I was thinking today about Pilot Edge and the subscription cost...and a thought dawned on me that I'd like to hear opinions from the far more informed than mine..

Pilot Edge really is a nice service, but it is rather expensive in my opinion ($329 a year for the best package).  Pilot Edge uses human controllers with delivers nearly 100% realism including even controller error and hearback errors.  The limitations of course are the cost due to having to pay humans an hourly wage and coverage limitations due to a single human can only control so many areas.  So basically, it is a limited premium service. 

Anyway, here's my question/thought: what is the feasibility of a remote server taking flight information from P3D/XPLANE (speed, altitude, track, heading, etc) and processing the data, then comparing the data with a submitted flight plan, followed by issuing back instructions?  The user would have a light weight client software that sending the data to the server, a program on the server reads the flight data and processes other data like weather, terrain, and airport data, and sends back instructions to the client for the user to follow.  The program on the server would get navigation data from Navigraph or FSaerodata, weather from HiFiSimTech, terrain and obstacles from Orbx for guidance.  These would allow global coverage, realistic procedures, and I imagine more affordable subscriptions.  The software would not be too complicated as there would be no artwork, but pure backend processing and a simple client end user interface.  Both IFR and VFR procedure could be modeled as the premise is still the same.

Seems to be a winning combination....I can't be the first to have thought of this...

So, where is it?!?!?! 

 

This is basically how all ATC addons and the built-in ATC work. Locally, without any need for a server or subscription.

Apart from VATSIM and the likes, that are doing it with actual people and for free.

Quote

The software would not be too complicated

This is where you are wrong I'm afraid. It is exactly as complicated as RC, PF3, ProATC, VoxATC and all the others, and it will have the same shortcomings. On the contrary, running this on a server where not one single user, but maybe thousands of users request their ATC at the same time will be a massive task. Not only to develop it, but also to run a big and powerful enough server to handle all that traffic at the same time. And if you plan on streaming audio from the server, the cost for bandwidth will be significant. Which is probably why PilotEdge have those prices in the first place. A reasonably powerful server with bandwidth to boot can easily cost  $20.000 per month. There is no real limit though, a really really powerful server will set you back $300.000 too - per month.

One aspect that is not to be dismissed IMHO is the content. To be realistic, ATC controllers all should have different voices. But getting those assets spoken by many different people with different accents, and then process them in such a way that you can build any ATC phrase from them, this will cost you. You need a professional sound designer for this, and you need decent actors with good voices.

Best regards

Edited August 22, 20187 yr by Lorby_SI

The price for PilotEdge has very little to do with infrastructure and everything to do with paying controllers to be there. 15x7 coverage....2 coverage areas with 1 of the areas have an extra controller online for some of the time. Use just about any pay rate above minimum wage and you'll see the costs add up very quickly.

In terms of this proposal, the problem is writing a solid voice recognition system, and something which doesn't suck at sequencing. Consider this request, "we'd like the VOR-A, then the ILS back at John Wayne."   There's that, and a nearly infinite number of other ways that you could paralyze a voice recognition system, making requests that you say in real life on literally any IFR flight.