Sega VR Revived: Emulating an Unreleased Genesis Accessory | Hacker News Reader

Sega VR Revived: Emulating an Unreleased Genesis Accessory

gamehistory.org

261 pointsecliptik73 comments

This is insane, I had no idea Sega almost launched a VR headset in 1993... and for $200?!

I'm not surprised it didn't work out, but I never imagined we had the tech even for low-resolution tiny color LCD screens back then, or the "intertial measurement unit".

I mean, I lived through those years and now I kind of have to retroactively alter what I thought was possible...

The Game Gear came out in 1990 (in Japan). I assume they used similar screens for the headset.

The tracking is what surprised me the most. Watching the trade show videos in the article, it's amazing how good it was with what was likely just an accelerometer.

Accelerometer is wishful thinking. The consumer headset I had back in that era used a couple of potentiometers on the sides. A stick ran through them and clipped onto your shirt at the shoulder. Turning your head left and right rotated them in phase (yaw) our out of phase (tilt)

Wow, that is super interesting honestly. I'm sure that was cheap as dirt to make, and reactive enough to pull off the effect. But still awkward as anything I'd imagine.

Almost reminds me of the Data Glove, the prototype idea for the power glove controller for the NES, using fingers and tilt sensors as button inputs to the console.

The idea was prototyped in a few days by an engineer by attaching drinkings straws from your wrist to the end of your finger, shining a light on one end, and having a photo resistor to read it on the other end to see how far you've bent a finger. Super cheap low latency finger movement detection. It was later replaced with fiber optics for reliability sake.

NASA's first VR headset from 1985 https://airandspace.si.edu/collection-objects/headset-virtua... used screens cannibalised from Citizen portable LCD TVs bought from Radio Shack: https://youtu.be/NF2Hd69-Tf4?list=PL256A5BBCA4A0714D&t=450

If I remember correctly, everyone in the company was super excited...then they came out of the prototype room and immediately canceled it.

I think it might have made people sick.

You are right, the article mentions it.

However, a likely factor in the cancellation was feedback Sega received from the Stanford Research Institute, which warned of headaches, dizziness, and sickness, particularly in younger users and children. In an episode of Retro Gamer Podcast, former CEO of Sega of America Tom Kalinske confirmed these issues as major factors in the decision to abandon the project.

Twenty-seven years later, nothing has changed.

$200 in 1993 is $360 in 2020 [1]. A 64 GB Oculus Quest 2 is $299.

[1] https://www.usinflationcalculator.com/

A couple of years later, Virtuality got roped into nearly releasing the Jaguar VR headset (with Atari) which moved the bar significantly: https://m.youtube.com/watch?v=WTWyKNN_Lgg

Wow, I had no idea the very obscure Jaguar console almost did that.

Very random Jaguar fact that I love: The plastic mold for the console was later used for a wall mounted dentistry tool, which used the cartridge port in some way.

Pic: https://www.startpage.com/av/proxy-image?piurl=http%3A%2F%2F...

Then you really missed out on a treat the [1]Nintendo Virtualboy released in 1995. I had a lot of fun playing it sadly it never took off. [1] https://m.youtube.com/watch?v=Po5iEqDDv3U

I wonder what sort of IMU could that have been? Back then MEMS was not a thing, so how did that work? I am puzzled how they managed to target 200$

Actually, the first MEMS devices predate this by decades, though the first MEMS gyroscopes and accelerometers were actually developed around the same time as the Sega VR, though not available commercially. The article does link to the patent for the motion tracker used, and it's pretty odd (and not MEMS): https://patents.google.com/patent/US5526022A/en

To quote from the patent:

> The tilt sensor includes a transparent gas and a transparent viscous fluid in a spherical shell, a light emitting diode (LED) mounted at the top of the shell, and four photodetectors mounted at the bottom of the shell. As the tilt sensor is rotated the path of the light cone emanating from the LED and refracting at the gas/fluid interface is altered, thereby altering the intensity of light incident on the photodetectors. The magnitude of the light incident on the photodetectors is processed to provide the tilt angles. A thin transparent disk floats at the gas/fluid interface to damp surface waves caused by reorientation of the device.

This is crazy

Yes, both you and ansible are right, raised good points, and I learned something today. My perspective was more in line with the fact that about decade ago, MEMS sensors got really cheap, so that enabled a whole slew of things that were inaccessible previously. Before, you didn't really have affordable, small options for these things.

But TIL that the paths were prepared a long time in advance for this.

MEMS accelerometers were definitely a thing. But commercial, low-cost MEMS was not yet a thing at that time.

Motorola was rolling out their first product around that time, which was derived from military-grade hardware. That was for automotive use (airbag sensor).

One of the most interesting articles I've read on Hacker News lately. I think it is well written because someone like me with zero experience developing videogames or even VR software was able to follow most of it (or at least I think I followed it :-) )

It is incredible the amount of stuff he had to solve to be able to work at the end. I think that if it were me, I would have finished the project happily after I surpassed the first obstacle, calling it a day. Kudos to everyone involved on this, really nice write up!

I really want to see someone do something similar: Allow you to view emulated VirtualBoy games in a VR headset so you can experience the actual 3D effects

Or for another option, Virtual Boy emulated on a 3DS:

https://github.com/mrdanielps/r3Ddragon

Note also that you can use the 3DS emulator Citra with a head-mounted VR display, and apparently it's pretty great:

https://www.reddit.com/r/Citra/comments/c20fu7/citra_in_vr_i...

So for maximum nesting doll fun, it's likely you could run the Virtual Boy emulator under Citra, and view it all on your VR headset.

Sadly r3Ddragon looks quite dead (no updates in five years) and isn't very far along (commercial games don't run properly, speed is often lacking).

This works flawlessly on the Oculus Go and the Oculus Quest:

https://sidequestvr.com/app/125

https://github.com/braindx/vbjin-ovr

Really fascinating. Along with Nintendo's failed attempt at VR, I'm starting to think if VR is really ever going to take off? It seems history is repeating itself. We all get excited about VR, but it keeps failing to go beyond being a fun gadget. A nice to have. Any thoughts on this?

It needs a Pokemon Go. I just got a Quest 2. $299, Apple quality. Works by itself or can plug in to a PC and be powered by its GPU and CPU. It's waiting for new experiences with mass appeal. Valve gave a Half Life incarnation.. it's waiting for the right new content.

For creators, there's already amazing stuff like Tilt Brush.

Part of the issue is that VR is great with super high resolution displays and 120Hz low latency simulation response. The hardware is still catching up to the dream, but it's a lot closer. It reminds me of the dream of the 1984 Macintosh or the NeXT cube, or the 1993 Newton. Software graphics rasterization on 8 MHz 68k / 20mhz ARM cpus.

I've had the original Vive and the Index since each of them came out, though I didn't regularly use them much until I got into VRChat. In a month I've already put more hours into VRChat specifically than the total I've done with other things in VR. I'm not exactly sure of its mass appeal, but it's definitely been the killer app for me and my friends.

I've become somewhat convinced that the killer features of VR are in how it works for social games. The ability to socialize in 3d space with regular body language present (your head and hands are positionally tracked) is great. The sense of presence from being in an all-encompassing scene in VR makes your mind feel more involved in social situations compared to a standard voice call. Having positional audio and the ability to see where other people are looking makes it easier to follow the flow of conversations. Groups can smoothly reform and break apart into multiple separate conversations just as easily as they can in real life meatspace gatherings, which is something that's much less natural in voice calls.

I agree that social is going to be the killer app for all the reasons you listed. I think the only missing piece is face and eye tracking so those cartoon avatars can have some emotionally relatable cartoon faces.

It actually surprises me how expressive just head and hand tracking can be. Just watched a friend do his stand-up debut in VR and I stopped seeing the avatar fairly quickly. I just saw him performing.

I bought the quest out of curiosity with low expectations. I’ve tried a few of the games, many of which were novelties but have been blown away by Eleven Table Tennis. It replicates maybe 98% of what it feels like to play ping pong and I’ve convinced at least five other friends and family to get a set so we can catch up, chat, and play some ping pong. I play it almost daily now and use it instead of phone calls to catch up with my dad.

That's super interesting. Most sports require significant resistance, but ping pong is fairly unique in the ball being so light you really barely feel it. Plus whether you're holding a paddle or controller isn't super different.

I'd never thought of trying that game but now it makes PERFECT sense. Thanks!

In the Quest 2 "welcome" app they give you a paddle and a ball, and it feels really cool to throw the ball and whack it.

It also puts them on a virtual desk in front of you, and I was having so much fun I tried to rest my elbows on the desk. The immersion of VR is the most unique aspect of the platform.

I wonder though of this is the case. If we look at early PCs, they grew relatively fast. A big driver for that growth was digital spreadsheets.

VR has been around for a while. Why hasn’t anyone built the software that skyrockets VR to the masses. Is it because it has no real value beyond entertainment?

For example, for me, having a huge workspace with infinite amount of displays is the way to go. It would be a killer app. Its availabe today for Quest, but I still havent made the jump. The reason being is that we seem to get closer to Mixed Reality devices and THAT platform seems to have no-brainer killer apps. Including pokemon Go.

I use an old android phone + google cardboard headset when I do drone surveys or inspection. It has become indispensable, as the drone camera gimbal is slaved to my head. Much easier to inspect hard to reach stuff in (or on) the manufacturing facility I work at this way.

I'm always sort of impressed that I am able to have this functionality using (literally) the trash of others.

Once the resolution gets high enough that I can read small text comfortably, I plan on removing monitors from my office all together and migrating to some sort of AR terminal multiplexer. The general idea being that I can throw the headset on and wrap myself in some sort of window manager. I use a whole pile of monitors during daily work and despite needing them all during certain development projects, 50% of the time, I've got 2-3 displays sitting idle which is annoying. Having a 3 dimensional bubble of monitor space would fix this nicely for me.

I know this is what work spaces are for, but I've never been very proficient switching between them. I prefer to have all my current projects visible if I look around me.

Can you provide information on how you set that up? (the slaved gimbal) I'd love to recreate that!

Base station with DJI support called Litchi. Runs on Android. Has a nice VR mode that auto integrates head tracking. Turn on, switch to VR mode, slide into your pick of cardboard style devices and off you go. It isn't fast enough response for racing, but is great for inspection/maneuvering a drone around inside and out of a crowded facility (slowly) without line of sight.

I use an old samsung galaxy for the ground station/dispaly. Works excellently for the purpose and cost me nothing.

For work I use a DJI mavic pro with updated firmware to remove some pesky restrictions and the DJI phone home functions.

If you really want a roll your own solution, check out Ardupilot with a Navio II flight controller. This will let you construct a full flying telepresence solution to your personal specifications. Embedded Linux and ROS built in too!

https://dji.retroroms.info/

Anyone who plays Star Wars Squadrons in VR with a Flight Stick, finds it hard to go back to non VR gaming. The quality can be really excellent now, and I think we have moved from the "Tech Demo" stage to Early Nintendo Stage, where there were only 4 or 5 great games but they were all worthy of obsessing over and iterating on.

Beat Saber I think could sustain the Oculus Quest at this point on its own.

When Half Lyfe starts running natively on the Quest we will probably see another explosion.

And if they get Mind Craft running on Quest there will be millions of kids playing VR for hours every day.

I never bought into VR but just got a Quest 2 a few weeks ago and it's amazing.

Feels like it was made by Apple, the screen resolution is entirely decent, it's ready to use out of the box, no PC required.

I haven't even used it for games yet, but I've already been blown away by watching movies/TV in a virtual theater on it, using Google Street View (the Wander app), and exploring a handful of virtual environments.

I think when retina-level screens take off (~4-8x current resolution, so 1080p keeps its full glory), it's going to become the norm for watching TV/movies by yourself (instead of on your phone or laptop) and watching with friends remotely.

When on-device graphics compare with PC graphics cards it'll become the norm for gaming too.

As for productivity, when the resolution comes, the big remaining question will be lens quality. The further off-center you look, the blurrier it gets. But maybe?

I'm interested in the productivity side myself, however I haven't had an off center problem. If I move my eyes more than a few degrees out of line they start to hurt so I even move my head when reading your comment. I honestly didn't think about it much until you mentioned it and I actually gave it some thought.

One of the things that I really like is the head tracking in headsets. I don't really need to look at 6 monitors at once but I want them there in my peripherals so that I can easily glance back and forth.

The accessibility of Oculus Quest really seems to have pushed it forward a lot. I suspect the low price of v2 will result in a double impact between new people buying it and another group entirely buying the old Quest used (I got one for just under half its retail price recently, mindblowing you can get that kind of tech that cheaply now)

I think there's something in mapping the mental model of interacting with your environment to the new controls for that environment, which isn't quite finished yet.

Humans are good at making tools invisible or subconscious extensions of our physical selves, and there's this moat that VR gets stuck in I think on the way to true integrated physical extensions. i.e., a hammer becomes consciously indistinguishable from the carpenter's hand.

TV-based gaming consoles have had 4 decades of continually improving the screen and the input device. How invisible do the newest Playstation/Xbox/Switch (pro) controllers feel in your hands now?

VR can be immersive, but in the real world outside the goggles you're not yet transparently / invisibly integrated into the physical things you're near. So you kinda don't yet trust it, and some of the cables pull you out of the experience, and things are heavy etc.

When VR can live-integrate the world around you transparently into your virtual world, and the problems with things that pull you out of that subconscious human-tool melding are fixed, then I think there's going to be some really epochal things happening.

Until then AR seems to approach that integration moat from the other side and already looks to be making its way into everyday use.

It's called proprioception. It's your body's sense of its position relative to itself and to an extent the sense of position in your environment. It involves your inner ear, eyes, sounds, and even sensation of air moving over body parts

VR headsets gives your eyes visual cues that don't necessarily match that of your inner ear and body positioning. Lag between your inner ear's sensation and what your eyes see also causes problems.

Additionally your eyes aren't just a pair of fixed cameras mounts in your skull. They constantly scan (saccades) a scene and change their focus and that's that's then constructed by your visual cortex. The world doesn't have a depth of field effect applied to it, your eyes do that for you. Forcing both eyes to focus on a particular focal plane causes a lot of strain in because your eyes are not identical.

Besides human sensory input the lag and lack of precision of the game's sense of your body really fucks with your proprioception. Using a traditional controller uses small muscles with fast reaction times and little of your large motive and balance muscles. You can split your focus easily on the events and visuals of the game and work the controls.

With VR you have to make a lot larger movements and use more of your muscles for balance. Most people are not highly trained athletes their game avatars are and do not have that level of coordination or muscle endurance.

VR basically short circuits a ton of your body's senses or gives them conflicting input. Some people can adapt, many can't adapt for long, and some can't adapt at all.

Your comment makes me wonder how much variation people experience when trying VR. Possibly quite a bit. I get what you are saying about different cues that are missing, but for me it caused very mild nausea, some odd dreams and a bit of mild disassociation that passed after a few sessions. I wonder if there is some neuroplasticity factor that allows some people to more easily rewire to the new sensations and still feel natural.

I think the more serious attempt at VR from that period was Sega's R360 - https://en.wikipedia.org/wiki/R360

I think VR could spark a comeback for arcades with really expensive immersion equipment and infrastructure. Maybe they will be more like The Void.

This was one of the most cherished things from my childhood. They had one in the old Sega/Trocadero centre in Leicester Square, the Mecca of arcades in the UK or London at least. An attendent straps you in, the airplane takes off and you tilt back, and then woosh a dog fight commences with you sideways, upside down, facing the ground... Utterly incredible.

Edit: I never knew only around a hundred of these were produced. How lucky I was to have one near!

VR is like Linux on desktop. There are people making solid money on it, there is big group of enthusiasts, good quality product exists, yet it does not go mainstream. Personally, I'm much more bullish on AR and its productivity uses. Pushing Linux analogy further it would be VR's 'Android moment'.

I have a first release Oculus Rift and ended up not really using if after the first few weeks because it mostly just felt like a cool tech demo and not really ready for regular use.

A friend recently brought over his Valve Index and using it felt like it was actually approaching something I could see using more than to just show off a fun gadget (as you put it). Really the big thing is the screens finally felt good enough that I was able to occasionally forget what I was looking at was because I had two monitors strapped to my face.

That said, I think I'm probably going to wait one more generation before I spend the money on a new headset. Hopefully the improvements in gen3 will comparable to those between gen1 and gen2.

From what you've already seen from the Rift, I would probably wait for the model with built in eye-tracking, a new lens technology (varifocal?) and either stand-alone or fully wireless PC linked. 2 more generations or about 6 years? I hope we get there before companies lose interest.

I recently bought an Oculus Quest 2 as my first VR headset. I've been using it about 2 hours a day for the past month. It is absolutely beyond being a fun gadget. I am completely convinced that some form of VR/AR is the future of computing and in many ways that future is here now. A month ago I thought it was going to be a fun toy at best and I was completely wrong.

I like to think of products in terms of the ah-ha moments they can deliver, so here are mine:

  - The sense of presence is incredible. When you slip it on, you go to another place. When you take it off, you are surprised you are still at home. Every time.

  - The intuition of using your hands and body to interact with the world reminds me of the very first time I used an iPhone touchscreen. The best games let you pick up familiar objects and interact with them in expected ways without instructions or tutorials.

  - The amount of information that can be displayed. A web browser the size of a room. A YouTube video the size of an IMAX screen. I am very interested in Facebook's Infinite Office app they'll be releasing for productivity work.

  - Social experiences that are absolutely unlike talking on the phone or video chat and very close to hanging out in person. The social dynamics are also changed and people tend to be much kinder, less like they are screaming into the twitter void and much like they would be in person. I think this has the potential to save us from the social media death spiral and get us talking to each other again like decent human beings. I have witnessed several actual heartfelt apologies when people get offended and it gives me hope.

I think there are a couple of hurdles still:

  - The social experiences currently can work, but without eye tracking and facial expressions you need a good amount of distracting game or group activity to cover over the gap. I think the facial expressions could work on lower fidelity cartoon avatars just fine, sort of like Apple's AR emoji.

  - The screen quality is pretty much there and the rendering quality from the mobile GPU could use a boost but it's also serviceable. The biggest issue is that the fresnel lenses currently used have too much distortion and blur and cause eye strain and make the headset fidgety and uncomfortable.

  - The headset is heavy and awkward to wear and put on. It really needs to be closer to a pair of big sunglasses.

  - The apps and games are still in their infancy, though I think this is largely a factor of # of units sold and the low amount of development dollars invested. Facebook and Valve have been making good contributions here but it will take more before there is a self-sustaining software ecosystem.

So overall, I'd estimate we are 2-3 generations of the current hardware and one spot-on social experience app away from this hitting truly billion unit potential, but it is already so, so close. It feels like smartphones in 2005 or the internet in 1996.

I 100% agree, it changed me from thinking of VR as a dumb gimmick to starting to seriously consider changing my career focus towards VR

I remember getting the virtual boy when it came out and had such high hopes for the current batch of technology but it still makes me feel kinda ill. I personally won’t buy in because after ten minutes I need a several day break

Probably some niche markets worth capturing, but we’ve yet to see a killer use case. Maybe one exists, but it’s hard to suggest that strapping a screen to your face is something that everyone will want to use every day. I liken it to a roller coaster: they’re really fun, but also really intense, so I don’t want to go on one very often.

Rather than just downvoting me for an informed opinion, reply and let me know why you disagree. So far the evidence (sales, years on market) is on my side.

I'm quite bearish on VR as a mass-market product, and to get my point across I usually replace the term "virtual reality" with "Nintendo 3DS strapped to your face". I think you've kind of touched on my point well.

    wptr++;         /*  Skip number of sprites in frame */
    wptr++;         /*  Skip hotspot offset */

    wptr++;
    wptr++;

>Those last 2 pointer increments are skipping over data that goes completely unused, and it seems that at some point the data was removed from the ANM2FPA.EXE output.

Yea... I wouldn’t have found that so easily I don’t think. I don’t just increment pointers like that, and trying to determine intent of someone else’s code where there are just two random increments... this guy must be a lot sharper than me!

He has the source code, I guess these comments are from original source code.

Oh I get that. My comment was still how easily this could be missed even with source.

Maybe it’s just me, but I rarely use ptr++ to jump over like that. I’m more likely to use ptr[1] that isn’t “destructive” to the location you started with.

It saves you needing a separate index variable. If you're doing something like LZSS decompression it makes the relative window offsets natural (absolute). Not sure I'd recommend coding like that today, unless ptr was a smart pointer that had bounds checking, but for the time period it was a perfectly fine technique.

But the destructive nature of ++wptr or whatever can be to its advantage, in that you never need to update any numbers when things change. The data is consumed/produced in the order specified, a property that is maintained even after new operations are inserted or existing operations are reordered.

(Whether you want this or not, who can say! - though I claim you typically do.)

This is amazing! Can't believe Sega had a VR Headset in 1993 and for only $200, it's a bit sad they couldn't get it out there.

I believe VR is coming very close to going main stream, probably in less than 5 years from now. Oculus Quest 2 is IMO by far the best overall VR device ever, and things will only improve from here. That's why I've recently made a career change to work full time on VR application development, I truly think this is the next big computing platform.

Surprised to discover that the Ono-Sendai of Neuromancer fame was an actual cyberspace company.

I've always been so fascinated with products like this which are way before their time.

A VR headset in 1993 for $200 USD!?

Other similar products include: Apple Newton, Microsoft SPOT Watch, General Motors EV1, Kodak digital camera, etc.

My question is, which products are out right now, which are way too head of their time?

Great invention referenced:

https://patents.google.com/patent/US5526022A/en

I mean Sega Master System had those cool 3D glasses with a missile defense type 3D game.

Is it just me or did the website designer go WAY overboard on the banner size and it’s practically unreadable on mobile in horizontal?... Why do people do this? Do they not care?

Edit: lolokhn https://i.ibb.co/9VcJtCn/00-ACECC5-A03-F-4988-8286-99-E7650-...

"Please don't complain about website formatting, back-button breakage, and similar annoyances. They're too common to be interesting. Exception: when the author is present. Then friendly feedback might be helpful."

https://news.ycombinator.com/newsguidelines.html

If you really want to rage try it on a phone in landscape. A logo and a hamburger icon...and it's half the screen!

It distract me reading the text, on PC.

Try https://alisdair.mcdiarmid.org/kill-sticky-headers/.

Nah, simply created a uBlock rules, took few seconds.

I have a global custom stylesheet with #header, #site-header, #footer etc hidden by default. 99% of sites are improved.

I don't think that modifying that much the website will improve my productivity.

I do use a lot of uBlock list that remove annoyance, but made some important content disapear sometimes...

They probably don't care, since you can just rotate your phone 90°.

Reply on news.ycombinator.com