New Valve / Steam handheld console "Steam Deck"

someone tried to emulate the performances of the valve console

 

I hope it performs a little better than that.

 

If I'm not mistaken it should be up to one-third faster thanks to being the latest architecture.

 

That thing has no decent positional tracking, though. When playing alyx and the demo he has normal controller taped at the bottom of his palm. So it is not an equivalent.

Also, that's not quite the same thing as what I had in mind, although if you try to implement basic data glove, that's one way to go about it. There are also minor issues with the design, for example having to fight resistance of the springs might be tiring, over time it'll need adjustments, threads will stretch and so on.

Speaking of relativity helmet, I looked into this one, and it is basically half-done.
There are no controllers, and it cannot properly align itself in environment, like Oculus.

It could be improved further, but likely will need more powerful computing platform, like RPI, installed into it.

 

Wouldn't that just require adding infrared LEDs?

 

it would require infrared leds, solid chassis instead of potentiometers being sewn onto a cloth glove, bluetooth module, and a software in a helmet to run computer vision and talk to the glove.

I'm certain that Oculus helmet actually talks to the controllers, because IR led patterns constantly change.

Unfortunately, my phone and webcam come with a fairly decent IR filter, meaning I have hard time seeing the IR leds on oculus.

Regarding gloves, I think you'd want more solid design than just this. Something like knight glove or freddy kreuger glove. You'd also want 3 dof per finger, but that is likely tall request.

Anyway, I wasn't even describing a data glove in the first place .Just a controller that would be attached to your hands while keeping fingers available for typing and so on.

 

Two years back (now), at the SF Maker's Fair, there was a student project where they used a strip of capacitance switches/sensors (maybe 6-8 or so on each finger) set up kinda like that video where they ran down the top of the finger and on the back of the hand. Then software basically interpreted their distance from each other and translated it to not just the closure, but the knuckle angle. And since it measuring instead a single value on the string one, you bend your finger (like pressing on a surface) and the avatar would show that. It was really cool. They paired it with a leap controller for spatial coords.
These were middle school kids using python and arduino parts.

 

I know it can be done in more than one way and with relative ease as well.

However, in most cases it also won't be dirt cheap and won't talk to existing applications. Only to the custom software.

 

What are we defining as dirt cheap here?

 

$10 is dirt cheap.
$20 is maybe dirt cheap.

The project ZombieGorrila described could easily be in ballpark of $500, by the way.

Basically, it is possible to throw together something that will sort of act like sort of motion controller (and yes, there are cheaper part like chinese arduino which is something like $0.50 ), but it will be one o a kind thing and won't be recognized by a VR headset without additional software and that will make it mostly useless.

For the record I wasn't even describing a data glove, I only said that it woudl be nice if oculus controller was designed differently could be used while keeping your fingers free. Meaning you could type and track hand position and easily switch to "gamepad mode".

 

No remotely. The sensors are about about 60¢ - $1 for a 10 pack (depending on size/type). A controller is around $2-3 each. A little solder some wire and a usb cable and the rest is just time. (maybe some cheap gloves). So about $10 or less in hardware. Maybe a couple of bucks extra if you want to add rechargeable batteries. If you want to add a leap motion controller (instead of adding to existing controller) you can get those for $30-$50 used. $100 new. 1 for desktop/static, 2 if you want to get fancy. Low/med end: <$10. High end ~$210.

 

To be fair, when speaking about turning it into a commercial product then that time isn't free, and nor is the development of the software component which would also have to go with it. But there's no reason either of those have to push it anywhere near the realms of $500.

 

Totally, I mainly brought it up because it was similar to the string/potentiometer homemade deal. With cheap 3d printers, and dirt cheap electronic components (and electronic printing/printers) prototyping this stuff is super cheap. Also there is a ton pre-existing OSS software and libraries that make cobbling this stuff together pretty easy.

That said, a commercial product has to be tested and certainly made robust. (the string glove thing will never be a commercial product). There is definitely overhead (and brand markup). But basically it would contain the same amount of components as an Apple Magic Mouse which is ~$100 which include the apple tax. . (you could probably hack most everything you need for one glove out of a MM.)

I would highly recommend anyone interested in DIY electronics to go to a Maker Faire (when the world is safe for it again). The student and amateur hackers are AMAZING. Most provide specs and source code as well.

Also, bear in mind the complete Oculus Quest 2 system is $300. At best a commercial quality set of sensor gloves as an add on would be probably around 100-200. (200-400 if it were an Apple product)

 

How big is the market for such a glove at $100-$200? Is it sufficient to setup a proper production facility, or is it too niche (say sales in the thousands)? Which is a bit of a chicken & egg problem, you don't have a big market until you have a big way to utilize them, and you won't find the big way to utilize them until they're there. Until some big boy business sells it bundled to other things to bootstrap the market.

 

You need to decide whether you're talking about data glove or about alternative controllers.

Because while I'd love to have a data glove, I was originally talking about better way to attach oculus controlllers to hands and said nothing about finger tracking.

Regarding data gloves, I can't find anything cheaper than $250, and the price can go up to $5000. Normally the price is "contact us".

So if someone makes a dirt-cheap data gloves, there will be a market for it.

Lastly, regarding production facilities, currently CNC and 3d printing are fairly accessible, so it could be made into a kit or list of parts.

Honestly, I'm a bit puzzled as to why nobody has made an opensource clone of Quest 2. I mean, we have relativity project, but it stopped the development halfway, and got sidetracked into neural network estimation instead of, you know, making controllers or making the device capable of orienting itself while you wear it.

 

Diy inside out tracking is the main problem i gathered. But relativity is half way there, we need to attach a decent single board computer able to said inside out tracking.

 

I was a bit surprised, when I saw that handheld reveal, pop up on youtube. First thing that came to mind, when I saw the design was the nintendo switch.

Anyway, is Valve making alot of inhouse, first party games for their hand held? Because doing that would be a smart idea.

 

I wouldn't be surprised if they made Half Life 3 a Steam Deck exclusive.

 

"Half Life 3: New Steam Deck exclusive"
"Genre: Match 3".

 

Could it be a loss leader hardware issue? Like Sony and nintendo used to do, selling the hardware cheaper than cost to make money on the game store? Or maybe selling it at a loss to break open the VR market? It is pretty darn cheap. well, comparatively. I got a Vive several years ago to develop for and play with. But it was pretty obvious that the cost and complexity wasn't going to affect the market size. I got a quest 2 about 3 months ago to play some games with friends and was super impressed on cost and ease of use. It actually felt like a game changer for the market.

This steam deck kinda feels like the steam box and stuff like that. It doesn't feel like it is solving a problem anyone really is asking to be fixed. Tablets are great for... well.. tablet gaming. There are some games I play regularly on my iPad, none of which I wish I could play on my TV. (though I can through airplay). And the games I play on my TV (via steam and others) do I really wish I could play on a tablet. It just seems in a weird space. Usually I am a bit of hardware junkie, but this just seems... meh.

 

Devil's Advocate: Before iPads came along I'm not aware of anyone asking for a bigger iPhone to play games on. In fact, I remember quite a few people wondering what the heck anyone was going to do with an oversized iPhone that didn't fit in a pocket. Similar deal with iPhones themselves a few years before that.

In most cases I agree with the "who's asking for this?" perspective. Occasionally, though, someone makes a thing that people don't want until after it's a thing.

Of course, I have no idea if this will be that.

 

There's use cases for the Nintendo Switch, besides the exclusives sold for it. A lot of those apply to the Steam Deck as well.
There's use cases for gaming laptops, some of those apply to the Steam Deck as well.
There's use cases for general purpose computers attached to TV's, the Steam Deck can probably do a bunch of those as well.
There's use cases for smaller tablets, some of those apply to the Steam Deck as well (it having a touch screen)

If you have some intersection of these use cases that isn't yet fullfilled by some hardware you have, then a Steam Deck may be worthwhile for you. I think I have enough to warrant getting one (for now, who knows 5 months from now).

There's a lot of context that makes it more or less useful. Like; do you live alone or with kids/partner/parents/others? Do you have a small apartment or a large house? Do you rent or own your housing? Are you richer or poorer? Do you already own a lot of games on Steam or not? Do you like gaming on a TV+couch or prefer desktop+chair? Are you used to using controllers or keyboard-mouse? Do you already own a Switch/xbox/ps/fat-game-pc/gaming-laptop/whatever?
It's definitely not for everyone.

 

I believe people still don't want them after. It's just that the manufacturers need to justify their marketing tactics so they usually put anything worthwhile in the "flagship" oversized devices. Back in the days when Nexus split into "small" (4") and "large" (5") and "gigantic" (6") devices, the two smaller ones were more popular. For me, personally, the 4.9"-5.2" screens were the ideal size. And I think people who "invented" phablets and tablets should be deported to deserted islands and never ever should make contact with them again. I guess it's just me.

 

GridPAD 1900 was released in 1989, apparently.

Spoiler

----------
Speaking of SteamDeck....

One of the interesting points raised on other forums I saw was that valve does not seem to possess much of hardware manufacturing capability.

They built a small number of Valve Indexes (over 100k, vs millions of quests),and apparently after 100k preorders of steamdeck, the shipping estimate is at Q2 2022.

Meanwhile swtich sold 84 million units worldwide.

With this kind of production speed difference, the likely scenario that SteamDeck will remain a niche product forever.

 

Well, Facebook is using the device to harvest motion data from users and force people into making Facebook accounts, so they are definitely selling at loss.

However, I'd expect someone to be able to make a borg headset at double or triple that price from chinese parts.

Relativity is a decent example. Basically, to get it closer to the quest you'd need a raspberry pi or orange pi, and two arduino clone based controllers.

 

With regard to manufacturing, it's really common for a business to outsource that, and doing so would be drastically reducing the cost of pretty much anything which is made in those factories. If Apple and Steam and Sony and MS and so on all had to build their own separate factories in order to make their things then the cost of all of that would need to be covered in the price of the goods. Building factories and employing staff to run them is a rather expensive exercise. There is of course still a bunch of cost in re-tooling a factory to make each specific thing, but that's a fraction of making a whole new factory.

So the fact that Valve doesn't "possess" the capability themselves isn't a negative.

The Switch hit retail release over 4 years ago, and didn't do it during a silicon shortage. That's hardly "meanwhile".

 

My prediction is, just like the crappy controllers and other junk theyve tried, so shall this...

 

Very good points. Though if were to bet on which company had consumer prescience, steam would be pretty much last on my list given their track record. Apple has pretty much made a business model out of predicting marketable needs/desires.

In fairness, switch did well enough to show that maybe there is a market for something in between. Though like all things nintendo, it comes with a caveat... if you can stick zelda on it, people will by it. (by people... I mean me).

 

Wow. My first non-apple laptop was a grid (it actually looks a little like that). Ugly as could be but basically indestructible, it was a super dependable little machine.

 

Frankly, i would be interested if i had the monies. Having a distinct machine to play, because i let the working software up most of the time, like chrome, unity and all. And i don't want to manage resources. And i want to move around to relax. And i have a backlog of games for which it is overkill thus perfectly playable without performance problems. And also for those weirdo game that chsnge resolutions and windows freaks out my desktop organisation.

 

Well it's not just the Switch's success, there's already several other handheld PCs similar to the Steam Deck. The Aya Neo's Indiegogo raised $2.3 million, the OneXPlayer raised just over $2 million. And the GPD Win 3 raised $3.3 million, which is more than the original Oculus Rift Kickstarter's $2.4 million. So I'd say that's a pretty strong indicator of interest for handheld PCs that can run Steam.

Plus, with devices like these and the Steam Deck, you could stick in every Zelda game along with Mario and anything else on it via emulators (and Retroarch is coming to Steam soon). All the games you already own but Nintendo isn't porting to Switch any time soon.

The problem the Steam Deck and all these devices are solving is basically the holy grail many people have been wanting for ages--a single all-in-one gaming device where you can have your entire library in one place, and play it anytime, anywhere.

 

That's a laptop
/jk

 

Personally I'm interested in this. I already spend most of my day in front of my desktop making games. Playing games in the same place for several hours more after work is not that enjoyable even if it is nice to play on a 2k screen at 144hz.

I'm not into AAA games either these days. So Steam Deck sounds ideal for my indie gaming needs and I can bring it wherever. Change the scenery.

A lot of people here seem to miss the portability aspect of it. It's way more portable than 15 or 17 inch gaming laptops with huge ass power bricks. A lot of people also seem to ignore the accessibility aspect of it. It looks like they're striving for a console like experience on PC which will appeal to casuals plenty. This has real potential to expand the audience for us indie developers on Steam, so I'm excited to see what will happen next.

 

I wonder how kids perceive it. If I whip that out on a subway would that be seen as something cool or geeky? It's one thing to play on your phone but there are some cultural differences how socially acceptable it is to play games with these larger devices. Obviously you may not care what other people think. I don't. But it could have a big effect on popularity of these larger devices.

 

I think it will meet with exactly the same ridicule as ngage and other bulky things have. Also you're a magnet for mugging. And the weight!

 

So basically the same responses to lugging around and playing a gaming laptop?

 

We can go further!

Spoiler

The only thing this setup is missing is powered exoskeleton.

Spoiler

Because the only true way to play beat saber is wearing this thing on a subway.

 

With VR on a head, so don't see people near

 

Yes, because that’s totally the reaction people have to someone playing on a Nintendo Switch in public or an iPad Pro...

The Surface Pro 7 weighs more than the Steam Deck (1.7lbs vs 1.47lbs)

 

Nah, this is prime for selling off the back of a lorry. A laptop is much harder to shift, by far. Don't ask me how I know.

 

Yeah. That would definitely make alot of sense. I was thinking the same thing.

 

STEAM trying to go mobile... it brings back dark memories ...

Blizzard is still continuing their downward spiral.

I recognize STEAM is trying to make the computer mobile instead of shoving a beloved PC game franchise over to mobile phones. But who is this device for?

The 64GB STEAM Deck costs about twice as much as my gaming pc. Most games/software don't use more than 2 processors... still. I have an i3 from from 2013 and a nvidia card from 2017 with 16 GB dual channel ddr3 800mhz RAM. My hard drive is not SSD. I bought this system earlier this year.

My system is outdated and still plays resource heavy modern AAA games and GPU and RAM intensive indie games made by developers who have great game ideas and less than ideal game development skills. Unity is so easy to use people who don't know what they're doing can make games that take up way more resources than necessary. On the other hand put Unity in the hands of a team of good software engineers, or a single good software engineer, and AAA game companies become pointless.

My point is there's not a good reason to have the best system/hardware available to play modern games. I get 45-60+fps on current games with 720p resolution with my outdated dual core system.

So making a super ultra modern pc that takes up less space on your lap, and may come with heat issues, is unnecessary.

According to statistics, if this is measured accurately, over 55% of the gaming industry's revenue each year is made by mobile games. I don't believe that is from upfront purchases of mobile games but is from in-game purchases. If that's true. But according to the same website measuring this it claims in the USA PC gamers spend the most money out of all the gamer categories; in the USA, (from Statistica website). Maybe in other countries STEAM Deck will catch on but in the USA where gamers are Windows/Linux desktop users first, console gamers second, and mobile gamers when they're on the bus I don't see it happening. Especially since one person with dementia is doing everything to ruin the USA economy in time for Christmas 2021.

The best gaming experience is using a Windows/Linux desktop computer. This Deck is reminding me of lunch box computers.

The link below goes to a "lunch box" computer. It's an old computer with Core 2 T5600 1.83GHz 2GB RAM 80GB HDD that costs $2,000 simply because it's portable. What, no laptops existed in 2008? Is this proof of a parallel Earth?
https://www.ebay.com/itm/1839069322...WSqyya33j6QUoxA4acaONRmxOgEGZogLfMLVGE4IPFUf4

Modern "lunch box" computers start at almost $3,000. And that's with an Intel® Celeron® G5905 Processor in them!
https://www.bsicomputer.com/products/fieldgo-m9-1760

And last, but not least, students in 2017 remade the lunch box computer...
https://www.lunchbox.com/blogs/lunchbox-news/lunch-box-computer

It seems the idea of making a modern computer fit in a lunch box, like the STEAM Deck, is one of those ideas so dumb it appears to be revolutionary at first but once put in practice flops hard. The idea is sooo awful it's forgotten by the masses. Then when someone else comes up with it in the next generation they think it's new.

 

[etc]

The Nintendo Switch really is a ridiculous form factor. Whoever made that is truly dumb and wasting years of their life. Just use a phone. Or use a proper PC. Why a "Switch" that is too big to be mobile but too small to be static? No one will remember that thing.

Edit: (/s if it wasn't clear)

 

Mate your 'gaming PC' has less power than my office work station...

And your ramble aside, considering the pre-order numbers that were leaked, I don't think this is a piece of hardware that will be forgotten. Hell I'm still mad it's not available in Australia!

 

My gaming computer purchase was less than my monthly car payment.

Changed resolution to 1366x768, before at 720p

Empyrion: Galactic Survival - 36 fps in forests and combat.

Fallout 4 - 39-49fps in combat in city with Textures at high, godrays at medium, and view distance maxed for everything. Higher fps (edited because I can't spell) everywhere else.

Kingdoms - 39-50fps with view distances maxed, textures and shadows set to low.

Power supply is 250 watts

And of course building games with Unity.

Nice for around $200.

 

It is amusing when people with strong opinions suddenly arrive in a month old thread.

Fallout 4 is 6 years old and is not a modern game at this point, and less than 50 fps on a six years old game in combat situation where there aren't even that many enemies is not good at all, especially when you can't even reach full hd (If you're running on a FullHD monitor you'll be seeing scaling artifacts. ). You honestly need to upgrade. A modern game would be Far Cry 6 or Microsoft Flight Simulator 2020, and not Empyrion/Fallout/Kingdoms.

Additionally the information like "games do not use more than 2 processors" is not correct. You REALLY want at least a quad core on a modern OS, otherwise it will be running poorly.

$200 for the system you listed is a waste. You could pay $400-500 and get PS5 or you could pay $300, and get Quest 2 which would give you VR. Instead you bought a frankenstein PC at a very high price.

 

Why? They're clearly happy with it, and your suggested alternatives cost significantly more.

Things aren't all about numbers. If it does what they want to do then an upgrade is what would be a waste.

 

Have you ever tried to play assassin's creed 1 at 640x480 because at normal resolution it doesn't perform well? (at release time).

I spent a couple of years sitting on AMD FX 6300. Thought it was "fine" as well, however after switching to more modern Ryzen, I got a HUGE performance boost in everything my PC does. His situation is similar. The person is using outdated underpowered hardware, and likely the machine is slow like hell. He doesn't know. And he won't know what's missing until he upgrades. That's why I suggest to upgrade.

Given that the person has no problem wasting $200 on old hardware and is paying the same amount for car downpayment every month, the options I listed should be affordable and offer better gaming performance (PS5) or access to something his computer can't do at all (Oculus 2, VR). However a better option would be to upgrade. Mind you the point here is not to get a "bleeding edge" computer but a "normal" one.

 

Some people just really like paying as little as they can for computer hardware, and have little interest in playing games at a smooth 60+ FPS on the highest settings. My best friend among them.

Since I bought my beefy rig in 2016 (i7 5960X, 16GB DDR4 RAM (now 32GB), Nvidia 980ti (now a 1080ti), SSD, NVME, etc), he's bought a well thought-out, price-pointed PC three times... and his hardware is still very much leagues behind my older rig. But that said, he's still spent less on those three computers combined than I have on my one, very expensive 4K gaming machine, not even including the screens.

Different strokes for different folks I suppose.

 

It is not just games, though. The OS becomes much more responsive and reacts to everything instantly. You'll want a better CPU if you're dealing with unity, unreal, or IDEs in general.

Regarding very cheap hardware I recall trying to make a frankenstein computer years ago. I think the whole thing cost about $30, that included everything including display, except keyboard (used parts say hi). The frankenstein lasted 6 months, then blew a capacitor on motherboard (or something like that) and died.

Speaking of lunchbox PC, if games are not a factor, then something like Raspberry PI is a possible option. You can install linux on it, then it will be usable for basic document edition, word processing and maybe even a bit of image editing. The architecture is not X86, though.