RIP Kinect

Status
Not open for further replies.
HoloLens costs $4k ... It is not a mass consumer product. It's basically a device for developers at this point.
I'm not sure if you know but Google glass was also not a mass consumer product @ $1500. Also hololens launched at $3k and still is https://www.microsoft.com/en-us/hololens/buy
but thats all moot as the point 'kinect is a success cause it gave us hololens' is a bit of a logical fallacy, going by that you could say WW2 was great cause it got the USA to the moon in 69, which most likely is true. :oops:
I could make a valid argument that hololens, windows mixed reality etc would be much further advanced today without kinect, if instead of pumping the 10s of millions they spent developing kinect into developing hololens/WMR, Do you disagree with this?
 
but thats all moot as the point 'kinect is a success cause it gave us hololens' is a bit of a logical fallacy...
Kinect gave us affordable 3D depth cameras powering Hololens's inside-out tracking. You claimed it was one of the biggest tech failures ever, even more than Kin and Nokia. How can one of the most significant advances in computer vision be considered a tech failure?
I could make a valid argument that hololens, windows mixed reality etc would be much further advanced today without kinect, if instead of pumping the 10s of millions they spent developing kinect into developing hololens/WMR, Do you disagree with this?
I think everyone would, because the money spent on Kinect fed directly into those other projects. It funded the development of 3D cameras without which Hololens as envisioned wouldn't exist. Without Kinect, pretty much the same money would have had to be spent. The tech also wouldn't have had the massive public 'beta test' that was clearly needed. Imagine Hololens launching with 3D cameras version 1, as per Kinect 360, and all the tracking problems that has. Through Kinect 360 and Kinect One, MS has the experience to make a far more robust and reliable camera-based tracking solution. And don't forget the huge sales and profits from Kinect 360 which helped fund further R&D. The 10s of millions saved as you suggest would have many many more tens of millions profits lost.

Kinect 360 was landmark tech that was a world-record breaking device making loads of money == not a failure
Kinect One was a significant improvement on the tech, preparing it for mainstream computer vision. As a tech it wasn't a failure because it worked, though as a product it failed because there was little reason to own one and it affected One's design, sales, and everything that goes with that.

In no way whatsoever is Kinect comparable to the biggest tech flops of history, as you argue. How can you equate Kinect 360's gangbuster sales and profits to the DOA Kin phone??
 
I'm not sure if you know but Google glass was also not a mass consumer product @ $1500. Also hololens launched at $3k and still is https://www.microsoft.com/en-us/hololens/buy
but thats all moot as the point 'kinect is a success cause it gave us hololens' is a bit of a logical fallacy, going by that you could say WW2 was great cause it got the USA to the moon in 69, which most likely is true. :oops:
I could make a valid argument that hololens, windows mixed reality etc would be much further advanced today without kinect, if instead of pumping the 10s of millions they spent developing kinect into developing hololens/WMR, Do you disagree with this?

There's absolutely no way you can know that augmented reality would be further along if Kinect did not exist. I also don't think you can say Kinect gave us hololens, but Kinect is a part of hololens, so the tech continues at Microsoft even if in a different form.

I do think it's pretty easy to say that Kinect led to a lot of research in the space of 3d tracking, both by microsoft, the people who hacked it into robotics etc and competitors that saw the market for tracking growing.
 
Last edited:
Oh my god. B3D is still in the phase that the massive and brutal failure of Kinect is everyones fault but ‘Kinects’ and people are just to stupid to know, and only the 3 die hard fans left know its full potential (stupid MS research devision...go ask here at B3D and make a thread: “Haaaalp, 10 visionary things B3D can teach us about Kinect utilisation we never thought about”)

And this argument about ‘stupid people don’t know how to use Kinect in a visionary way’ is pretty much not worthy for B3D honestly:

You can bring this argument for every single tec failure in the past ever created.

Kinect wasn't a tech failure. The tech was incredible. So you fall at the first hurdle.

And it's not "people's" job to know what the tech can achieve, it's "people's" job to spend their hard earned money on the things that seem worthwhile. What MS thought they could use the tech to deliver was insanely mundane and worthless and almost entirely unrelated to gaming - quite insane until you remember they had forgotten they were actually selling a games console.

Plenty of people had ideas about how to use Kinect in new and interesting ways, but MS wasn't interested in supporting them. So you fall again at the final hurdle.
 
Latency and inaccuracy kills. I suppose iphone x front sensor will be the most widely used kinect like sensor in near future.
 
Latency and inaccuracy kills. I suppose iphone x front sensor will be the most widely used kinect like sensor in near future.
Very Kinect-like. A chunk of the original Kinect technology came from Israeli 3D company PrineSense who were bought by Apple a few years back.
 
Very Kinect-like. A chunk of the original Kinect technology came from Israeli 3D company PrineSense who were bought by Apple a few years back.

The camera tech that was used. But the hardware wasn't what made Kinect revolutionary in the world of motion tracking. It was the software and API stack.

Once Microsoft showed that it was possible to do fairly accurate real time motion and object tracking in 3D with relatively inexpensive tech for the consumer market, many other companies got very serious about it very quickly.

Regards,
SB
 
The camera tech that was used. But the hardware wasn't what made Kinect revolutionary in the world of motion tracking. It was the software and API stack.

With respect, that's nonsense. We use Kinect sensors in our 3D scanning lab and Microsoft's commitment to developing the APIs faded out pretty damn quickly and not a single bug has been fixed in the last 3 years and there are plenty. The sensors aren't reliable on modern Intel chipsets with semi-random disconnect/reconnects and much of the functionality that Microsoft supports in their APIs is way behind commercially-licensed algorithms that work on a wider variety of equipment. The really telling thing is the lack of Microsoft patents; they're offering their implementation on published research, much of which was undertaken during the development of MRI scanners.

In terms of the keystone Kinect features, it was realtime 3D scanning and that is entirely sensor hardware tehy Microsoft licensed wholesale from PrimeSense and their R&D is built on decades of published information. Near-realtime 3D scanning has been a goal of radar and sonar designers since the mid-1960s and vast amounts of papers have been published on this. Former radar and avionics engineer here. PrimeSense replaced radar/sonar input with I/R .

Kinect is cool, but isn't some Microsoft technology revolution. Like Apple with iPhone, Microsoft packaged up a bunch of [then] new technologies and built a decent product out of it. Unlike Apple (but very like Microsoft) they then left it to die.
 
I thought only the Xbox 360 Kinect used hardware tech from PrimeSense and the Kinect One used different.
 
I thought only the Xbox 360 Kinect used hardware tech from PrimeSense and the Kinect One used different.

PrimeSense pioneered 3D mapping based on I/R sensors and were later bought by Apple - before Xbox One/Kinect 2 was publicly a thing. Subsequent to this a fair few companies got into 3D I/R technology working around the few patents that PrimeSense had, just as happened withe Apple and "multi-touch". Any licensing agreements that Microsoft had with PrimeSense prior to Apple's acquisition would have had to been honoured subsequent to the sale and by the time Kinect 2 was a thing, 3D I/R sensors were more widely available. Just like multi-touch.
 
Ah, that slightly explains all the issues reviewers are having with iPhone X "Face ID", since it's using PrimeSense (Kinect X360) tech.

Just going by my own consumer usage impression of the Kinect X360 and Kinect One, the Kinect One is significantly improved.
 
Ah, that slightly explains all the issues reviewers are having with iPhone X "Face ID", since it's using PrimeSense (Kinect X360) tech.

I've learned the hard way that it's safer to buy in on the second generation of any significant new product range. Especially Apple. :yep2:
 
Kinect for Xbox One is a time of flight camera, based on tech from a company called Canesta. Canesta was purchased by Microsoft and is probably the basis of Hololens as well.
This. World's first TOF camera. MS brought it to market. If they hadn't backed it through Kinect, would anyone else have? Amazing tech without an application at the time. It needed someone to support the tech to get people imagining what could be done with it. Now it's proven, it'll find its way wherever.
 
This. World's first TOF camera. MS brought it to market. If they hadn't backed it through Kinect, would anyone else have? Amazing tech without an application at the time. It needed someone to support the tech to get people imagining what could be done with it. Now it's proven, it'll find its way wherever.

Not the first, but the first mass consumer.
 
In no way whatsoever is Kinect comparable to the biggest tech flops of history, as you argue. How can you equate Kinect 360's gangbuster sales and profits to the DOA Kin phone??
The reason its a bigger flop than Kin, is cause MS tried Kin it didnt work out and then abandoned it, they did not repeat the same mistakes and bring out a Kin 2, they actually learnt their lesson.
This is exactly what MS done with Kinect 2 with the Xbox One, like I said in a previous post Kinect 1 was worthwhile, as it was them experimenting which is often worth a gamble, True they never released a single compelling/must game that used the tech. but OK.
So after failing to prove the necessity of kinect with the xbox 360 or a compelling reason to have it they then decided to include it in every single Xbox one, which
1. increased the price
2. decreased the consoles power
3. greatly reduced the number of places they could sell Xbox one in (IIRC only 8 countries or something for months)
perhaps these 3 things wouldnt of been so bad if they kinect had a large benefit to offset them, but it didnt and there were no indications that this would change, they had been working on kinect for what? 3-5 years before the xbox one (*).
They then release the xbox one with kinect bundled, yet no compelling reason why its there (the only reason I heard from ppl on forums was voice commands which could easily be done with a mic built into the console or plugged in, no need for kinect at all)
You and I would be expecting, "oh they must be working on some amazing stuff thats gonna show why kinect is a must have" yet months/years later, nothing compelling. This is why its a bigger failure than Kin

(*)the latency was the killer, the #1 reason why it failed as a gaming device, if they could get that down to mere milliseconds then you could make valid reasons of using it as a game device, I knew this years ago, so why didnt the Xbox engineers? Its not rocket science
 
Status
Not open for further replies.
Back
Top