Why we stand so far apart & why audio chat needs to be adjusted


#1

Yes, it could be we’re all introverted and shy geeks, but I think the real answer is (and all the prose previous to and including this parenthetical remark is just to appease the minimum character threshold) is Field of View. In RL we can stand much closer to each other because we have a wide angle field of view, 180 degrees horizontal, 114 degrees binocular (https://en.wikipedia.org/wiki/Field_of_view). The horizontal FOV of 1st person screen view is approximately 90 degrees. The best case FOV of the Oculus CV1 and HTC Vive is 110 degrees, so that means a lot of uncomfortable head turning has to happen when people crowd together, hence we move further apart (almost 4x the usual distance) to get everyone in our tech limited field of view.

Now this is easy enough to do (backing away), but it means that for similar VR outdoor hangout scenarios mimicking an RL gathering, we tend to be 8-10m apart instead of 2m. Again that is not a problem visually, but it is a problem with audio chat. What happens with the standard audio attenuation is that voices will seem much quieter and the attenuation of the individuals in our crowd is accentuated because of the larger distance ratios of near/far. This can be fixed by reducing the audio rolloff (aka domain attenuation) parameter. It is presently 0.18 and I think the default needs to be 0.045.

I recommended this long ago for the sandbox and it stuck for a while before it got set back. This post is just me articulating the reason why it wasn’t just an idle request. I made my domains use 0.045 attenuation as the default. For inside spaces, I create zones and lower the attenuation even more, down to 0.02 because meetings in walled rooms create a near field effect.


#2

Wonder what it would take to let individuals adjust that at their whim in Interface. I.e. override the domain’s settings for that for themselves alone.


#3

https://alphas.highfidelity.io/t/voice-volume-voice-low/10524


#4

I was thinking of your problem, Richardus, when I wrote my post. We should try having a conversation in a domain where the attenuation can be lowered to 0.045 and then we can determine if doing that works around the low volume problem.


#5

I think any adjustments to sound on a domain should be done on the interface (viewer) level, in other words the user should be able to adjust sound levels that they want but, for the sounds that are sent from the audio mixer, control should be up to the owner of the domain.

I am all for being able to set the attenuation to what ever a domain owner feels is right. But what I do not want here is settings in Interface (viewer/client) that override this.

And the same goes for gamma/lighting and even view settings like camera/1st/HMD/3d person field of view, fly/no fly, tp etc.

If you don’t want to or can not experience the domain/build the way the person who made it intened, then maybe it would be best to not experience it at all.


#6

This is an important thing that Twa_Hinkle brought up and that is that the mix and the effects are artistic statements. There may be a reason why a domain or a zone in a domain is murky or over staturated or whatever. The scene creator wants it that way to set up a specific experience.

That said, the unadorned attenuation default is off and it shows off voice chat not as well as it could be. People trying out their own first created domain should have a decent experience.

Also the overall sound level is something that has no endpoint absolute value, so it needs to be adjustable. There really needs to be an overall volume control in interface; that, or interface needs to run +12dB higher than it does today. Only then will there be enough dynamic range to turn it down in a system mixer.


#7

Not agree complete, the end user need to have cntrol on the sound and voice volume as seperate control.

If high fidelity is getting build by amateur creators like secondlife. There will be many volume setting wrong on domain.

Simple reason that many not have good hardware or knowledge to adjust everything correct. So when it sounds good for you, the lther is not hearing anything or get the ears blow off.

In my case with high fidelity voice is way to soft, and sounds blow my ears off. Both sre so out of balance now that i never can fix it. sound is supressing voice only more if the play both.

I do not hope that high fidelity is for professionals only, because then it could be better with audio.


#8

HF seems to be targetted for professionals and large business game developers and virtual world creation companies, so yes, they are the ones who set things up. Nonetheless, I recall these same process issues when Windlight was introduced in SL. Region designers were appalled that the end user had total control over the look of their region creations. Likewise, RP designers were appalled that their players could alter the audio rolloff and so take advantage of listening in to their ‘war-enemies’ over the virtual hill. In simpler words, it broke the game dynamics.

So there are definite reasons to lock down the look and feel. Conversely, open areas should be adjustable by the client for the reasons you mentioned, Richardus. This problem speaks to a larger XUI/UI deficiency and lack of architectural depth. Domains (more specifically, zones) need to be abe to set operating conditions for audio, visual FX, for capabilities and privileges, and whether these settings are locked or not. Zones have a few settings, not nearly as rich a set as there should be. Lockouts are generally not enforcable because there are external means to modify brightness, gamma, color balance, audio levels, albeit it, those means are messy at best when done through OS system controls.

But, to get to my point, I was adressing one specific issue as to the reason why our voice chat seems so low and spotty, and offering a spot solution that did work well in the past and a suggestion to add a general audio level setting to interface.