SteamVR Tracking HDK

SteamVR Tracking HDK

SteamVR Tracking Technology Update

50 Rate up
< >
LittleV.Qr Oct 12, 2017 @ 12:26am 
What's the price
frstrtr Jul 23, 2017 @ 8:59am 
End of june were passed, any updates? We need those new base-stations!!! :steammocking:
FEBER Jun 19, 2017 @ 9:38pm 
How can we get the new station as soon as possible as we want to build compatible hardwares?
roothorick Jun 6, 2017 @ 2:05pm 
I'm okay with the asymmetric backwards compatibility, personally. There was an expectation (though no guarantee) of full backwards compatibility, but current owners are highly unlikely to have a reason to upgrade their base stations.

We have the backwards compatibility we actually care about; I'll change out the HMD and controllers as I get tempted by new technology and features, but my Vive bases will be mounted to my walls for as long as they continue to function. By the time they DO fail, I'll probably have replaced the rest of the setup with 2.0 gear at a minimum, if not 3.0.
Lhun Jun 6, 2017 @ 12:57pm 
I realize BOLL that is the case, but if my math and logic is correct it should be possible for the older photosensors to handle maybe 2-4 max of the NEW devices pulses by using the +- pulse modulation as the "sync flash" with a little bit of logic.

You would have to work with "half of half" the 1-5mhz modulation bandwidth, but I don't see why you couldn't do it. We already have different modulations for the sync flash.
BOLL Jun 6, 2017 @ 12:42pm 
It's a split, new controllers and headsets will be able to use old base-stations, but old devices can not use the new base-stations. It makes sense as making the new base-stations backwards compatible would in principle turn them into 1+2 devices with the LED flash back in place, just like the early PS3 that had the PS2 chip built into it.

If we want the cost to go down for next generation VR, so more people will adopt it, we will just have to accept the loss of complete backwards compatibility. This is the cost of being an early adopter.
Graham J Jun 6, 2017 @ 12:24pm 
While I appreciate the forward movement of this excellent tech, fragmentation of tracking devices seems like a bad thing. Making these 2.0 devices backwards compatible should be a priority.
Lhun Jun 6, 2017 @ 11:38am 
So, knowing this, I still think that older devices could be made to work with the new lighthouses via a little logic, but I could be wrong (but only up to 2, maybe 4?) If the current "flash, sweep xy, flash, sweep yx, flash..." method allows us to determine which station is b or c, couldn't the data coming into the TS3633's from modulated + - IR be read and used in the same way? Use the + code as "sweep" and - as "pulse"? volo plz. I know the possibility of interference is greater, but this doesn't seem impossible.
Lhun Jun 6, 2017 @ 11:27am 
In the new device, a carrier frequency modulation can be placed on the laser light itself, probably using a gaussian transform or something similar (this is how bluetooth "channels" work) Rather than reading the modulated timing "flash" pulse and determining which base station it came from, based on the more broad modulation of channel c and b, the new TS4231 device can read a positive AND negative beam state each (modulated). I'm assuming valve is using a "+" state and a "-" state for each channel's IR emitter, and modulating the lasers so each one is unique, no longer requiring the modulated "flash". The TS4231 can simultaneously determine which base station is hitting the emitter, and there's 5mhz more available bandwidth for modulation. Assuming 5mhz allowed for two base station frequencies and a sync, 10mhz allows for at least 4 frequencies width, and adding 2 states multiplies that further.
Lhun Jun 6, 2017 @ 11:24am 
Oh, I get it, looking at the diagram. To simplify, the new photodiode supports FSK. (frequency shift keying).
The previous diode used OOK or ASK (probably ASK due to the mentioned interference issues from the clock timer pulse)
So, the TS3633 photodiaode (binary OOK/ASK) receives an "on" pulse when it "sees" light, and is assumed to be "off" when it doesn't see light. There's two frequencies the gen1 base stations produce for the TIMING flash (channel b and channel c), which is detected by both the base station's single TS3633 and the HMD's. This is modulated. The current diode detects either "lighthouse sweep" intensity or "timing flash"(b or c) or "nothing."