Audio quality

[Mike Sweeney]

Appears voice technology concerns may vary by member for pilot-ATC communications.

US with New Zealand and US with South Africa = no voice quality or noticeable lag issues.

(sample size= 7,000 hours pilot; 3,000 hrs as ATC)

... voice technology concerns may not be across-the-board if some members remain unaffected.

Edited August 4, 2018 at 12:54 AM by Guest

[Andreas Fuchs]

I do want to echo Mike's observations. Yes, I have observed voice-issues, but they were not caused by bad connections, but rather by faulty setups (wrong device, microphone too far/close to mouth).

[Oliver Gruetzmann]

I do want to echo Mike's observations. Yes, I have observed voice-issues, but they were not caused by bad connections, but rather by faulty setups (wrong device, microphone too far/close to mouth).

Sure, it's usually something like that.

My main concern with the current codec is, that it's not tolerant regarding setup errors. Put your mic too far away? You'll drown out in some kind of noise. Put it too close or set the volume too high? Distorted voice.

 

Even the voice calibration seems strange theese days. Put a slider / meter combination or (if it really needs to be there) AGC.

 

And from when I started in 2006 I always had occasional dropouts in voice, which seem to come from the VVL itself. Have never, in all those years, been able to get rid of those using several different PCs and notebooks.

 

I know how to get it right, but I'm sure not everyone does. Lots of room for errors, with a tiny littly spot labelled "perfect".

[Scott Bickford]

I wonder if clients are still based on old RogerWilco, which uses UDP. UDP is transmission is not guaranteed where tcp is guaranteed. This is probably why a lot of people say, when I use Skype it works perfectly, but this I get breakups etc.. (and the other ATC network is perfect to the point of not being "real", but this one I get breakups-too real, but too much).

 

RW was the frontrunner in voice with gaming. https://en.wikipedia.org/wiki/Roger_Wilco_(software)

[Kieran Samuel Cross]

I wonder if clients are still based on old RogerWilco

 

Yes

[Trevor Hannant]

 

It is being developed as we speak - someone is currently actively working on it

[Daniel Morris 1132365]

Literally the voice codec is the only thing that puts me off flying in VATSIM,

if you have to cut older interfaces like squawkbox and leave behind fs9 users - fine, it's about time they updated from a sim made in 2003. There are probably more people willing to return given a new codec vs the amount lost from moving to a new one...

edit/p.s.: just checked up on the MELP codec that vatsim uses, it was developed in 1995 just for reference

 

"Something is being done"

"It's in development"

from one corner of the room for about 6 years with nothing to show

(source of this same conversation with another user)

 

"there's nothing wrong with it"

"It's your hardware, this is fine"

from another corner

[Matthew Cianfarani]

Daniel,

 

I'll reiterate what I said in my previous post that you quoted. A large amount of work is underway. We cannot make things happen overnight, especially as we are all volunteers.

 

The concerns you hold are heard - And I am sure you will be pleased with the result.

[David Zhong]

UDP has nothing to do with the issues. All state of the art voice protocols run off UDP including Skype. "Reliable" in the context of what TCP provides is not necessarily desired in real-time applications.

[Nicholas Cavacini]

As previously stated, UDP is preferred for (almost) all real and near real time applications because while TCP guarantees delivery and will sequence packets in order, in real time applications, it makes no sense to have the overhead that TCP has and it makes even less sense to have a packet arrive late as that section of the transmission is now in the past. You can build reliability on top of UDP at a higher OSI level but again, you'd be adding latency and overhead.

[Daniel Morris 1132365]

The concerns you hold are heard - And I am sure you will be pleased with the result.

 

I come from a medium sized volunteer organization so I understand this, Rome wasn't built in a day; I accept that, looking at the history of squawkbox 3: it took 3 months from start to the first flying test. Vatsim's been stagnant for yearS with moving forward from this

 

My frustration comes for the love of flying online, the concept of vatsim is awesome, I'm more p[Mod - Happy Thoughts]ionate that I want to see vatsim do well!

 

Is there anything you can show us? tell us? interact with the community with dev blogs, videos. Let's hear the struggle and we might stand with you

[Oliver Gruetzmann]

I wonder if clients are still based on old RogerWilco, which uses UDP. UDP is transmission is not guaranteed where tcp is guaranteed. This is probably why a lot of people say, when I use Skype it works perfectly, but this I get breakups etc.. (and the other ATC network is perfect to the point of not being "real", but this one I get breakups-too real, but too much).

I think those breakups are related to the way the VATSIM implementation (or the codec) is buffering the audio. You hear those breakups even when playing an automated Euroscope-ATIS locally - exactly at the same time. This happens if the ATIS reaches a specific length, think it is around 60 seconds. And if I'm not totally wrong, there was a similar effect in VRC (don't have it installed currently, so I can't test )

 

I'm not 100% sure, but those sound exactly like the breakups on frequency. Most likely this comes from the codec, not the transport.