|
EbolaIvory posted:They said at OC6 that any quality USB C or w/e should work. LINK cable is "optimized" or some poo poo. It's a fibre optic cable, it's about as optimized as it can get with near-zero latency.
|
#
?
Nov 16, 2019 01:24
|
|
|
|
| # ? Apr 20, 2024 00:26 |
|
|
Neddy Seagoon posted:It's a fibre optic cable, it's about as optimized as it can get with near-zero latency. Are normal DisplayPort cables used in every other headset fiber optic too? I don't get why the Link cable needs that when every other wired headset uses regular rear end display cables with no latency issues.
|
|
#
?
Nov 16, 2019 01:48
|
|
|
Calipark posted:Are normal DisplayPort cables used in every other headset fiber optic too? I don't get why the Link cable needs that when every other wired headset uses regular rear end display cables with no latency issues. It has to do the work of the DisplayPort and USB cable that PCVR headsets use.
|
|
#
?
Nov 16, 2019 02:06
|
|
|
Calipark posted:Are normal DisplayPort cables used in every other headset fiber optic too? I don't get why the Link cable needs that when every other wired headset uses regular rear end display cables with no latency issues. It has nothing to do with latency. The engineering on the cable is all about providing a long, thin, wire that can provide power to the Quest. According to Oculus there are no other USB cables that are 3m long and specced to provide enough power to keep the Quest charged. So if you used other cables you'll probably need to make sure your Quest is charged, and your play time will be limited by the battery life.
|
|
#
?
Nov 16, 2019 02:08
|
|
|
Calipark posted:Are normal DisplayPort cables used in every other headset fiber optic too? I don't get why the Link cable needs that when every other wired headset uses regular rear end display cables with no latency issues. At VR lengths latency is not the issue. However, since the cable will carry enough power to charge a Quest and it’ll bend, you definitely want it to have a strong, reliable signal with minimal interference. I think it’s probably entirely possible to get away with using a conventional cable, in most cases, but I bet using the link cable where you have USB-C on both ends is likely going to be more reliable in the worst case.
|
|
#
?
Nov 16, 2019 02:12
|
|
|
Calipark posted:Are normal DisplayPort cables used in every other headset fiber optic too? I don't get why the Link cable needs that when every other wired headset uses regular rear end display cables with no latency issues.
|
|
#
?
Nov 16, 2019 02:14
|
|
|
Mr Phillby posted:I'm guessing the usb cable has lower bandwidth than a standard displayport (largely due to the quest's hardware I assume, displayport over usb is a thing apparently) so either a standard copper usb either adds a small amount of latency or Oculus want to sell you some gold plated cables because people will buy them. Given Carmack's involved, I'd say they just wanted to go for broke on the most ideal cable possible. Fibre optic doesn't have to deal with things like signal noise or latency over distance, so it's about as good as you can get in that regard. Hell, if Oculus really want to, they can sell the drat thing in just about any size they want.
|
|
#
?
Nov 16, 2019 02:25
|
|
|
BMan posted:John Carmack will not, in fact, program an AGI Well not with that attitude
|
|
#
?
Nov 16, 2019 02:26
|
|
|
Finally got Virtual Desktop working well on my Quest in spite of my Google Wifi mesh network absolutely sucking. Plugged a used 30 dollar router that can give me 5ghz into my bridge and am using it as a dedicated VD streaming SSID. I played Gorn and it felt flawless to me, except for the part where I hit the wall multiple times with my knuckles. Is there any way to have Gorn warn you even more than normal about Guardian boundaries? Maybe I need to change the sensitivity I use for beat saber and make it more sensitive again.
|
|
#
?
Nov 16, 2019 02:46
|
|
|
Chin Strap posted:except for the part where I hit the wall multiple times with my knuckles. So working flawlessly then  I managed to swing downward and bash my controller into my nightstand. Surprised I didn't break the ring on the controller from it.
|
|
#
?
Nov 16, 2019 02:48
|
|
|
Chin Strap posted:Finally got Virtual Desktop working well on my Quest in spite of my Google Wifi mesh network absolutely sucking. Plugged a used 30 dollar router that can give me 5ghz into my bridge and am using it as a dedicated VD streaming SSID. I played Gorn and it felt flawless to me, except for the part where I hit the wall multiple times with my knuckles. You can turn on "strict chaperone" and it'll automatically pause as soon as you cross the line. Personally, when playing Gorn or B&S, I set my chaperone's outline to always show (look up Advanced Settings VR) and that helps me regularly position myself with plenty of room in front, relative to the enemy I'm facing.
|
|
#
?
Nov 16, 2019 02:54
|
|
|
Mr Phillby posted:I'm guessing the usb cable has lower bandwidth than a standard displayport (largely due to the quest's hardware I assume, displayport over usb is a thing apparently) so either a standard copper usb either adds a small amount of latency or Oculus want to sell you some gold plated cables because people will buy them. Neddy Seagoon posted:Given Carmack's involved, I'd say they just wanted to go for broke on the most ideal cable possible. Fibre optic doesn't have to deal with things like signal noise or latency over distance, so it's about as good as you can get in that regard. Hell, if Oculus really want to, they can sell the drat thing in just about any size they want. Neddy Seagoon posted:It's a fibre optic cable, it's about as optimized as it can get with near-zero latency. Neddy Seagoon posted:It has to do the work of the DisplayPort and USB cable that PCVR headsets use. Fiber-optics does not have latency advantages when the distance is measured in meters. It's useful for long distance because it doesn't attenuate as much so you need fewer repeaters boosting the signal, or when something else would otherwise affect signal integrity, such as enough AC power to charge a Quest. Also, unless you specifically have the electronics for it, you can't exactly ingest an external device's displayport and put it on your device's screen. It's a data link, not a graphics input, and this isn't changed by it being copper or optical. It's not VirtualLink.
|
|
#
?
Nov 16, 2019 03:58
|
|
|
https://www.youtube.com/watch?v=kiMWhWCs8sc H3VR's update this week is really good. Though I got my hopes up for nothing on that fictional double-action at the end of the firing range segment; I was hoping for the Trigun one  . The one in the proving grounds makes up for it a bit though  .There's also now support for Liv and ForceTube apparently.
|
|
#
?
Nov 16, 2019 04:00
|
|
|
From everything Oculus said at oc6 it's not using display port over usb at all. It's a video stream that is being encoded on the PC and then decoded in the quest. It's interleaved so it's not being sent as complete frames helping to reduce latency. The bandwidth requirements Carmack mentioned were actually pretty low and nowhere close to maxing out usb3.
|
|
#
?
Nov 16, 2019 04:28
|
|
|
When I'm chrome casting on my quest, and I take off the quest it immediately stops casting. This is annoying when having a party with a bunch of people switching off the quest. Can I disable this behavior?
|
|
#
?
Nov 16, 2019 07:25
|
|
|
Skyarb posted:When I'm chrome casting on my quest, and I take off the quest it immediately stops casting. This is annoying when having a party with a bunch of people switching off the quest. Can I disable this behavior? There's a little sensor in the face mask that detects when you're wearing it, just tape a little piece of paper over it during the party and the headset will think you're always wearing it.
|
|
#
?
Nov 16, 2019 08:32
|
|
|
mashed_penguin posted:From everything Oculus said at oc6 it's not using display port over usb at all. It's a video stream that is being encoded on the PC and then decoded in the quest. It's interleaved so it's not being sent as complete frames helping to reduce latency. The bandwidth requirements Carmack mentioned were actually pretty low and nowhere close to maxing out usb3. Makes it sound like playing PC games on the Quest is a total after thought... Even though it seems on paper to be a huge bullet point for it. Here's hoping that the second gen Quest has native display input. Or I guess by that time wireless will be acceptable with low cost hardware so they'll just do that. Having a really solid $300 entry level unit that does both PC linked (wired or otherwise) and solo play sounds like exactly what VR needs right now. And from what I hear the Quest is already crazy solid for the price point.
|
|
#
?
Nov 16, 2019 09:20
|
|
|
Neddy Seagoon posted:It's a fibre optic cable, it's about as optimized as it can get with near-zero latency.
|
|
#
?
Nov 16, 2019 15:26
|
|
|
I was so impressed by VD now that I'm off a mesh network that I'm not even excited for link right now. What is the length of cat5 cable from my WiFi router to my network bridge would start actually causing signal degradation or slowdown? I would like to keep my router hard wired to my main bridge but put it in a place farther away in the house.
|
|
#
?
Nov 16, 2019 15:43
|
|
|
Doctor w-rw-rw- posted:Also, unless you specifically have the electronics for it, you can't exactly ingest an external device's displayport and put it on your device's screen. It's a data link, not a graphics input, and this isn't changed by it being copper or optical. It's not VirtualLink. "I said the work of", not a direct replacement. It's still gotta manage bandwidth carrying a similar amount of data. I do think the Link cable, from a technical standpoint, was just meant to go for broke and erase as many variables as possible on paper. I'm actually curious to see if the modules in them are 5GB/s or 10GB/s, as neither are outside of the price range of the cable and it'd futureproof the heck out of it for a fair while if they're 10GB/s ones. Combat Pretzel posted:Fiber optics doesn't mean anything. More so on short distances, because the transceivers are there and introduce their own latencies. IIRC, copper is faster on short distances than fiber because of that. I forgot what the cutoff is. I think it's around the 100m mark when things start to go sideways for copper versus optical fibre iirc? Chin Strap posted:I was so impressed by VD now that I'm off a mesh network that I'm not even excited for link right now. Unless you're running it to another house down the street, you'll be fine. Neddy Seagoon fucked around with this message at 16:09 on Nov 16, 2019 |
|
#
?
Nov 16, 2019 16:07
|
|
|
I think Oculus is using fiber optic for other characteristics, like light weight.
|
|
#
?
Nov 16, 2019 16:51
|
|
|
Chin Strap posted:I was so impressed by VD now that I'm off a mesh network that I'm not even excited for link right now. They really need a better acronym.
|
|
#
?
Nov 16, 2019 17:04
|
|
|
Fuzz posted:They really need a better acronym. One of the products I worked with had a commonly used module with the acronym STD.
|
|
#
?
Nov 16, 2019 17:58
|
|
|
Played more Stormlands. It keeps being good.
|
|
#
?
Nov 16, 2019 18:15
|
|
|
Skyarb posted:When I'm chrome casting on my quest, and I take off the quest it immediately stops casting. This is annoying when having a party with a bunch of people switching off the quest. Can I disable this behavior? I think there’s a setting somewhere that puts the Quest to sleep after a set number of seconds after you take it off instead of immediately.
|
|
#
?
Nov 16, 2019 19:12
|
|
|
Neddy Seagoon posted:
This is because of signal attenuation, not latency. In either case, the latency of your cable is a drop in the bucket compared to the switch it's connected to.
|
|
#
?
Nov 16, 2019 19:40
|
|
|
Calipark posted:Makes it sound like playing PC games on the Quest is a total after thought I mean... it is. They've been pretty open about that. The Quest is a standalone headset with no display input that they managed to retrofit a solution into after the fact, it wasn't built from the ground up to allow PC games which is why it's taken months for the feature to be activated. It's still drat impressive that it exists at all, considering nothing else even attempts to straddle that line.
|
|
#
?
Nov 16, 2019 19:50
|
|
|
Electrons travel through copper wire at the speed of light, just as Wifi radio waves do. What slows both down is the switching latency of the routing equipment. Other than that the only delays are the encoding/decoding.
|
|
#
?
Nov 16, 2019 20:45
|
|
|
Zero VGS posted:Electrons travel through copper wire at the speed of light, just as Wifi radio waves do. What slows both down is the switching latency of the routing equipment. Other than that the only delays are the encoding/decoding. Not to say it materially changes your point, but electrons move super slow and even wave propagation is like 2/3 the speed of light depending on what dielectrics are present and a bunch of other stuff.
|
|
#
?
Nov 16, 2019 20:58
|
|
|
Delta-Wye posted:Not to say it materially changes your point, but electrons move super slow and even wave propagation is like 2/3 the speed of light depending on what dielectrics are present and a bunch of other stuff. The biggest impact in latency on high speed (10/25gb) interconnects is the SFP media adapter performance. Optical interconnects have always been lower latency than copper. We are talking nanoseconds here, completely meaningless for any relevant performance scenario except HPC for algo stock trading.
|
|
#
?
Nov 16, 2019 21:51
|
|
|
Delta-Wye posted:Not to say it materially changes your point, but electrons move super slow and even wave propagation is like 2/3 the speed of light depending on what dielectrics are present and a bunch of other stuff. rage-saq posted:The biggest impact in latency on high speed (10/25gb) interconnects is the SFP media adapter performance. Optical interconnects have always been lower latency than copper. We are talking nanoseconds here, completely meaningless for any relevant performance scenario except HPC for algo stock trading. Both of these are true, but I appreciate Delta-Wye for being pedantic about it.
|
|
#
?
Nov 16, 2019 22:04
|
|
|
Leal posted:Next up, forklift VR https://store.steampowered.com/app/939450/Forklift_Simulator_2019/ My dreams have come true!
|
|
#
?
Nov 16, 2019 23:13
|
|
|
Leal posted:https://store.steampowered.com/app/939450/Forklift_Simulator_2019/ As someone who used to drive a forklift, this extremely speaks to me.
|
|
#
?
Nov 16, 2019 23:24
|
|
|
Without the risk of actual physical damage and harm, this is not fun.
|
|
#
?
Nov 17, 2019 00:23
|
|
|
Delta-Wye posted:Not to say it materially changes your point, but electrons move super slow and even wave propagation is like 2/3 the speed of light depending on what dielectrics are present and a bunch of other stuff. OK, if it's half the speed of light, that would be about a foot per two nanoseconds. So across 10 feet of USB-C or Wifi it should be about 20 nanoseconds? You'd only need 999,980 more nanoseconds to reach 1ms of latency, just to put in into perspective for people here. I don't know how you can say electrons move "super slow". The time for the GPU to render the frame, NVENC hardware to encode it, and the Quest GPU to decode and display it comes in around 15-30 milliseconds, and that's where the most gains stand to be made (for example from a new generation of GPUs with yet faster encoding, or saving on decode time with more techniques like the slicing that Virtual Desktop uses). The switching speed of the transistors in the the PC's NIC card, the router's CPU etc, make up a tiny fraction of the total latency here. edit: For people wondering about the concept of the Oculus Link, it is using a miniaturized version of the circuitry of an SFP like the PCB in the middle here:  So each plug end of the Oculus Link has something like that to convert electrical 1's and 0's into basically "morse code" sent out with a laser diode, and picked up with the photodiode on the other end of the cable. This allows them to use fiber optic polymer plastic to carry the signal, the cable is similar to that plastic in those kid's light toys like these:  As other posters touched on, the benefit is that light is immune to interference/crosstalk unlike copper wire. There are negative/positive copper wires in the Oculus Link for carrying power, but power doesn't get disrupted by radio interference and crosstalk like data over copper can. Because the optical cables can be small, it leaves room for the copper power wires to be large enough to deliver power to the Quest without the voltage dropping too much (thin wires would lose voltage as resistance/heat along the length of the wire). The overall cable can be lighter and more durable overall (localized strain, such as a office chair rolling over the cable, might tear thin copper data wires more easily than the thick copper power wires and optical polymer). I'm not a professor in this poo poo or anything, but I spend a lot of time behind a soldering iron. Hopefully that's illuminating. edit 2: If anyone wants the "Oculus Link" technology in an HDMI cable and has some extra money, you can get HDMI cables from Monoprice that go insane lengths and have a lifetime warranty, I use one at my house. They have USB extensions too but I don't recommend those as I've had compatibility issues. Here's the HDMI though: https://www.monoprice.com/product?p_id=21568 Zero VGS fucked around with this message at 01:34 on Nov 17, 2019 |
|
#
?
Nov 17, 2019 01:07
|
|
|
Goons are saucy about the properties of electrons.Zero VGS posted:Electrons travel through copper wire at the speed of light. Drift velocity of an electron is a bit over a inch per minute or so. It takes several hours for an electron to enter your house, pass through the wiring, through your wife's vibrator, and then back out to the distribution lines while you post on the internet. Hardly the speed of light! The interesting number is how fast waves propagate, but even that is some fraction of the speed of light. Like I said, useless pedantry but that statement isn't correct.
|
|
#
?
Nov 17, 2019 02:07
|
|
|
tl;dr:Doctor w-rw-rw- posted:At VR lengths latency is not the issue.
|
|
#
?
Nov 17, 2019 02:38
|
|
|
VR Megathread: powering your wife's dildo is slower than you think
|
|
#
?
Nov 17, 2019 04:07
|
|
|
Unless you plan to have a 1000km long cable for your quest, the difference between fiber and copper doesn’t really matter.
|
|
#
?
Nov 17, 2019 04:09
|
|
|
|
| # ? Apr 20, 2024 00:26 |
|
|
IDK why you guys are all hung up about the speed of electrons and latency, isn't the whole thing explicitly that USB 3 cable lengths are severely limited by impedance and superspeed being very susceptible to interference and crosstalk? You can't really get passive USB 3 cables longer than 2 or 3m, which is pretty short for VR purposes, and the official Oculus optical cable is going to be like 5m? Anyway, the Polybius OpenVR update came out like yesterday, has anyone gotten that to work? When the launcher pops up and I hit the VR button, it momentarily pops up the SteamVR theater and then closes. I don't even have the theater enabled
|
|
#
?
Nov 17, 2019 04:22
|
|
