end up using them with one hand, which looks ridiculous. As you'll end up one-handing semi auto shotgun, antimaterial rifles and all other things. Which is the definition of "non-immersive".

In general, it looks like the games so far hasn't really nailed rifle handling in VR in general. Gun play with a pistol is fairly well done in Gun Club. Archery apparently translates into VR very well, as even in Skyrim VR godawful implementation of it, you shoot one arrow and you immediately get it. For two-handed melee weapons, there's Blade and Sorcery which is probably a gold standard. Rifles, however are always awkward and feel weird.

 

And as long as the tech relies on everyone doing this, it's highly unlikely to take off.

 

Relying on arm swim locomotion is a sure way to kill any sales for a VR shooter.

We were the first VR shooter to implement stick locomotion and its still the only way to enjoy VR Shooters.

 

If you use our two handen wepoans with one hand the recoil is unmanageable. We have also talked about kicking the firearm out of the users hand but not implemented yet.

 

I could be wrong here, but the 'old gloves' (Dexmos) look like the bulk is the motors/cabling for force feedback, something those newer Plexus gloves dont/didnt have. The Plexus gloves are just haptic feedback on the fingertips.

Ive not used haptic or force-feedback gloves in a very long time (over 20 years), but Im curious as to how they've come on, because they actually felt.... bloody weird. FF in particular; you're 'holding' a weightless void the shape of an object instead of a solid representation of the object, and the haptic 'touch' system I remember was vibration based, which, again, just felt weird.
Neither were 'immersive'.

Are newer ones any better?

 

I asked above the poster but didn't get an answer

 

(opinion)
For a proper force feedback, you'll probably need an exoskeleton. Because extra weight on hands can be physically tiring so you'd want to support it with something.

While this can be certainly done, given the price tag of an industrial articulated robot arm, it is not something that will be accessible to everybody, and is not something worth pursuing.

 

Closed, because it outlived any value it.