This is the perfect technology for North Americans! Using this thing, you can make walks down city streets pleasant and even pretty, no matter how run-down the neighborhood you are walking through! The bums become invisible, unless you are about to step on them. Everything is clean, no faded paint, no broken windows, no graffiti, no visible signs of urban decay. And if it doesn't look broken, why bother to fix it? Every car you see is brand new and clean, and some of the consumer products you see might seem a little brighter and nicer, since they are 'your brand.' Targeted marketing takes on a whole new meaning when your perception of reality is software driven. This is the sort of technology hinted at in "The Congress". There, they used drugs and chemically enhanced experiences to create 'perfect worlds' for the residents of the City, this gizmo lets you see only what you want to see, and not one pixel more. We won! We defeated reality!
Holy hell this guy is very realistic. Most of the companies are so marketing-based it seems like they're making promises with the complete intention of not managing to keep them. I don't think any of use believe them anymore than they do. But this guy, he's got a realistic concept of cost, development time, systems limitations. If anyone's going to succeed in this, I think it'll be him.
When people say inside-out tracking nowadays they really mean markerless inside-out, this is why it causes some confusion as it's, according to my Wikipedia knowledge, based on where the sensors are in relation to the tracked object. Which with the Vive makes it inside-out (but not markerless) and the Rift outside-in. Pretty straight forward, when using a few more labels to describe things. Main stream just says inside-out for anything markesless though :p which feels like a normal thing to happen with technical terms.
I really think the scanning mechanisms used in the demo game Scanner Sombre could make for a useful tool in mixed reality hardware today, as it allows to user to specify the depth of detail they need in an area by changing the concentration of many beams from a LIDAR based scan, then having each environmental reflection mapped into the system.
I'm just curious, did he notice how Pickle Rick's... um... "tip"... was being slightly covered up by his mike? It kinda looks like the Incredible Hulk is happy to see him and it's about to get REALLY friendly with his face.
people seem to miss-atribute inside out tracking. Vive is inside out. only really Rift is outside in. this is a third technology where inside out optics are used.
Controller issue - Over thinking things. If each controller has the same sensors as the headset, the controller doesn't need to render anything. All it needs to do is give the headset/computer it's position. So all the controller needs to do is output positional and hand input data. With the headset pushing out it's own IR laser blaster, this can be simplified by having each controller having a dual camera setup in the same configuration as the samsung 360 camera. That with an accelerometer for general controller orientation (not position. Just tilt) the cameras wouldn't need to be high res.
Highly mistaken. you can build a motion tracking system for $40 now (look up eyemotionos). it was maybe 15 years ago that you could get a webcam with face tracking for like $80. A nintendo wii controller is able to track an IR blaster (lightbar, but was stationary) for way $60 when the wii was a thing and that only had an IR sensitive camera (only showed a black picture with white view-able IR points. low rez). To have it view 200+* with any camera is all lensing. To make sense of all that is software. I could see something like a future 5th gen raspberry pi being powerful and portable enough to fit in a controller.
It's an interesting technology but I don't think this is good enough to replace tracking with external help. Interesting compromise would be to allow for arbitarilly placed external stations to improve precision and controller tracking. The positional system presented here could be easily used to track the location of those helper stations.
What if the vive was meant to have inside out tracking and that was meant to be their ground breaking thing for the vive but it for some reason they had to delay it so they just brought out the front facing camera
I don't see the problem with the tracking issue with the controllers. I have seen (and visited) many companies who use a hybrid solution with other sensors that keep track on these things. So I think many people should open their vision (no pun) or even combine their knowledge to work together in these cases.
One more comment about costs. Although sometimes those are special chips and solutions, it all comes down as just a matter of numbers. That the ONLY reason why other things (like in your cell phone) are so cheap.
Vive's psuedo-inside-out solution can and should be expanded by additional lighthouses. Theoretically it should be possible to expand the range of your playfield near limitlessly with adding more units. I wish HTC would explore that more.
Its not HTC but Valve, the tracking technology is theirs. And they have already solved it -> steamcommunity.com/games/steamvrtracking/announcements/detail/1264796421606498053 The TS4231 part uses 5 components per sensor, so it is cheaper to place at each sensor. More importantly, though, it provides a burst of data per laser or sync hit. Using that data allows information to be transmitted on the laser itself, which we can use to learn about the source of that laser. This new capability to encode information in the laser is significant for two reasons: It allows support for more than two base stations, and thus larger tracking volumes. It allows a base station to function without including a sync blinker, which is the source of most of the interference between base stations (and is also a significant driver of base station cost.) We call this technology sync-on-beam.
I really don't want this as a part of PC VR. Norm and Jeremy have brought up the problem of controllers being inferior when using this approach a couple of times, and what it'd require to be on par with the solutions we've already had for a year. They're not stressing it, in order not to be opinionated on video, but it's dead obvious it's going to be an inferior. No amount of weird cameras on the headset is going to solve controller occlusion. If magnetic tracking worked it would've been used everywhere already - Sixense was first on the market and got superseded by better solutions. Valve's tracking is dead simple, cheap to make (HTC's base station and controller prices could be 3x less and still make proper profits), it's nigh perfect in operation, and it's free for everyone to license. PC tracking solved. I hope that SteamVR and Oculus tracking prevail with keeping the bar on what's acceptable tracking and others be forced to follow or get pushed out. This will be great for mobile, but keep it out of PC ecosystem please, Microsoft is already about to make it difficult and we don't need more of these to make proper VR that we started with become a minority of the market. This poses serious limitations onto developers, a lot of existing games will be unplayable with these, and new ones will have to be dumbed down. Translating into: new customers buying new cheap VR kits will be frustrated en masse, VR will suddenly and unjustly be considered as "it sucks" just because some companies are greedy to capture their pie of the market even if it means they're messing it up for themselves and everyone else. It's great that they're working on this stuff, but they'll want they R&D money back and won't be consciencious not to bring the PC VR experinece a step back. We as consumers have to take stance - spend $100-200 more and have a proper VR experience, and educate people around you on what to buy; as developers - don't bend to support an inferior device. Looking forward, it's easy to see a form of base stations is here to stay. Wireless VR will require antennas to be mounted similarly to how Vive's base stations are mounted. Going forward, Kinect-like full body tracking or even a consumer version of holoportation (th-cam.com/video/7d59O6cfaM0/w-d-xo.html) will also require mounting the cameras in the same fashion (ideally it'd all be a single device). And they look badass in the corner of your room, you wanted future sci-fi gizmos, here they are!
puedes aprender viendo los simpsons y/o padre de familia y similares en ingles. despues de un tiempo vas a acostumbrar el oido y te encontraras buscando palabras que no conoces en el diccionario.
You guys need to make a Rick and Morty vid. Interesting the creators.... and why they have a big following of loyal fans It woud get Alot of veiws! Pickle Rick!
It seems like the industry is really hostile toward lighthouse tracking. I wonder why that is, considering its basically perfect (just need two outlets).
Posting 3 years later, but I would have replied similar back then... I'm in a rented apartment and don't want to deploy Lighthouse in the "appropriate" way for this apartment -- to keep them out of the way and off of tripods would require involving my landlord. WMR is inside out tracked -- and while far from perfect, I paid less for it than I would for a Valve-tech system. I loved Vive when I set it up at work, but it's just not suitable for my life situation. I'm likely not alone. :/
@@ivucica yeah thats fair. im lucky to have ceiling vents and magnets so I stick them up that way, but I've definitely come around to inside out tracking since I posted this.
Jacek Jagosz the Leap Motion is really not a finish product, its more or less a prototype sell as a finished product. Too many problems poor sdk support in several game engines and way to less range to do proper IO tracking. The Leap is a neat idea, but not very well thought out.
leap motion had an update called orion ...the tracking was much better ....they have a new sensor with wider fov....and plans for different sensors ....and r talking to manufacturers to use the new optimized sensor for integrating in headsets blog.leapmotion.com/charting-the-course-for-mobile-vr/
Can't get excited about inside out tracking due to its limitations, I feel it’s a step backwards. Within VR, rest your hand on a VR table containing a pyramid of playing cards. Look away from the table then move your hand through the playing cards. When you look back at the table with an outside in system then the pyramid will have turned into a pile of cards. With inside out your hand will be behind the pyramid. Now think about the issues if you are using trackers on your feet to provide locomotion. Inside out has no way of seeing your feet where as this is not a problem with outside in (Rift / Vive). Issues become worse if you are a curvy female. Adding more cameras to inside out certainly helps but can’t solve the issue. The cost of ownership is not good either. The Rift sale price 399, normal 499 and the Vive 599. These new systems come in at 399-499? This is a stepping stone between cardboard and the Rift / Vive yet is hitting Rift pricing. Sure these systems are great for putting bunny ears on your cat but thats going to get dull fast. To be honest it's hard to think of any long term fun use of MR that isn't strictly adult(s) only. As for gaming we already have dumbed down PC gaming due to console performance and gamepad controllers. We don't need dumbed down PC VR due to the restrictions of inside out tracking. Perhaps professional use can save the system but even then I can't beleive it would be the best solution.
I did recognise that 'Adding more cameras to inside out certainly helps but can’t solve the issue.'. At what point does it get silly though? Adding cameras to the knees so that feet maybe tracked, perhaps the VR bra for the more shapely lady? MS are working on adding outside in tracking to the system as an option but that also adds to the cost taking it above the Rift and possibly the Vive. The only plus that these MR headsets have for general use is the higher resolution of 1440x1440 vs 1080x1200 of the existing Rift / Vive. This will provide an argument for productivity as text will be a little sharper but also has the downside during gaming where more pixels have to be pushed. 1440p/60Hz is a struggle for oc'd 980Ti / 1070 with some newer games while the headsets target 90Hz to reduce motion sickness.
stop bringing up the holo lens when it comes to VR!, AR (or mixed reality) is nothing like VR, AR (or mixed reality) is taking the existing 3D space you are in, and projecting something on top of that, so basicly if you dumb it down to simple terms, AR is a camera that allows you to overlay stuff on top in real time (think of it as a more advanced version of what snapchat filters do) where VR is creating a whole new virtual envoirment to submerge you in, the 2 are completly diffrent concepts so for the love of gaben dont compare things holo lens does when talking about VR
It was so refreshing hearing this guy give honest answers to their limitations and problems instead of dodging questions.
The best VR and AR reporting in the industry .
This is the perfect technology for North Americans!
Using this thing, you can make walks down city streets pleasant and even pretty, no matter how run-down the neighborhood you are walking through!
The bums become invisible, unless you are about to step on them.
Everything is clean, no faded paint, no broken windows, no graffiti, no visible signs of urban decay.
And if it doesn't look broken, why bother to fix it?
Every car you see is brand new and clean, and some of the consumer products you see might seem a little brighter and nicer, since they are 'your brand.'
Targeted marketing takes on a whole new meaning when your perception of reality is software driven.
This is the sort of technology hinted at in "The Congress".
There, they used drugs and chemically enhanced experiences to create 'perfect worlds' for the residents of the City, this gizmo lets you see only what you want to see, and not one pixel more.
We won!
We defeated reality!
Holy hell this guy is very realistic. Most of the companies are so marketing-based it seems like they're making promises with the complete intention of not managing to keep them. I don't think any of use believe them anymore than they do. But this guy, he's got a realistic concept of cost, development time, systems limitations. If anyone's going to succeed in this, I think it'll be him.
I am tickled pink that I discovered this channel. right up my alley.
Great content! Keep up the good work!
Yes! Woohoo I asked about a month or so ago for a structure sensor review and you guys delivered. Love this show
When people say inside-out tracking nowadays they really mean markerless inside-out, this is why it causes some confusion as it's, according to my Wikipedia knowledge, based on where the sensors are in relation to the tracked object. Which with the Vive makes it inside-out (but not markerless) and the Rift outside-in. Pretty straight forward, when using a few more labels to describe things. Main stream just says inside-out for anything markesless though :p which feels like a normal thing to happen with technical terms.
google doesn't need occipital for arcore ....they got tango....and can integrate it whenever they like....and tango is basically the same thing
Great interview and discussion. Pushing the boundaries, one hurdle at a time. Thanks!!
I really think the scanning mechanisms used in the demo game Scanner Sombre could make for a useful tool in mixed reality hardware today, as it allows to user to specify the depth of detail they need in an area by changing the concentration of many beams from a LIDAR based scan, then having each environmental reflection mapped into the system.
A Google image search of the game would give a better idea of what I mean, lol
This was really good!
I'm just curious, did he notice how Pickle Rick's... um... "tip"... was being slightly covered up by his mike? It kinda looks like the Incredible Hulk is happy to see him and it's about to get REALLY friendly with his face.
can we have an update to that? Is this implemented already in current VR generations already?
people seem to miss-atribute inside out tracking. Vive is inside out. only really Rift is outside in. this is a third technology where inside out optics are used.
The Windows Mixed Reality HMDs does this right?
djorgen yeah, and I think Google Daydream Standalone HMDs aswell.
Sounds like the 'middle-out' algorithm would be the perfect solution...
middle out .....so with lighthouses attached to drones flying around u
Silicon valley Season 1 episode 8
need the mean jerk time
Putting a em-field emitter near our head??? Is it safe?
What about the new Windows Mixed Reality Headset, when you are going to try it?
Ew.
Ew.
Controller issue - Over thinking things. If each controller has the same sensors as the headset, the controller doesn't need to render anything. All it needs to do is give the headset/computer it's position. So all the controller needs to do is output positional and hand input data.
With the headset pushing out it's own IR laser blaster, this can be simplified by having each controller having a dual camera setup in the same configuration as the samsung 360 camera. That with an accelerometer for general controller orientation (not position. Just tilt) the cameras wouldn't need to be high res.
Highly mistaken. you can build a motion tracking system for $40 now (look up eyemotionos). it was maybe 15 years ago that you could get a webcam with face tracking for like $80. A nintendo wii controller is able to track an IR blaster (lightbar, but was stationary) for way $60 when the wii was a thing and that only had an IR sensitive camera (only showed a black picture with white view-able IR points. low rez). To have it view 200+* with any camera is all lensing. To make sense of all that is software. I could see something like a future 5th gen raspberry pi being powerful and portable enough to fit in a controller.
It's an interesting technology but I don't think this is good enough to replace tracking with external help. Interesting compromise would be to allow for arbitarilly placed external stations to improve precision and controller tracking. The positional system presented here could be easily used to track the location of those helper stations.
What if the vive was meant to have inside out tracking and that was meant to be their ground breaking thing for the vive but it for some reason they had to delay it so they just brought out the front facing camera
A tetherless inside-out tracking system is also going to need GPS assuming that your play space is the Earth's surface.
Norm, you better test the Pimax 8K...it's the real deal. Kickstarter is coming Sept 19!!
Always forget to comment, love this!!
I like the smartwatch squad, they know their stuff...
I don't see the problem with the tracking issue with the controllers.
I have seen (and visited) many companies who use a hybrid solution with other sensors that keep track on these things.
So I think many people should open their vision (no pun) or even combine their knowledge to work together in these cases.
One more comment about costs. Although sometimes those are special chips and solutions, it all comes down as just a matter of numbers.
That the ONLY reason why other things (like in your cell phone) are so cheap.
and the vive has a usb port on the top for accessories
Single camera location tracking would be amazing in Snap Spectacles 2.0
What about neural network tracking algorithms ?
Does anyone know the intro song name and artist?
Vive's psuedo-inside-out solution can and should be expanded by additional lighthouses. Theoretically it should be possible to expand the range of your playfield near limitlessly with adding more units. I wish HTC would explore that more.
Its not HTC but Valve, the tracking technology is theirs. And they have already solved it -> steamcommunity.com/games/steamvrtracking/announcements/detail/1264796421606498053
The TS4231 part uses 5 components per sensor, so it is cheaper to place at each sensor. More importantly, though, it provides a burst of data per laser or sync hit. Using that data allows information to be transmitted on the laser itself, which we can use to learn about the source of that laser.
This new capability to encode information in the laser is significant for two reasons:
It allows support for more than two base stations, and thus larger tracking volumes.
It allows a base station to function without including a sync blinker, which is the source of most of the interference between base stations (and is also a significant driver of base station cost.) We call this technology sync-on-beam.
I really don't want this as a part of PC VR. Norm and Jeremy have brought up the problem of controllers being inferior when using this approach a couple of times, and what it'd require to be on par with the solutions we've already had for a year. They're not stressing it, in order not to be opinionated on video, but it's dead obvious it's going to be an inferior. No amount of weird cameras on the headset is going to solve controller occlusion. If magnetic tracking worked it would've been used everywhere already - Sixense was first on the market and got superseded by better solutions. Valve's tracking is dead simple, cheap to make (HTC's base station and controller prices could be 3x less and still make proper profits), it's nigh perfect in operation, and it's free for everyone to license. PC tracking solved. I hope that SteamVR and Oculus tracking prevail with keeping the bar on what's acceptable tracking and others be forced to follow or get pushed out.
This will be great for mobile, but keep it out of PC ecosystem please, Microsoft is already about to make it difficult and we don't need more of these to make proper VR that we started with become a minority of the market. This poses serious limitations onto developers, a lot of existing games will be unplayable with these, and new ones will have to be dumbed down. Translating into: new customers buying new cheap VR kits will be frustrated en masse, VR will suddenly and unjustly be considered as "it sucks" just because some companies are greedy to capture their pie of the market even if it means they're messing it up for themselves and everyone else. It's great that they're working on this stuff, but they'll want they R&D money back and won't be consciencious not to bring the PC VR experinece a step back. We as consumers have to take stance - spend $100-200 more and have a proper VR experience, and educate people around you on what to buy; as developers - don't bend to support an inferior device.
Looking forward, it's easy to see a form of base stations is here to stay. Wireless VR will require antennas to be mounted similarly to how Vive's base stations are mounted. Going forward, Kinect-like full body tracking or even a consumer version of holoportation (th-cam.com/video/7d59O6cfaM0/w-d-xo.html) will also require mounting the cameras in the same fashion (ideally it'd all be a single device). And they look badass in the corner of your room, you wanted future sci-fi gizmos, here they are!
Just curious what happened to Will?
ugh, it's really kind of frustrating because you can tell the technology is nearly there but no body's nailed it down yet
how long before you can just put your phone into a headset link it to the pc via rift cat and just use camera for tracking.
i've been trying to find this for too long
But what about middle out?
any news about Magic Leap project?
dev kit maybe in December or after that .....and literally no other news
Augmented ARKit or ARCore with sensors is what Google Tango was
Those cursor-on-screen moments...
Anyone else notice a buzzing sound?
patchnauty sounds like their mic gain is too high
I would love someone to set up a park, much like laser tag, using this tech for a shooter game!
I have the feeling you are betraying Leap Motion right there. There current Integrated hand sensor is solid with nothing up to wishes.
I just want the OASIS
someone should definitely make something called the oasis
pumpuppthevolume not until we have actually good VR technology.
I’m not sure how into VR i will be until they can figure out how to do it without motion sickness. :/
did he call them "trash controllers"?
Is Occipital dead?
SUBTITLES IN SPANISH PLEASE. I AM LATINO AMERICANO AND NOT SPEAK ENGLISH.I LIKE THEIR CONTENT, THEIR VIDEOS ARE VERY GOOD.
Esau Mendoza learn English
Bartosz Olszewski I'LL DO THAT. THANKS FOR YOUR ANSWER.
And turn off capslock
puedes aprender viendo los simpsons y/o padre de familia y similares en ingles.
despues de un tiempo vas a acostumbrar el oido y te encontraras buscando palabras que no conoces en el diccionario.
Frank Winterlong gracias por tu consejo. Lo tomaré en cuenta.
Make a armor stand for king aurthers armor
Dope
It's pickle Riiiiiccck!
aljowen pickle rick BIG DICK !!!
oculus quest boys where we at?
The dream 1:19
Pokemon go Next Level !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
You should make a marionette puppet.
You guys need to make a Rick and Morty vid. Interesting the creators.... and why they have a big following of loyal fans
It woud get Alot of veiws!
Pickle Rick!
Odd Ryan no dead memes are not funny
can i pls get your shirt? thank you
You need LIDAR
Bradley Coopers
haha pickle rick xxxxdddDDDDDDDDD nice shirt
It seems like the industry is really hostile toward lighthouse tracking. I wonder why that is, considering its basically perfect (just need two outlets).
Posting 3 years later, but I would have replied similar back then...
I'm in a rented apartment and don't want to deploy Lighthouse in the "appropriate" way for this apartment -- to keep them out of the way and off of tripods would require involving my landlord. WMR is inside out tracked -- and while far from perfect, I paid less for it than I would for a Valve-tech system.
I loved Vive when I set it up at work, but it's just not suitable for my life situation.
I'm likely not alone. :/
@@ivucica yeah thats fair. im lucky to have ceiling vents and magnets so I stick them up that way, but I've definitely come around to inside out tracking since I posted this.
10 toes down
Leap Motion, anyone?
Jacek Jagosz the Leap Motion is really not a finish product, its more or less a prototype sell as a finished product. Too many problems poor sdk support in several game engines and way to less range to do proper IO tracking. The Leap is a neat idea, but not very well thought out.
leap motion had an update called orion ...the tracking was much better ....they have a new sensor with wider fov....and plans for different sensors ....and r talking to manufacturers to use the new optimized sensor for integrating in headsets blog.leapmotion.com/charting-the-course-for-mobile-vr/
I see dead vr
Can't get excited about inside out tracking due to its limitations, I feel it’s a step backwards.
Within VR, rest your hand on a VR table containing a pyramid of playing cards. Look away from the table then move your hand through the playing cards. When you look back at the table with an outside in system then the pyramid will have turned into a pile of cards. With inside out your hand will be behind the pyramid. Now think about the issues if you are using trackers on your feet to provide locomotion. Inside out has no way of seeing your feet where as this is not a problem with outside in (Rift / Vive). Issues become worse if you are a curvy female. Adding more cameras to inside out certainly helps but can’t solve the issue.
The cost of ownership is not good either. The Rift sale price 399, normal 499 and the Vive 599. These new systems come in at 399-499? This is a stepping stone between cardboard and the Rift / Vive yet is hitting Rift pricing.
Sure these systems are great for putting bunny ears on your cat but thats going to get dull fast. To be honest it's hard to think of any long term fun use of MR that isn't strictly adult(s) only. As for gaming we already have dumbed down PC gaming due to console performance and gamepad controllers. We don't need dumbed down PC VR due to the restrictions of inside out tracking.
Perhaps professional use can save the system but even then I can't beleive it would be the best solution.
I did recognise that 'Adding more cameras to inside out certainly helps but can’t solve the issue.'. At what point does it get silly though? Adding cameras to the knees so that feet maybe tracked, perhaps the VR bra for the more shapely lady? MS are working on adding outside in tracking to the system as an option but that also adds to the cost taking it above the Rift and possibly the Vive.
The only plus that these MR headsets have for general use is the higher resolution of 1440x1440 vs 1080x1200 of the existing Rift / Vive. This will provide an argument for productivity as text will be a little sharper but also has the downside during gaming where more pixels have to be pushed. 1440p/60Hz is a struggle for oc'd 980Ti / 1070 with some newer games while the headsets target 90Hz to reduce motion sickness.
Pickle Riiiiiick
PICKLE RICK !!!!!!! Yesss !!!
does he keep it 10 toes down while his fam is in therapy?
Theme music sucks. Sorry, it does.
stop bringing up the holo lens when it comes to VR!, AR (or mixed reality) is nothing like VR, AR (or mixed reality) is taking the existing 3D space you are in, and projecting something on top of that, so basicly if you dumb it down to simple terms, AR is a camera that allows you to overlay stuff on top in real time (think of it as a more advanced version of what snapchat filters do) where VR is creating a whole new virtual envoirment to submerge you in, the 2 are completly diffrent concepts so for the love of gaben dont compare things holo lens does when talking about VR
They're not that different at all. Black out an AR display and you have a VR headset. They're just 2 points on the same spectrum.
31 minutes? Come on guys, I know you have a lot to say but this could have been trimmed down a bit
Lul
1000th
That guy is soooo boring , I fell asleep.
hahaha .. i was riveted. great content guys!
it was perfectly engaging and not too long episode..............when u play it at 1.5x the speed
i keep forgetting the speed trick .. thanks for reminding me Mr pump
Top/first comment
First