First up, if you're looking for my upcoming Vive unboxing video, this isn't it!
When my Vive arrived earlier this month, I'd decided to let it sit in its box for a while. The most recent communication regarding official support was "We are working on it but it's not ready yet," and I had a lot of other work to focus on.
About a month ago, an OSVR contributor had mentioned in the OpenHMD IRC channel that OSVR had a driver that interfaced with Valve's Vive driver and allowed OSVR to support the Vive under Linux. I didn't have time to look into it, but was glad to know that even if Linux support wasn't being advertised as ready by Valve, that there was something tangible to work with.
The Vive's official launch came and went without advertised Linux support and it seemed that those users who did have Vives hadn't managed to jump through the correct hoops to get it functioning on Linux until last night when Linux user and developer SketchStick nudged me to take a look at some successes he'd had.
TMA (Too Many Acronyms)
Before going any further, we're going to need to work through some terms. Here's a rough glossary!
I'd also like to pop a disclaimer here. The stuff contained within this article will not magically give Vive support to Linux apps that weren't compiled with OpenVR support, and it doesn't make doing that any more accessible to developers. The value here is primarily for Linux based developers who are looking for examples to work with, and for F/OSS driver developers who are looking for working examples to observe and learn from.
Getting Things Working
The OpenVR SDK ships with an application called HelloVR. It renders a matrix of cubes around the HMD's camera and demonstrates how to interact with the OpenVR API. This example program has some Linux specific code in it, but needs additional tweaking before it will compile and behave happily on Linux (looking at the issue list in the OpenVR SDK's GitHub repo, it seems that there may be hurdles for other platforms as well).
Last year, in an effort to simplify maintenance of his OpenVR AUR package, Christoph Haag forked the Valve OpenVR repo and attempted to keep Linux support for the example applications behaving by testing with the Oculus Rift DK2 he had at the time. After selling his DK2 (more on that a bit later), Christoph continued to keep his fork in sync with Valve's repo, but didn't have any real opportunity to continue testing it to ensure functionality.
These steps have been adapted from instructions originally written by SketchStick as we were working through this stuff last night. Nice work!
To download Christoph's OpenVR SDK fork and build the HelloVR application, run the following commands (note, that you'll need CMake 3.1.0 or greater, as well as SDL2 and GLEW development packages):
Next, you'll want to plug in all those lovely Vive cables and make sure that the device has appropriate permissions. You can either run chmod like so:
Or set up udev rules for usb devices using the vendor codes 0bb4 and 28de as suggested in the OpenHMD readme (this is persistent and probably not a good idea given how early things are right now):
The next step is to make sure that the hellovr app and OpenVR commands have access to the environment and libraries they expect. If you don't already have it installed, you'll need to install SteamVR. This is done by either installing SteamVR from the VR or Tools category in your Steam library, or by using this URL steam://install/250820. Note that at the time of writing, the SteamVR beta branch does not work.
SketchStick has put together the following shell script which assumes that Christoph's OpenVR fork has been cloned into ~/openvr, that Steam has been installed to ~/.local/share/Steam and that the SteamVR depots have installed to ~/.local/share/Steam/steam/SteamApps/common/OpenVR (this stuff varies! Right click on SteamVR in your Steam library, select Properties, switch to the Local Files tab and click on the Browse Local Files button to see where SteamVR/OpenVR has installed to):
To use this script to set up the environment, run the following command in a terminal (it will allow $steam, $openvr and $steamvr to be used as variables in that terminal until it is closed):
It's not essential to use this script, but for the rest of the document, I will assume that you are (expand out the paths as necessary).
From here, we'll need to verify that the Vive is correctly detected by OpenVR.
This will output information on any detected OpenVR devices. You may see a large number of "ioctl (GFEATURE): Broken pipe" lines being output. For now, these are safe to ignore.
The output we care about should look something like this:
If you see a device and driver line that indicate that a Vive is present, we're ready to move forward with starting up the vrserver process and running the HelloVR app.
Start the vrserver process as a background task.
Open a new terminal and set up the SteamVR environment as we did before.
And finally, run the HelloVR app via a launch script that Christoph has provided.
At this point, we should have a working HelloVR app and be able should see something like this:
On the right is the vrcompositor window, which draws what you'll see through the Vive. The image is duplicated and offset for each eye. On the left is a normal monitor with a preview window that runs at a significantly lower frame rate and resolution that allows you to see what's going on without putting the device on.
Interestingly, the preview window shows significant lens compensation distortion while what's displayed through the device doesn't. This seems to indicate that the compositor isn't quite doing everything that it's meant to be doing.
HelloVR is also meant to use the OpenVR API to retrieve 3D models of the controllers and Lighthouse base stations, which are positioned in 3D space, although at the time of writing, I haven't been able to get this aspect to behave correctly.
Anybody who goes through this is likely to notice that there's a significant amount of latency (my gut says it's input latency, but it's difficult to discern the difference between that and render latency). At this point in time, I'm uncertain whether it's introduced by the HelloVR app, the OpenVR abstraction layer, the Vive driver, or even just launch options for vrserver, but it does suggest to me that there's still a little work left to do before the Vive can be considered "Linux ready".
A Little More Background
As soon as I had a stereoscopic image and headtracking working, I sent an email to Christoph to invite him to offer some comments on his work.
Without a device to test with himself and with the dominant understanding being that the Vive can't run on Linux, Christoph was surprised to learn that his efforts to get an OpenVR sample application allowed myself and SketchStick to demonstrate our Vives running on Linux with the existing OpenVR drivers. Equally surprised at his perseverance with maintaining his fork without a device, I asked Christoph about his interests and motivations with regard to VR.
Looking Forward
Christoph's History of Openness chronicles some details of Oculus' history that I wasn't previously aware of and reflects a level of disappointment that I think we all feel - not only over Oculus' apparent indefinite postponement of Linux support, but also for Valve and HTC's decision to launch the Vive without advertised Linux support.
Oculus dropped Linux support at the time that they shifted towards direct rendering mode. Assuming that there weren't any external factors involved in discouraging Linux support (something I'd rather not entertain in this article), it seems fair to assume that there were hurdles with regards to getting that performant and stable that were large enough to justify shelving Linux support in favour of focusing on a single platform.
The HelloVR app doesn't use direct rendering mode either, and it could be easy to assume that Vive's Linux support is a victim of the same prioritisation that lead to Oculus stepping back from Linux. In this case at least, we can see evidence of continued Linux related updates to OpenVR listed on SteamDB, with the most recent changes to OpenVR and Vive/Lighthouse related drivers happening within the past three weeks (change #1920962).
Looking even farther back, we can see that the Linux OpenVR libaries and binaries have continued to be updated alongside other platforms for the past two years (all the way back to change #395316 when vrserver_linux as added).
I want to sorely and strongly stress again that even though this article is about how to get the Vive to run, and that it highlights continued work on Linux support for the Vive, that now is not the time to run out and order a device or to hassle developers who aren't already shipping OpenVR support in their Linux builds. Both of those are important steps, but can't happen until either official vendor support for the Vive on Linux is announced and promoted as ready and stable, or until there is a feature-complete F/OSS driver available.
In the coming weeks, I'm sure that we'll dig around and discover Linux/Vive compatible apps, better identify the cause of (and possibly resolutions for) the latency issue I mentioned previously, and of course, we'll see progress on OpenHMD's Vive branch as well. Hopefully the proper release of Vive's Linux support is also on the near horizon.
Exciting times are ahead!
When my Vive arrived earlier this month, I'd decided to let it sit in its box for a while. The most recent communication regarding official support was "We are working on it but it's not ready yet," and I had a lot of other work to focus on.
About a month ago, an OSVR contributor had mentioned in the OpenHMD IRC channel that OSVR had a driver that interfaced with Valve's Vive driver and allowed OSVR to support the Vive under Linux. I didn't have time to look into it, but was glad to know that even if Linux support wasn't being advertised as ready by Valve, that there was something tangible to work with.
The Vive's official launch came and went without advertised Linux support and it seemed that those users who did have Vives hadn't managed to jump through the correct hoops to get it functioning on Linux until last night when Linux user and developer SketchStick nudged me to take a look at some successes he'd had.
TMA (Too Many Acronyms)
Before going any further, we're going to need to work through some terms. Here's a rough glossary!
- API: Application programming interface, a documented way of interacting with a program or library (to allow interaction without necessarily having access to the library or program's source code)
- SDK: Software development kit, a set of tools for creating applications (typically containing a set of libraries, API headers and examples)
- VR: Virtual reality \o/
- HMD: Head mounted display (typically with positional and/or rotational tracking)
- Positional tracking: The detection and capture of real world movement for use in software
- Rotational tracking: The detection and capture of real world rotation for use in software
- Vive: The HTC Vive, a HTC manufactured VR system created in collaboration with Valve
- Lighthouse: The infrared base station system used by the Vive to provide positional tracking, has also been used as a name for the entire Vive system
- Extended mode: Mapping a HMD display as a traditional monitor (derived from "extended desktop")
- Direct mode: A method for drawing graphics directly to a HMD rather than via the window manager
- SteamVR: Valve's collection of Steam and OpenVR libraries, drivers and demos
- Steam VR mode: A VR friendly version of Steam's "Big Picture Mode"
- OpenVR: A proprietary VR driver abstraction layer which supports the HTC Vive, the Oculus Rift, Razr Hydra, typically by interfacing with vendor provided drivers
- OSVR: An open source VR driver abstraction layer by Razer, Sensics and other partners, which supports the HTC Vive, Oculus Rift, Leap Motion and other devices, typically by interfacing with vendor provided drivers
- OpenHMD: A Free/Open Source Software VR input driver which supports the Oculus Rift DK1 and 2 and other devices
- vrcmd: An OpenVR binary providing command line access to information on OpenVR devices (and maybe other things)
- vrserver: An OpenVR binary that can be run as a service to provide access to OpenVR devices
- vrcompositor: An OpenVR binary that is used to render to OpenVR devices and take care of a bunch of VR specific concerns (distortion, etc.)
I'd also like to pop a disclaimer here. The stuff contained within this article will not magically give Vive support to Linux apps that weren't compiled with OpenVR support, and it doesn't make doing that any more accessible to developers. The value here is primarily for Linux based developers who are looking for examples to work with, and for F/OSS driver developers who are looking for working examples to observe and learn from.
Getting Things Working
The OpenVR SDK ships with an application called HelloVR. It renders a matrix of cubes around the HMD's camera and demonstrates how to interact with the OpenVR API. This example program has some Linux specific code in it, but needs additional tweaking before it will compile and behave happily on Linux (looking at the issue list in the OpenVR SDK's GitHub repo, it seems that there may be hurdles for other platforms as well).
Last year, in an effort to simplify maintenance of his OpenVR AUR package, Christoph Haag forked the Valve OpenVR repo and attempted to keep Linux support for the example applications behaving by testing with the Oculus Rift DK2 he had at the time. After selling his DK2 (more on that a bit later), Christoph continued to keep his fork in sync with Valve's repo, but didn't have any real opportunity to continue testing it to ensure functionality.
These steps have been adapted from instructions originally written by SketchStick as we were working through this stuff last night. Nice work!
To download Christoph's OpenVR SDK fork and build the HelloVR application, run the following commands (note, that you'll need CMake 3.1.0 or greater, as well as SDL2 and GLEW development packages):
git clone https://github.com/ChristophHaag/openvr.git
cd openvr
mkdir build
cd build
cmake -DCMAKE_BUILD_TYPE=Release ../
make
Next, you'll want to plug in all those lovely Vive cables and make sure that the device has appropriate permissions. You can either run chmod like so:
sudo chmod a+rw /dev/hidraw*
Or set up udev rules for usb devices using the vendor codes 0bb4 and 28de as suggested in the OpenHMD readme (this is persistent and probably not a good idea given how early things are right now):
echo KERNEL=="hidraw*", ATTRS{idVendor}=="0bb4", MODE="0666"
echo KERNEL=="hidraw*", ATTRS{idVendor}=="28de", MODE="0666"
udevadm control --reload-rules
The next step is to make sure that the hellovr app and OpenVR commands have access to the environment and libraries they expect. If you don't already have it installed, you'll need to install SteamVR. This is done by either installing SteamVR from the VR or Tools category in your Steam library, or by using this URL steam://install/250820. Note that at the time of writing, the SteamVR beta branch does not work.
SketchStick has put together the following shell script which assumes that Christoph's OpenVR fork has been cloned into ~/openvr, that Steam has been installed to ~/.local/share/Steam and that the SteamVR depots have installed to ~/.local/share/Steam/steam/SteamApps/common/OpenVR (this stuff varies! Right click on SteamVR in your Steam library, select Properties, switch to the Local Files tab and click on the Browse Local Files button to see where SteamVR/OpenVR has installed to):
#!/bin/bash
export openvr=~/openvr
export steam=~/.local/share/Steam
export steamvr=$steam/SteamApps/common/OpenVR
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:\
/usr/lib/:\
/usr/lib32/:\
$openvr/lib/linux32/:\
$openvr/lib/linux64/:\
$steam/ubuntu12_32/steam-runtime/i386/lib/i386-linux-gnu/:\
$steam/ubuntu12_32/steam-runtime/amd64/lib/x86_64-linux-gnu/:\
$steamvr/bin/linux32/:\
$steamvr/bin/linux64/:\
$steamvr/drivers/lighthouse/bin/linux32/:\
$steamvr/drivers/lighthouse/bin/linux64/
To use this script to set up the environment, run the following command in a terminal (it will allow $steam, $openvr and $steamvr to be used as variables in that terminal until it is closed):
source ./steamvr.sh
It's not essential to use this script, but for the rest of the document, I will assume that you are (expand out the paths as necessary).
From here, we'll need to verify that the Vive is correctly detected by OpenVR.
$steamvr/bin/linux64/vrcmd
This will output information on any detected OpenVR devices. You may see a large number of "ioctl (GFEATURE): Broken pipe" lines being output. For now, these are safe to ignore.
The output we care about should look something like this:
QuoteDevice 0 - lighthouse.LHR-1F0D0FBC - Vive MV by HTC - HMD - generic_hmd
Left Projection: left=-1.396245, right=1.242144, top=-1.467562, bottom=1.464816
Right Projection: left=-1.251338, right=1.392499, top=-1.474031, bottom=1.464144
Lens Center: (0.546726, 0.499251), (0.457286, 0.497308)
Driver lighthouse : 1 displays
Vive MV (Serial number LHR-1F0D0FBC)
Driver oculus : 0 displays
Driver oculus_legacy : 0 displays
Driver null : 0 displays
If you see a device and driver line that indicate that a Vive is present, we're ready to move forward with starting up the vrserver process and running the HelloVR app.
Start the vrserver process as a background task.
$steamvr/bin/linux64/vrserver --keepalive
Open a new terminal and set up the SteamVR environment as we did before.
source ./steamvr.sh
And finally, run the HelloVR app via a launch script that Christoph has provided.
$openvr/build/samples/hellovr_opengl/run_hellovr.sh
At this point, we should have a working HelloVR app and be able should see something like this:
On the right is the vrcompositor window, which draws what you'll see through the Vive. The image is duplicated and offset for each eye. On the left is a normal monitor with a preview window that runs at a significantly lower frame rate and resolution that allows you to see what's going on without putting the device on.
Interestingly, the preview window shows significant lens compensation distortion while what's displayed through the device doesn't. This seems to indicate that the compositor isn't quite doing everything that it's meant to be doing.
HelloVR is also meant to use the OpenVR API to retrieve 3D models of the controllers and Lighthouse base stations, which are positioned in 3D space, although at the time of writing, I haven't been able to get this aspect to behave correctly.
Anybody who goes through this is likely to notice that there's a significant amount of latency (my gut says it's input latency, but it's difficult to discern the difference between that and render latency). At this point in time, I'm uncertain whether it's introduced by the HelloVR app, the OpenVR abstraction layer, the Vive driver, or even just launch options for vrserver, but it does suggest to me that there's still a little work left to do before the Vive can be considered "Linux ready".
A Little More Background
As soon as I had a stereoscopic image and headtracking working, I sent an email to Christoph to invite him to offer some comments on his work.
Quote"My changes are actually quite small. Valve definitely started porting the sample to linux (some #ifdef __linux__ etc.) but it seems halfway through apparently they stopped. It's weird because this is basically what makes it build and run.
Well, and they have no build system for linux so I threw some cmake stuff together."
Without a device to test with himself and with the dominant understanding being that the Vive can't run on Linux, Christoph was surprised to learn that his efforts to get an OpenVR sample application allowed myself and SketchStick to demonstrate our Vives running on Linux with the existing OpenVR drivers. Equally surprised at his perseverance with maintaining his fork without a device, I asked Christoph about his interests and motivations with regard to VR.
Quote"I personally had preordered an Oculus Rift DK2 because they sorta had linux support and were bought by Facebook so I thought they'd surely deliver and I could play with unity or unreal. Well, they didn't deliver and I sold my DK2 and wrote History of Openness to express my disappointment with oculus. The SteamVR oculus plugin worked more or less, but all of SteamVR's engine integration is only for windows too, so there is that.
As long as all of the interesting stuff is confined to proprietary platforms that are tied to windows I don't really care that much about VR right now.
I am hoping for the next OSVR HDK to have a display that is on par with Vive and Rift though, then I would likely buy one.
Still the trouble remains that while Oculus and Valve only support VR on windows, almost all VR applications will be windows-only too, so stuff like LibreVR/Revive and a cross platform SteamVR would remain pretty useless, at least until wine can really support low latency translation.
My experience with oculus-wine-wrapper was that it works for some unity applications, but not all and that the latency is a bit off...
Well, back when I bought the DK2 I made this subreddit, but there is not a lot of content.
Since I don't like oculus anymore I thought I'd rather have a subreddit name like this, but.. yea, I need to look for some content."
Looking Forward
Christoph's History of Openness chronicles some details of Oculus' history that I wasn't previously aware of and reflects a level of disappointment that I think we all feel - not only over Oculus' apparent indefinite postponement of Linux support, but also for Valve and HTC's decision to launch the Vive without advertised Linux support.
Oculus dropped Linux support at the time that they shifted towards direct rendering mode. Assuming that there weren't any external factors involved in discouraging Linux support (something I'd rather not entertain in this article), it seems fair to assume that there were hurdles with regards to getting that performant and stable that were large enough to justify shelving Linux support in favour of focusing on a single platform.
The HelloVR app doesn't use direct rendering mode either, and it could be easy to assume that Vive's Linux support is a victim of the same prioritisation that lead to Oculus stepping back from Linux. In this case at least, we can see evidence of continued Linux related updates to OpenVR listed on SteamDB, with the most recent changes to OpenVR and Vive/Lighthouse related drivers happening within the past three weeks (change #1920962).
Looking even farther back, we can see that the Linux OpenVR libaries and binaries have continued to be updated alongside other platforms for the past two years (all the way back to change #395316 when vrserver_linux as added).
I want to sorely and strongly stress again that even though this article is about how to get the Vive to run, and that it highlights continued work on Linux support for the Vive, that now is not the time to run out and order a device or to hassle developers who aren't already shipping OpenVR support in their Linux builds. Both of those are important steps, but can't happen until either official vendor support for the Vive on Linux is announced and promoted as ready and stable, or until there is a feature-complete F/OSS driver available.
In the coming weeks, I'm sure that we'll dig around and discover Linux/Vive compatible apps, better identify the cause of (and possibly resolutions for) the latency issue I mentioned previously, and of course, we'll see progress on OpenHMD's Vive branch as well. Hopefully the proper release of Vive's Linux support is also on the near horizon.
Exciting times are ahead!
Some you may have missed, popular articles from the last month:
Most games suck, so will VR games... but the awesome games will be even more awesome with VR ;)
But ultimately hardcore VR really isn't for the majority of the general public, just like movies are of course better in the cinema than on a small TV with commercial breaks, but the latter is still what the majority (quite intentional, because of course they could go to the cinema) is using.
The real great thing about VR is that it is an enthusiast hardware driven by hardcore games and porn, which pushes the technological boundaries and puts the foundations for Augmented Reality (AR) in place. Because all the tech is very similar, but we are still a quite a bit off achieving good AR. And If you ever read Suarez' Daemon books ( https://en.wikipedia.org/wiki/Daemon_%28novel_series%29 ) you will know that proper AR is a game-changer like the Internet once it arrives.
Last edited by Julius on 19 May 2016 at 2:35 pm UTC
But ultimately hardcore VR really isn't for the majority of the general public, just like movies are of course better in the cinema than on a small TV with commercial breaks, but the latter is still what the majority (quite intentional, because of course they could go to the cinema) is using.
The real great thing about VR is that it is an enthusiast hardware driven by hardcore games and porn, which pushes the technological boundaries and puts the foundations for Augmented Reality (AR) in place. Because all the tech is very similar, but we are still a quite a bit off achieving good AR. And If you ever read Suarez' Daemon books ( https://en.wikipedia.org/wiki/Daemon_%28novel_series%29 ) you will know that proper AR is a game-changer like the Internet once it arrives.
Last edited by Julius on 19 May 2016 at 2:35 pm UTC
0 Likes
Quoting: bubexelSo you picked out that one sentence and ignored the actual point I was making. Well done.Quoting: Mountain ManDoes VR make games more "immersive"? Possibly.Saying that is obvious you have no idea what you talking. At moment you try VR you will understand the meaning of immersive.
Does VR make games more "immersive"? OMG YES! I'm inside the game!
And the reason I said "possibly" is because I have no problem with becoming immersed in a traditional gaming experience (for that matter, I have no problem becoming immersed in a book!). VR provides a different experience, but is that necessarily more immersive? I think it's open for debate.
But anyway, the point is this: does VR make games better? You could just as well ask if 3D makes movies better. I'm not convinced that it does.
Last edited by Mountain Man on 21 May 2016 at 11:17 am UTC
0 Likes
Quoting: Mountain ManBut anyway, the point is this: does VR make games better? You could just as well ask if 3D makes movies better. I'm not convinced that it does.
I don't think that it can inherently make games better, but I do think it can open doors for different kinds of games that couldn't otherwise exist (unlike 3D movies, which don't offer fundamentally different ways of viewing/interacting with an film).
0 Likes
I look at VR this way;
Remember when you pretty much had a 1 button joystick and a black and white TV? Then we got 8 bit computers, so we had a whole keyboard to interact with our games, but still had that 1 button joystick. Then we moved onto a mouse+keyboard, and the games could be a lot more interactive. Not really because of our methods of controlling the systems, but due to the systems becoming more capable as well. Then we started getting gamepads with multiple buttons as a choice, then HOTAS flight sticks, and steering wheels, all things that make interacting with the software more more realistic. It certainly sucks to drive a car with a gamepad. Now we have VR, which gives the head tracking portion of being able to look around. No longer are you turning left and right with a gamepad, well unless you want to. You also get something that no flat screen can really give you, depth. They've tried for years without luck.
If the Leap Motion can take off and get implemented in more games/software the possibilities are pretty endless. So yeah, VR really is a new way to interact with software, and it is almost the ultimate way to do so.
It was mentioned being immersed in a novel. Think Virtual visual novels (which from what I understood, the Abbot's book is such a thing.) Could you imagine an visual novel version of A Game of Thrones? Granted that might be something the porn industry would be into!
By the way, even just using Virtual Desktop to watch episodes of Game of Thrones on a gigantic screen is worth while!
Remember when you pretty much had a 1 button joystick and a black and white TV? Then we got 8 bit computers, so we had a whole keyboard to interact with our games, but still had that 1 button joystick. Then we moved onto a mouse+keyboard, and the games could be a lot more interactive. Not really because of our methods of controlling the systems, but due to the systems becoming more capable as well. Then we started getting gamepads with multiple buttons as a choice, then HOTAS flight sticks, and steering wheels, all things that make interacting with the software more more realistic. It certainly sucks to drive a car with a gamepad. Now we have VR, which gives the head tracking portion of being able to look around. No longer are you turning left and right with a gamepad, well unless you want to. You also get something that no flat screen can really give you, depth. They've tried for years without luck.
If the Leap Motion can take off and get implemented in more games/software the possibilities are pretty endless. So yeah, VR really is a new way to interact with software, and it is almost the ultimate way to do so.
It was mentioned being immersed in a novel. Think Virtual visual novels (which from what I understood, the Abbot's book is such a thing.) Could you imagine an visual novel version of A Game of Thrones? Granted that might be something the porn industry would be into!
By the way, even just using Virtual Desktop to watch episodes of Game of Thrones on a gigantic screen is worth while!
0 Likes
Quoting: slaapliedjeSo yeah, VR really is a new way to interact with software, and it is almost the ultimate way to do so.Nah, that would be direct neural interfaces or whatever utopic/nightmarish (take your pick) sci-fi cliché we'll get in the future. Whatever lets us actually feel like we're moving about and interacting with the virtual environments without actually flailing about and bumping into stuff in the real world. Rolling our heads around wearing clumsy isolation helmets with screens glued to our eyeballs isn't quite there yet.
0 Likes
:D That's why I said almost. Certainly a neural interface into 'cyberworld' would put you there. But then again, while we're probably not terribly far away from having such tech, getting over the 'ewww, you want to place a chip in my brain with how crappy software is known to be?' problem.
Guess it could be like Surrogate where they were connected to machines that let them control a physical avatar. That could definitely be interesting, especially thinking how big of assholes people are on the Internet now, imagine anonymously controlled bots running around.
Guess it could be like Surrogate where they were connected to machines that let them control a physical avatar. That could definitely be interesting, especially thinking how big of assholes people are on the Internet now, imagine anonymously controlled bots running around.
0 Likes
Quoteexport steamvr=$steam/SteamApps/common/OpenVR
Should the OpenVR at the end not be SteamVR?on my system that is
export steamvr=$steam/steamapps/common/SteamVR
However I also get
QuoteVR_IsRuntimeInstalled returned false to indicate that the runtime is not installed.when running vrcmd
Thanks
0 Likes
Quoting: michaeltoalYes, as mentioned in the article, there are a bunch of different possible path/case combinations. You'll have to use whatever is appropriate for your system.QuoteShould the OpenVR at the end not be SteamVR?export steamvr=$steam/SteamApps/common/OpenVR
on my system that is
export steamvr=$steam/steamapps/common/SteamVR
Quoting: michaeltoalHowever I also get
QuoteVR_IsRuntimeInstalled returned false to indicate that the runtime is not installed.when running vrcmd
Thanks
That seems to suggest that your paths aren't set correctly (either the script is wrong, or some libs may be in the wrong place). I've not had a chance to come back to VR stuff since I wrote this article though, and I can't speak to the current state of things (OpenVR and SteamVR have both been updated since).
0 Likes
Thanks for that.
I mentioned changing export steamvr=$steam/SteamApps/common/OpenVR to /SteamVR because it caused me a little confusion and when checking paths I saw a SteamVR folder. It was just in case you wanted to avoid someone (me) thinking it needed to be pointed to the git clone of Christoph Haag's OpenVR fork.
I have checked all the paths over and over, so I assume something is wrong in my setup of steamVR.
Do you have a working installation?
If you do would you mind tarballing it up?
I would be in your debt.
Michael.
I mentioned changing export steamvr=$steam/SteamApps/common/OpenVR to /SteamVR because it caused me a little confusion and when checking paths I saw a SteamVR folder. It was just in case you wanted to avoid someone (me) thinking it needed to be pointed to the git clone of Christoph Haag's OpenVR fork.
I have checked all the paths over and over, so I assume something is wrong in my setup of steamVR.
Do you have a working installation?
If you do would you mind tarballing it up?
I would be in your debt.
Michael.
0 Likes
Quoting: michaeltoalDo you have a working installation?
If you do would you mind tarballing it up?
I would be in your debt.
My Vive has been packed away since I wrote this article and won't be pulled out again until after I'm done with the project I'm working on at the moment.
Since Linux support is not advertised as being ready for use, I'm wary of distributing builds of anything myself - especially if I'm not in a position to personally verify that it works with current versions of SteamVR/OpenVR.
Do you have any output beyond "VR_IsRuntimeInstalled returned false to indicate that the runtime is not installed." when running vrcmd?
0 Likes
Currently working on Winter's Wake, a first person text adventure thing and its engine Icicle. Also making a little bee themed base builder called Hive Time :)
I do more stuff than could ever fit into a bio.
See more from me