Upgrade Odyssey to G2 moderate hardware?

[LithiumR]

I’m thinking of upgrading my HMD from the original Odyssey to the Reverb G2 BUT I’m not ready to upgrade my rig yet. My current specs are Intel I7 7820HK at 4.3Ghz, Nvidia 1080, 32Gb ram. I run DCS okay(ish) now on the odyssey. Been running it a couple years with this setup. 
 

I’m wondering how the trade off goes between pushing more pixels in the HMD vs being able to run less super sampling. Would I just break even? Net gain? Net loss? If I could just break even and end up with the same quality I have now minus the Odyssey’s SDE I’d be good with that but my fear is ultimately having to dial it down so much that the final image would still be better in the Odyssey. So I guess my question is, Does being able to drop the ss in SVR make up for the load of pushing more pixels?
 

At some point I’ll end up upgrading my rig but that might be a while. I feel like I just built this one only a couple of years ago! DCS is the only thing I run that it can’t keep up with lol (or barely keeps up with). 
 

Just curious what everyone thinks. Thanks. 

[speed-of-heat]

if your perf is "ok"  now then it's all about the pixels.

 

Providing you are not going above 2304000  (your current resolution if you are running at native in the oddy) pixels per eye you should be fine in steam VR ... the native resolution for the G2 is 4665600, and 9808128 in steam at 100% res... so in steam you will need to set  your CRM for DCS at about 20-25% at least to begin with but start there and work you way up.

 

HTH

[LithiumR]

23 minutes ago, speed-of-heat said:

if your perf is "ok"  now then it's all about the pixels.

 

Providing you are not going above 2304000  (your current resolution if you are running at native in the oddy) pixels per eye you should be fine in steam VR ... the native resolution for the G2 is 4665600, and 9808128 in steam at 100% res... so in steam you will need to set  your CRM for DCS at about 20-25% at least to begin with but start there and work you way up.

 

HTH

I’m running 160% (1832x2288) in per app in sp, 150% (1776x2216) in mp with pd @ 1.0. General in steam is at 100%. With shader mods and shared parser I hold 45fps; approx 12ms frametime on cpu; and about 19ms frametime on GPU (depends on server... that’s pretty typical on GS). I also run reshade sharpen filter. It’s okay but I do get the double image if I look at my 3 or 9 and roll. 

Edited August 25, 2021 by LithiumR

[speed-of-heat]

about 40 or 30 % then depending on if you are in SP or MP based on those numbers 

[Baldrick33]

I have previously had an Odyssey+ and now have a G1 and the clarity difference is amazing.

 

I run the G1 at 100% which is 2196x2156 so not too much difference to yours.

 

Note that the G2 lenses are optimised in SteamVR to run at higher resolution although the panel is the same resolution, so 100% is something just over the 3000x3000 mark as I recall. I did get a G2 but ended up selling it and keeping the G1 as I couldn't see any real improvement without hurting performance and frankly cranking up the resolution on the G1 was so close. I use speakers so don't care about the audio.

 

Just throwing this out there in case a decent G1 was available at a good price .

[Nodak]

G-2 from an odyssey, colors aren't quite as rich on the G-2, but close, the clarity you get will never match no matter what you do to an odyssey.  Odyssey has the best cockpits ever with those rich colors and its close up clarity is not bad, but put on a G-2 and your suddenly flying heads out of the pit, occasionally scanning instruments.  Gaining the outside world is well worth it.  My rigs a bit less than yours but upgraded the card to 2070Super upon gaining the G-2.  You can divide your frames in half to get a close estimate on how your G-2 will pull.

[The_Nephilim]

16 hours ago, LithiumR said:

I’m thinking of upgrading my HMD from the original Odyssey to the Reverb G2 BUT I’m not ready to upgrade my rig yet. My current specs are Intel I7 7820HK at 4.3Ghz, Nvidia 1080, 32Gb ram. I run DCS okay(ish) now on the odyssey. Been running it a couple years with this setup. 
 

I’m wondering how the trade off goes between pushing more pixels in the HMD vs being able to run less super sampling. Would I just break even? Net gain? Net loss? If I could just break even and end up with the same quality I have now minus the Odyssey’s SDE I’d be good with that but my fear is ultimately having to dial it down so much that the final image would still be better in the Odyssey. So I guess my question is, Does being able to drop the ss in SVR make up for the load of pushing more pixels?
 

At some point I’ll end up upgrading my rig but that might be a while. I feel like I just built this one only a couple of years ago! DCS is the only thing I run that it can’t keep up with lol (or barely keeps up with). 
 

Just curious what everyone thinks. Thanks. 

Well while you may not be able to run the G2 at native res you can gedt by @50% res in steam vr even with all settings in DCS down to minimal it will look way better then the O+ I had a simialr run as you  Odyssey +,, G2,, Rift S,, Back to a G2 and I have a 1080GTX..

[The_Nephilim]

Well I had messed with DCS for another few hours tweaking and here is what I came up with.

 

1. Resolution in Steam VR @ 2232x2184 which came out to be 50% I set it in per application, still 1:1 ratio and with DCS PD at .8.. here are my settings in DCS:

 

 

 

[LithiumR]

7 hours ago, The_Nephilim said:

Well I had messed with DCS for another few hours tweaking and here is what I came up with.

 

1. Resolution in Steam VR @ 2232x2184 which came out to be 50% I set it in per application, still 1:1 ratio and with DCS PD at .8.. here are my settings in DCS:

 

 

 

Is there a benefit to running Pd .8 with higher ss vs Pd at 1.0 with lower ss?

[speed-of-heat]

None at all… the only thing that matters is the total number of pixels driven… just be aware that pd 0.8 is not the the same as ss 80% they are different measurements. 
 

jabbers does a really good video explaining this 

 

 

[The_Nephilim]

19 minutes ago, speed-of-heat said:

None at all… the only thing that matters is the total number of pixels driven… just be aware that pd 0.8 is not the the same as ss 80% they are different measurements. 
 

jabbers does a really good video explaining this 

 

 

Yes I am well aware of this video but I am not trying to do this.. I just wanted to set the Res in Steam vr at 50% or my native res so I keep the 1:1Pixel ration then drop the PD too .8 I am not trying to do the .5 --500% trick.. 

[speed-of-heat]

the same effect is achieved by leaving pd @ 1.0  and reducing the res in steam to 32%.

 

if your issue is that you want to keep your headset at 50% for other games, be aware that steam has a crm multiplier for each game… which if your base resolution is at 50% in steam  then the crm would be 64%  … you can also continue to use pd 0.8 and ss 50%.

 

it makes no difference where you tweak the numbers… the only thing that matters is the total number of final pixels.PDSSCalc.zip

 

steam provides more granular control than pd, so you might find you can eek out some more pixels improving your image quality. Ymmv.

 

 

 

 

Edited August 28, 2021 by speed-of-heat

[The_Nephilim]

5 hours ago, speed-of-heat said:

the same effect is achieved by leaving pd @ 1.0  and reducing the res in steam to 32%.

 

if your issue is that you want to keep your headset at 50% for other games, be aware that steam has a crm multiplier for each game… which if your base resolution is at 50% in steam  then the crm would be 64%  … you can also continue to use pd 0.8 and ss 50%.

 

it makes no difference where you tweak the numbers… the only thing that matters is the total number of final pixels.PDSSCalc.zip

 

steam provides more granular control than pd, so you might find you can eek out some more pixels improving your image quality. Ymmv.

 

 

 

 

 

OK I just thought keeping the 1 at 50% and just lower the PD in DCS.. I will experiment with 32% and see.. I am doing the math and trying to keep it down the pixels I mean.

[speed-of-heat]

In my experience changing it in two places causes a lot of confusion, hence the reference to the jabbers video. But providing you understand the math, then it really doesn’t matter which way, but, whenever I’m explaining to someone I really recommend only using one variable, especially because the a pd of 0.8  and ss of 80% sound like the should be the same…

[The_Nephilim]

Well I had tried it at 32% and it seemed to look a bit better then the other way I was doing it.. I am just trying to figure out how many pixels a 1080GTX can push comfortably..

 

I seen someone mention the can game on a 3440x1440 monitor with a 1080 so I did the math and it is 4,953,600 pixels.. I am currently at 3,118,432. now I am sure it is a bit more complicated then that I am just trying to determine a round about number of pixels I should be able to push..

 

So far it has worked and I think I might be able to squeeze a bit more out of the old 1080gtx..

[speed-of-heat]

its not a straight map from monitor resolution... firstly the number is per eye... so it double, and then there is a fudge factor to deal with VR STUFF TM (R) ... i normally allow about 1/3rd or 3 x ...for back of fag packet comparison to the 2d 

[The_Nephilim]

7 hours ago, speed-of-heat said:

its not a straight map from monitor resolution... firstly the number is per eye... so it double, and then there is a fudge factor to deal with VR STUFF TM (R) ... i normally allow about 1/3rd or 3 x ...for back of fag packet comparison to the 2d 

Ahh yes I keep forgetting to 2x the rez.. But I was just trying to do this more methodically then just guessing..

[speed-of-heat]

I understand, I was just never able to isolate anything, that would account for the difference, in a scientific way … observation was the best method I had …so …

[speed-of-heat]

have you found any way to adjust the digital vibrancy for just the G2?

[j9murphy]

I two have a O+ and have considered going to the G2... for those that have done that did you miss the wider FOV of the O+?

[Baldrick33]

8 minutes ago, j9murphy said:

I two have a O+ and have considered going to the G2... for those that have done that did you miss the wider FOV of the O+?

I went from O+ to Pimax 5K+ to Reverb G1. Never missed the FOV of the Pimax, the clarity of the Reverb was (is) simply amazing in comparison!

[speed-of-heat]

34 minutes ago, j9murphy said:

I two have a O+ and have considered going to the G2... for those that have done that did you miss the wider FOV of the O+?

Nope.

Let me clarify I would love a wider fov, but not at the sacrifice of pixel density.