Oculus Rift and DCS World Discussion

[Frusheen]

Installed the 980ti no noticeable difference in DCS some increase in P3D about 10% to 20%.

 

R/

Daniel

 

I'm only seeing about 40% load on my 980ti with DCS in the rift and cpu has one core at about 70-80%. Some serious optimization needed.

[Enduro14]

I'm only seeing about 40% load on my 980ti with DCS in the rift and cpu has one core at about 70-80%. Some serious optimization needed.

 

Well unless they rebuild the engine I don't see much improvements unless they incorporate nvidia game works VR. It's the same problem in nvidia 3dvision. Engine despite how pretty and awesome it is just is not made nor optimized for 4K,VR,3D or multiple screens when it only runs off one cpu core. Hopefully something happens otherwise it will be same old story similar to Arma 3. A more powerful gpu makes no difference when you hit that proverbial wall of none cpu optimization which we have and still encounter with Dcs world amongst other titles.

Edited January 9, 2016 by Enduro14

[Farlander]

Well unless they rebuild the engine I don't see much improvements unless they incorporate nvidia game works VR. It's the same problem in nvidia 3dvision. Engine despite how pretty and awesome it is just is not made nor optimized for 4K,VR,3D or multiple screens when it only runs off one cpu core. Hopefully something happens otherwise it will be same old story similar to Arma 3. A more powerful gpu makes no difference when you hit that proverbial wall of none cpu optimization which we have and still encounter with Dcs world amongst other titles.

 

And give the amd users even more crap? no f*** that..

[TOYKILLA]

Well apparently most Asus boards don't pass, because they have AsMedia USB controllers. You are lucky, yours doesn't. From what I have read the test is not very reliable at all. I don't know what CPU you have, but it is failing Sandy Bridge CPU's that are way faster than some CPU's it passes and recommending buying lower rated CPU's. I read it just looks for Ivy Bridge +, which is kinda simplistic and inaccurate. I mean some nicely overclocked Sandy Bridges will smoke some stock Ivy's easily. Its probably about as accurate as Windows Experience Index.

 

Asus Owner just buy a USB 3.0 PCI-E card for 25 bugs and the Rift runs fine :-)

[hansangb]

Well unless they rebuild the engine I don't see much improvements unless they incorporate nvidia game works VR. It's the same problem in nvidia 3dvision. Engine despite how pretty and awesome it is just is not made nor optimized for 4K,VR,3D or multiple screens when it only runs off one cpu core. Hopefully something happens otherwise it will be same old story similar to Arma 3. A more powerful gpu makes no difference when you hit that proverbial wall of none cpu optimization which we have and still encounter with Dcs world amongst other titles.

 

 

I don't know. The improvement from 1.2 to 1.5 was fantastic. No reason to think they maxed out w/o a complete rewrite. Not just yet, anyway.

[Frusheen]

Well unless they rebuild the engine I don't see much improvements unless they incorporate nvidia game works VR. It's the same problem in nvidia 3dvision. Engine despite how pretty and awesome it is just is not made nor optimized for 4K,VR,3D or multiple screens when it only runs off one cpu core. Hopefully something happens otherwise it will be same old story similar to Arma 3. A more powerful gpu makes no difference when you hit that proverbial wall of none cpu optimization which we have and still encounter with Dcs world amongst other titles.

 

Seems to me that GameworksVR wouldn't make things any better. Based on the usage I am seeing since upgrading my graphics card the bottleneck seems to be the CPU.

 

Still it's alpha software and I'm sure we will see more improvements as time goes on.

[Warthog_Farmer]

So I have a 980Ti and a Rift on the March list :) Super happy...

 

Still a little peeved about the whole Pascal thing though. It will make the sale of the 980Ti a little harder for sure. Go early and cash in (but not know how Pascal will perform in the real world) or take a hit on the price but give Pascal time... Hmmm Hmmm Hmmm

 

Anyway - since we're discussing the engine... wasn't Edge supposed to offload CPU to GPU where possible? And we're still seeing 40% GPU with 80% CPU.

 

This seems like a gateway to performance improvements. I'm guessing it's pretty difficult to offload to GPU if this hasn't been implemented already?

 

#needmoreframesforRift

[Enduro14]

Seems to me that GameworksVR wouldn't make things any better. Based on the usage I am seeing since upgrading my graphics card the bottleneck seems to be the CPU.

 

Still it's alpha software and I'm sure we will see more improvements as time goes on.

 

Game works would make it better as long as the developers added it to their core engine otherwise it's just bolt on tech yes and make no difference.

[Enduro14]

So I have a 980Ti and a Rift on the March list :) Super happy...

 

Still a little peeved about the whole Pascal thing though. It will make the sale of the 980Ti a little harder for sure. Go early and cash in (but not know how Pascal will perform in the real world) or take a hit on the price but give Pascal time... Hmmm Hmmm Hmmm

 

Anyway - since we're discussing the engine... wasn't Edge supposed to offload CPU to GPU where possible? And we're still seeing 40% GPU with 80% CPU.

 

This seems like a gateway to performance improvements. I'm guessing it's pretty difficult to offload to GPU if this hasn't been implemented already?

 

#needmoreframesforRift

 

That is million dollar question..

[Kuky]

You can be stationary and still tell the direction. Levels can be exactly the same if the sound is coming directly from above/below/ahead/behind and you can still tell without moving, because your outer ear and your body is changing the frequency response.

 

Sorry but not correct... I have just done an experiment with my wife to prove I am right. I made her sit in a chair and told her not to move and keep her eyes closed... I was holding my mobile and let it ring and I kept moving it around her head, in particular I want to see if she can tell when I moved the mobile from in front, to right above her head (keeping the distance same) and guess what... she couldn't tell it was above her head. When I asked her where the phone is, she pointed in front of her and said "somewhere there". Reason for this is, without any extra info she could get (no eye sensors and no secondary position to make sound volumes comparison) both hear ears heard pretty much the same volumes on both ears, and as I started with phone in front of her, and she heard same sound volumes when phone was above her head, she assumed it must be in front again... that frequency change and stuff you guys think matters so much (as if we have hearing as good as bats) do not matter to our ears at all.

 

Only because we are able to move the hear, even so slightly, we are able to get different levels of sound change, hence we can tell straight away if we are looking more or less towards the sound source, and that is all there is to it. Keeping fully still and sound source not changing, we can only approximately tell side on which sound is coming from but definitely not pin-point it exactly as you guys think we can, to be able to do this, once more, you need at least one more reference (either memory, or sight, or another listening position).

 

If you heard for the first time some sound you have no idea what it is (you hear this sound for the first time in your life) and you only stay still, you will NOT be able to tell where its coming from or how far it is, you can only tell side its coming from (left/right) and until you start to move hear around, you will not be able to pin-point the sound source).

 

You guys still want to believe otherwise, be my guest, I think I have done enough to prove I am in fact right about this.

Edited January 9, 2016 by Kuky

[kam]

I'm only seeing about 40% load on my 980ti with DCS in the rift and cpu has one core at about 70-80%. Some serious optimization needed.

 

Is that in Nevada?

 

I'm seeing more like 10-12% CPU usage and 60-100% GPU..

[S3NTRY11]

You guys still want to believe otherwise, be my guest, I think I have done enough to prove I am in fact right about this.

It really sounds like you're set on relying on your own anecdotal evidence, which is not the best way to go.

 

I've read your posts, and you really need to listen to the people that obviously have a more thorough understanding.

 

You are incorrect. Sorry.

[Kuky]

It really sounds like you're set on relying on your own anecdotal evidence, which is not the best way to go.

 

I've read your posts, and you really need to listen to the people that obviously have a more thorough understanding.

 

You are incorrect. Sorry.

 

No, I haven't set my mind based on my own anecdotal evidence, what I have done is have a theory which proved to be right by doing an experiment, which is always the right way (as long as the way you an experiment is correct and you are excluding all other factors)

[Frusheen]

Is that in Nevada?

 

I'm seeing more like 10-12% CPU usage and 60-100% GPU..

 

Are you looking at a total % for all cores or 10-12% on a single core?make sure you view each core.

 

I've done a lot of experimenting this evening and can't recall which map I was on for those figures. I'll have to double check.

[S3NTRY11]

No, I haven't set my mind based on my own anecdotal evidence, what I have done is have a theory which proved to be right by doing an experiment, which is always the right way (as long as the way you an experiment is correct and you are excluding all other factors)

And in this 'experiment', what is your control?

To name but one problem with it.

 

End game is this: if 3D spatial audio was already a solved problem, we wouldn't be having this conversation.

[Kuky]

And in this 'experiment', what is your control?

To name but one problem with it.

 

- I used impartial person (my wife who isn't into gaming at all didn't know anything what I wanted to do, so her brain couldn't try to compensate or make up things to mess the experiment outcome)

- I eliminated influence of eyesight (to concentrate on hearing only)

- I did not tell my wife where I will be moving the mobile phone (sound source) as if I did, her brain would know where to expect it, hence the third source (the brain expectation or memory) was excluded also.

 

this all means she had to concentrate on her hearing only, and without any movement, to determine where the sound is coming from, and as I expected she couldn't tell when the phone was moved from front to above her head, based on hearing only.

[S3NTRY11]

- I used impartial person (my wife who isn't into gaming at all didn't know anything what I wanted to do, so her brain couldn't try to compensate or make up things to mess the experiment outcome)

- I eliminated influence of eyesight (to concentrate on hearing only)

- I did not tell my wife where I will be moving the mobile phone (sound source) as if I did, her brain would know where to expect it, hence the third source (the brain expectation or memory) was excluded also.

 

this all means she had to concentrate on her hearing only, and without any movement, to determine where the sound is coming from, and as I expected she couldn't tell when the phone was moved from front to above her head, based on hearing only.

And you didn't tell her what you were testing I take it?

 

Either way, that's Mythbuster-level rigour there, so all good, believe whatever you like, I guess.

[S3NTRY11]

I'm with Kuky on this, what he is saying is correct. I do think everyone is actually on the same page but Kuky is right about our ear drums only sensing left and right sound.

 

As we experience the world around us of course our bodies learn to tilt our heads to accomodate this stereo effect to pinpoint sounds. Also frequencies of known sounds paint a mental picture of the location through learned behaviour.

You instantly can visualise a helicopters flight path based on the pitch (high pitch means it's approaching) and you expect it to be above you. Also sounds behind you are less clear.

If I ever experience a loud sound that I am unfamiliar with it takes me a while to work out what it is and where it is coming from. This involves a lot of head tilting/turning and running it through my brains data bank of sound comparison.

 

So apologies, but I'm with Kuky, at the end of the say we are stereo animals, but with a great sonic library to help determine sound source. I believe all this surround experience in the headsets is based on very clever design that attempts to simulate 3D sound, but it relies heavily on the years of sound experience the user already has.

No need to apologise for agreeing with someone.

 

Yes, sound enters the ear from left and right. Most implementations of sound in games operate on a single plane. I don't even know how true it is for DCS to be honest, but I do know that what has been modelled to date in most sound-engines has not been binaural.

[Shmick]

Kuky is right about our ear drums only sensing left and right sound.

 

I'm not totally convinced by this logic (but haven't gone looking for papers or done my research :P), because each ear is made up of a multitude of sensors (hairs) in different positions and with different resonant frequencies

[Kuky]

I'm not totally convinced by this logic (but haven't gone looking for papers or done my research :P), because each ear is made up of a multitude of sensors (hairs) in different positions and with different resonant frequencies

 

hairs are not sensors, ear drum is. Please do the experiment I did and let me know your findings, but make sure you do it properly.

[Shmick]

My understanding of human physiology might need some improving!

[S3NTRY11]

hairs are not sensors, ear drum is. Please do the experiment I did and let me know your findings, but make sure you do it properly.

 

My understanding was that those hairs feed through to the drum?

 

Also, you can't "do it properly", outside of a controlled environment with rigorous methodology.

 

That just sounds like "agree with me, or you're doing it wrong".

Edited January 10, 2016 by S3NTRY11
Consolidation of posts

[S3NTRY11]

.

Edited January 10, 2016 by S3NTRY11
Sorry, needless double-post

[hannibal]

You guys still want to believe otherwise, be my guest, I think I have done enough to prove I am in fact right about this.

 

hard to discuss when you havent used DCS in VR, and a VR demo that has correct spatial audio. its ok brother, we are just discussing.

 

what i will say is that oculus has an "audio SDK" for developers. if audio is what you described as even for VR, then there is no any need for an audio SDK.

[Kuky]

My understanding was that those hairs feed through to the drum?

 

Also, you can't "do it properly", outside of a controlled environment with rigorous methodology.

 

That just sounds like "agree with me, or you're doing it wrong".

 

Well I am no expert on the ear physiology its basically ear drum vibrating, and there's some bones connected to it that vibrate because of it, transfer the sound vibrations via some hair to the nerves or something that ultimately goes to the brain. If you meant the hair on the outside of the ear then no, they are there just as natural defense for dirt to get trapped instead of going into your ear, they don't do anything with sound capture or transmitting it to the brain.

 

And by properly I mean do the experiment by eliminating other sources of information the person should be exposed to, anything beside sound itself, and no movement. That's all there is to it.