HP Reverb G2 Announced - next Gen

[speed-of-heat]

I currently have a Reverb v2 G1, it drives the same number of pixels as the g2, on my current hardware and settings ... i get good "consistent 45 FPS"across most scenarios except in some very heavy missions. I use a PD of 1 and my SS is typically at 100%.. i can read the cockpit fine...

 

PD of 1.7 would have a massive impact on performance, but you simply don't need it with the reverb, increasing SS in steam makes much more sense if you want enhanced clarity as it provides a more granular level of control

[boedha68]

Hi, everybody:

 

 

I only use virtual reality on DCS, and I'm just about to pre-order a new Reverb G2 because it looks like it's going to be the best to use in simulation, and take advantage of the 100 Euro discount on the pre-purchase.

But first I would like to ask a question to those of you who currently have a Reverb:

 

My equipment is a Rayzen 3700X, with 32 GB at 3600MHz and an RTX 2080.

Do you think this is enough equipment to handle the new G2?

 

Reading forums around, I see that the graphics are fair, but people who have 1080 Ti and 2080 "just" say that it is enough - logically a 2080 Ti will go better, since it is known that for GPU VR you always want the best and it will still be little?

 

Currently I fly with a Rift S, always OffLine and in light missions, and I am happy with the performance as long as I don't go to PD higher than 1.2. This way I get 40 fps with active ASW on smooth NOE flights.

 

As people justify that the Reverb does not need to increase the PD they say that the performance even improves, but of course they usually come from flying with high PDs: 1.7 or even more in their previous HMDs, hence my doubts...

 

I look forward to your contributions, which will always be welcome...

 

Regards

 

 

45 FPS straight you can get with the HP reverb G2 with your system now!

 

 

I play 45 - 90 FPS with the reverb g1 now. See system in signature.

 

 

Because i will the best of the best i gonna upgrade at the end of the year. But will it Nvidia or AMD... time will tell.

 

 

90 FPS DCS on high settings is a dream. I'll hope it finally comes true. :D

[sze5003]

My MB does have a front panel USB C-Port, which is hooked up to front of my case.

 

My Quest with Link works just fine in it so hopefully my Reverb G2 will also.

Yea hopefully it does. I looked up the specs of my motherboard and my USB C port only supplies 3 Amps of power. So as long as yours puts out 6 watts you should be good.

 

Looks like I'll need to use an adapter and also have the headset plugged into a power outlet.

Edited July 23, 2020 by sze5003

[dburne]

Yea hopefully it does. I looked up the specs of my motherboard and my USB C port only supplies 3 watts of power. So as long as yours puts out 6 watts you should be good.

 

Looks like I'll need to use an adapter and also have the headset plugged into a power outlet.

 

Just looked up the specs on my Z390 Dark MB, that USB -C front panel port I want to use is listed as 5V 3 Amp. None of the USB headers on this MB goes above 5V. The manual does say if more is needed to just use a powered USB Hub though.

[X-31_VECTOR]

Just looked up the specs on my Z390 Dark MB, that USB -C front panel port I want to use is listed as 5V 3 Amp. None of the USB headers on this MB goes above 5V. The manual does say if more is needed to just use a powered USB Hub though.

 

Unless I'm forgetting my Ohm's Law [nerdiest thing I've said today, which is saying a lot], 5V at 3 amps is 15 watts. So maybe you're fine?

[dburne]

Unless I'm forgetting my Ohm's Law [nerdiest thing I've said today, which is saying a lot], 5V at 3 amps is 15 watts. So maybe you're fine?

 

Ok thanks, will see it probably will be fine.

[X-31_VECTOR]

It's all pretty confusing, with USB-C being a connector and not a USB standard. Example: USB 3.1 -- which is what my mobo's Type C connector is -- is supposed to deliver up to 20 volts at 5A (= 100W).

 

Yet my motherboard's specs say its USB 3.1 port can deliver 3A of current, not 5A. So... is that 3A at 20 volts (60W)? Or is the power delivery spec for USB 3.1 just the maximum, and anything less than that is fair game to be called USB 3.1 as long as the data throughput requirement is met?

 

Not being smart enough to puzzle it out, my plan when my G2 arrives is to plug it into the USB-C port and just see what happens.

[sze5003]

It's all pretty confusing, with USB-C being a connector and not a USB standard. Example: USB 3.1 -- which is what my mobo's Type C connector is -- is supposed to deliver up to 20 volts at 5A (= 100W).

 

Yet my motherboard's specs say its USB 3.1 port can deliver 3A of current, not 5A. So... is that 3A at 20 volts (60W)? Or is the power delivery spec for USB 3.1 just the maximum, and anything less than that is fair game to be called USB 3.1 as long as the data throughput requirement is met?

 

Not being smart enough to puzzle it out, my plan when my G2 arrives is to plug it into the USB-C port and just see what happens.

Yea my z370 E Asus board has some usb A 3.1 ports + type c and it says they support 3A power output. The usb C is on the same section as the other two USB A ports.

 

I may as well try it out and if not I'll just have to find a bigger extension cord to plug the damn headset in when I get one.

[Peace Puma]

Which one implies more graphic load? The HP reverb glasses or a 4K monitor (3840x2160). regards

[speed-of-heat]

The reverb is 2160 x 2160 x 2

[Peace Puma]

The reverb is 2160 x 2160 x 2

 

????

[X-31_VECTOR]

????

 

...hence more pixels to push to the Reverb than a 4K monitor. Hence more "graphical load" (assuming there is no efficiency gain in basically rendering the same scene twice, once for each eye). There are also complications in VR like super sampling and reprojection, but I am not nearly smart enough to know how those impact the GPU versus CPU.

[Thadiun Okona]

????

 

 

It's easy to figure out. Multiply the total pixels first then by the refresh rate and you get the graphical load in terms of pixels per second that need to be rendered

 

4k: [3840 x 2160] x 60 = 497,664,000

 

G2: [2160 x 2160 x2] x 90 = 839,808,000

 

Conclusion: G2 at 90hz is roughly 1.7x the load of 4k at 60hz.

[Peace Puma]

Thanks

[Gryzor]

It's easy to figure out. Multiply the total pixels first then by the refresh rate and you get the graphical load in terms of pixels per second that need to be rendered

 

4k: [3840 x 2160] x 60 = 497,664,000

 

G2: [2160 x 2160 x2] x 90 = 839,808,000

 

Conclusion: G2 at 90hz is roughly 1.7x the load of 4k at 60hz.

 

Conclusion-> 1,7x .... get a 3080ti

[VitS]

Except that for VR it's twice as much CPU work because game actually renders 2 different frames

[speed-of-heat]

Except that for VR it's twice as much CPU work because game actually renders 2 different frames

 

i think thats covered ...

G2: [2160 x 2160 x2] x 90 = 839,808,000

 

im almost never bound by cpu... except in the heaviest of missions

Edited July 23, 2020 by speed-of-heat

[X-31_VECTOR]

im almost never bound by cpu... except in the heaviest of missions

 

What's your preferred method of monitoring GPU and CPU utilization? FPSVR seems to show GPU headroom even when I suspect one core is slammed, so I'm looking for something else to give me a better picture of what's really going on under the hood. Windows Performance Monitor? AIDA64? Thanks!

[Burnerski]

I'm aiming for 45 FPS. 90 FPS is not gonna happen with todays hardware. If a video card comes out that is capable of 90 FPS then my cpu may be the limiting factor.

[speed-of-heat]

What's your preferred method of monitoring GPU and CPU utilization? FPSVR seems to show GPU headroom even when I suspect one core is slammed, so I'm looking for something else to give me a better picture of what's really going on under the hood. Windows Performance Monitor? AIDA64? Thanks!

 

I use FPSVR as well and i'm more looking at the frame times for CPU than the state of the core . which usually shows me 2 or 3 ms of headroom, e.g. frame times between 6 and 9 ms for the CPU

 

I'm aiming for 45 FPS. 90 FPS is not gonna happen with todays hardware. If a video card comes out that is capable of 90 FPS then my cpu may be the limiting factor.

 

pretty much this

[The_Nephilim]

It's easy to figure out. Multiply the total pixels first then by the refresh rate and you get the graphical load in terms of pixels per second that need to be rendered

 

4k: [3840 x 2160] x 60 = 497,664,000

 

G2: [2160 x 2160 x2] x 90 = 839,808,000

 

Conclusion: G2 at 90hz is roughly 1.7x the load of 4k at 60hz.

 

 

Well, according to this method on my Odyssey Plus I have roughly 414,720,000. I am near the limit for a 1080GTX as I get most times in MP around 45-55FPS..

 

So, I am guessing if I did buy the G2 I would most likely need a new video card as well since my video card is near maxxed out with the O+? I doubt this 1080 could push twice the pixels??

 

where would I find info on where to get the amount of pixels a 1080 can push per second. from what I see wouldn't this caluclation be bandwidth based for the cable capability?

Edited July 24, 2020 by The_Nephilim

[Waxer]

Remember guys, the solution to improve frame rates is not just to buy a more expensive graphics card. There are plenty of graphics settings where you can compromise on "quality". And oftentimes the quality sacrifices are barely noticeable.

 

Anti aliasing and shadows tend to be the two biggies where an intermediate sweet spot gives you visual quality without being too much of a GPU resource hog.

[imacken]

I've said it a zillion times before, but IMO, you can tweak till the cows come home, but the only things that REALLY affect performance are MSAA, SteamVR SS/PD and Shadows. Everything else is fairly inconsequential.

[X-31_VECTOR]

Well, according to this method on my Odyssey Plus I have roughly 414,720,000. I am near the limit for a 1080GTX as I get most times in MP around 45-55FPS..

 

So, I am guessing if I did buy the G2 I would most likely need a new video card as well since my video card is near maxxed out with the O+? I doubt this 1080 could push twice the pixels??

 

where would I find info on where to get the amount of pixels a 1080 can push per second. from what I see wouldn't this caluclation be bandwidth based for the cable capability?

 

The G2 will have a 60Hz mode (as opposed to 90), as well as a mode to set it to half its native resolution, presumably to help folks with less-than-cutting-edge hardware.

 

Of course, I can see little point in buying a G2 just to run it in half resolution, but it might provide a bridging solution if you get the G2, find you can't get it to run smoothly, and then need some time to convince the wife that you need a 3080Ti more than the kids need new shoes.

[sze5003]

The G2 will have a 60Hz mode (as opposed to 90), as well as a mode to set it to half its native resolution, presumably to help folks with less-than-cutting-edge hardware.

 

 

 

Of course, I can see little point in buying a G2 just to run it in half resolution, but it might provide a bridging solution if you get the G2, find you can't get it to run smoothly, and then need some time to convince the wife that you need a 3080Ti more than the kids need new shoes.

Yep that 3080ti will be pricey. Especially if you need to swap out other components too. I recently read some article where it said the 3080ti needs a more powerful psu connector, could be a 12pin instead of 8. So for some people that have modular psu's I guess you add another cable if you have the room and ability.

 

Otherwise you may need another psu. Not so expensive in reality but it's always a pain to swap out psu's.