Future airport db update

[joefremont]

Hi Flyers

It has often come up, why are you using this old code for an airport and for a very long time my answer has been:

We understand that the ICAO codes for many airports have changed but the rule we use is that if an airport exists in FSX we will use its code for that airport, no exceptions . This is because that is the world most pilots on our system fly in, even the latest P3D uses those same codes. When the majority of pilots on our system use a sim with updated codes we will revisit this rule.

Is that still true? So I did a quick analysis of flights done in the last 30 days and the results of that are as follows:

By Flight:

FS2004 9.3%
FSX 30.8%
P3Dv3 2.9%
P3Dv4 27.8%
X-Plane 29.1%

By Pilot

FS2004 7.2%
FSX 27.5%
P3Dv3 3.0%
P3Dv4 26.9%
X-Plane 34.8%

And if you were wonder, average number of flights per pilot

FS2004 17.7
FSX 15.3
P3Dv3 13.2
P3Dv4 14.1
X-Plane 11.4

This means the number of pilots who are flying in a world as defined by old FSX codes is 64.6%, still clearly a majority, but the world is about to change. Microsoft is working on a new flight sim, (FS2020) that will have updated airport codes, Lockheed has just announced that P3D Version 5 will be released next week and the announcement said it would have:

- Fully updated globe with the latest airport and geographic data including sloped runways.
- Updated over 24,000 airports with new data across the entire globe

Read more about P3D5 here https://prepar3d.com/home-slider/2020/04/115906/

So, if you assume those 26.9% of pilots who use P3D4 will switch to P3D5 that would give 61.6% of pilots flying in a world with new airports.

So the question is, is it time to update the airports?

Your opinions please.

[AMV228]

My sense is that P3Dv4's 64-bit, was compelling for many with the previous ever-increasing oom experiences. Not sure that even dx12 will be quite that powerful a draw. And, with the new MS offering on the horizon, along with the uncertainty of real-world economy, I would guess there will be some tendency to hesitate before making an immediate jump.

That said, in my view it is a good time to prepare and design.
My impression is that the database structure is somewhat "de-normalized," and it might be worthwhile to consider a little move towards normalization in this area. In practice, that would mean splitting off the codes into a separate table:

• Airport table would have a unique airport_id number as primary index, maybe also a field for "original code" to identify only for management/maintenance purposes.

• Airport_Code table would have code+airport_id as primary index. Would thus allow multiple codes for each airport.

All airport lookups by code would then reference through a join between Airport_Code+Airport tables. Would remove the "Why are we still using the old code?" questions and choosing between simulators. (Or, at least make dealing with the issue much easier.)

(I do appreciate the amount of work that might be needed to make this change. So, just an offering.)

[zeke78]

Well, my humble opinion is yes, time to slowly but surely move forward. I don't have access to the statistics as Joe does, but I'm guessing more people at some point will transition and it'll open new possibilities for flights as well.

Joe, as it ever been considering having multiple airport databases? Say a pilot, more likely the VA, could have the option of choosing which database to choose from, FSX or P3D. Obviously that would limit its pilots to what sim that would have to use, but it would be an option.

Rgds,

Tiago

[Cat]

I keep my own "table" as it were as Active Sky Next weather data uses current airport codes. One way older systems like mine can be used with newer codes is the simple use of fsxmap.com. It's a great tool for showing every default airport in FSX. All you have to do is type in the city name and it will list all the airports there. So if/when FSAirlines does "flip the switch" to current airport codes (which by the way is a HUGE undertaking), older systems using FSX still have this reference guide:

https://fsxmap.com/

[joefremont]

Now that MSFS has been out for a bit I ran the numbers again, so for the last 30 days.

FS2004 7.4%
FSX 22.2%
P3Dv3 1.2%
P3Dv4 12.4%
P3Dv5 8.71%
X-Plane 19.0%
MSFS 29.1%

Total with Updated sim: 56.8%.

I have always said I would update the codes when a majority of pilots are using an updated sim I would update the codes and over the last 30 days about 57% have been using sims with updated code, which I define as P3D5, XP10, XP11 and MSFS(FS2020), but to make sure this is not just a flash in the pan I will wait till the 60 day average is above that and that number is 48%.

I still don't have the airport extract from MSFS yet but I do have the extract from P3D5 which of the airport lists I have access to is the most complete. I will try to make a tool that can show the differences in codes.

Interesting side notes in terms of usage, comparing the September to January, as a percent of total flights P3D is down 38%, Xplane is down 42%, FSX is down 25%, but the total number of flights is higher for September so in absolute numbers, P3D is down 15%, XPlane is down 25% and FSX is actually up 2%. I sort of expected that MSFS would impact P3D users more than FSX users, P3D users are more willing to try new things than FSX users so I would expect more of them to jump to MSFS, but it seams that more X-Plane users switch that P3D users.

[joefremont]

On the idea of using the P3Dv5 codes as a starting point I created a page that will show you what the changes would be if we were to make the update.

On the website from your VA overview page you will see an option on the right "Airport Update 2020" click on that it will show you a list of all airports whose codes would change. If your VA has routes departing from that airport they will show up on the top of the table in red while airports you have no departing flights would show below.

If we do this change, airports that exist in FSX that have codes that conflict with our standard code would have a -10 suffix.

I will probably setup a poll asking the members if we should make this update or not.

[BVU-951]

Please update to MSFS asap
Thanks
Alan

[joefremont]

A couple new pieces of information.

First I now have the airport data from MSFS! Pete Dowson released a new version of MakeRwys that can read MSFS so we now have the data and I have imported it into the database.

Second we now have two months in a row were we have more pilots using sims that have updated codes than use the old FSX codes so I think the time to upgrade the airport database has come.

I have updated the "Airport Update 2020" page to reflect using MSFS codes rather than P3Dv5 codes, I have also created a "Airport Code Differences" page, linked to both from the VA overview page and the Database -> Airports page where you can select a simulator version and it will show you the airports that have different codes, with the airports that your VA has routes to on the top in red.

[BVU-951]

Joe I notice that you think the new code for Sywell is EGII , as Sywell is my local airfield I have to tell you that the code is still EGBK, EGII is a small airfield at Pitsford .
I can confirm that Sywell is still EGBK on MSFS.
Alan

[joefremont]

Joe I notice that you think the new code for Sywell is EGII , as Sywell is my local airfield I have to tell you that the code is still EGBK, EGII is a small airfield at Pitsford .
I can confirm that Sywell is still EGBK on MSFS.
Alan

Ok, I will double check things and see what went wrong.

[joefremont]

Joe I notice that you think the new code for Sywell is , as Sywell is my local airfield I have to tell you that the code is still EGBK, EGII is a small airfield at Pitsford .
I can confirm that Sywell is still EGBK on MSFS.
Alan

The issue here was one of false reporting. My merging program sees two airports less than 3 miles apart and thinks they are the same airport. so it though EGBK was both EGBK and EGII and my report saw that MSFS though EGBK was also called EGII in MSFS and though the name was changing, I fixed the report but need to see if I can not have it think EGII is EGBK.