Continuing positive impressions and feedback

[c.b.powell]

Today I planned and flew an IFR flight from San Diego up to Palm Springs -- and had an absolutely delightful time. For me, PilotEdge's greatest benefit is to give me a safe place to practice IFR work and to increase my skills. I can make my mistakes in the sim instead of the real world.

The Cardinal is a fast, complex airplane that's easy to get behind, and early in the flight I had a course deviation -- and immediately got called on it and scolded. But I should emphasize: this is exactly the way it should be. Real life controllers wouldn't forgive that, nor should PilotEdge controllers. I think PE can provide that level of realism and authenticity without sacrificing the enjoyability of the experience; in fact, I think the realism augments the enjoyability. So thank you. Every error I make now is cheap; errors aloft may not turn out to be. Stay strict!

I also had a rather more extreme & humorous deviation when I shed a wing due to a misconfigured V-speed limit. I had checked the "remove flying surfaces in over-speed" option under Operations & Warnings -- now unchecked. Keith, thanks for your understanding on that and letting me jump back in on the airway where I was previously. I rather hope to lose no more flying surfaces en route.

By the way, a minor point: real-world weather synchronization seems to be a bit of a recurring problem. I had real-world weather activated in XP, yet somehow ATIS/AWOS winds and altimeter settings failed to match what ATC had. I'm unsure of why this is or how to correct it. It's just a tad embarrassing to request runway X because of what I think the winds are, and to be advised that runway Y is in use. Not a big deal, just mentioning.

Thanks for a great flight! I'm having an immense amount of fun with PilotEdge.
Chris P.

[Daddy O]

We had a similar thread going somewhere, talking about the difference between altimeter settings online and within your own aircraft. Sometimes the reported barrometer setting would make your indicated altitude off by a couple hundred feet.

Flying with "remove flying surfaces" enabled is a good thing. Your craft may not have had a misconfigured vmax setting. The heavier you are, the lower your vmax speed. With this option enabled also gives you more realistic damage on bad landings. I like to enable all the hard stuff like birds, vmax, real weather, time of day, etc. If you can fly the sim at max difficulty then you should be ready to handle a c152 on a bright sunny day.

[Keith Smith]

I will write more later, but wings should not fall off in normal maneuvering, regardless of weight. Sounds like misconfiguration to me, unrelated to his handling.

[c.b.powell]

I will write more later, but wings should not fall off in normal maneuvering, regardless of weight. Sounds like misconfiguration to me, unrelated to his handling.

Just to follow up, the problem was indeed a misconfiguration in Plane Maker. The negative-G limit was mis-set to 0.00; the first nose-over of any force was guaranteed to damage the structure.

I grasp Daddy O's points regarding realism. I think it depends on what you want out of the sim. My goals are sharpening my IFR skillset and gaining the associated boost in IFR confidence, with the subgoal of flying some planes I don't have ready access to.

Chris P.

[Daddy O]

If the neg G limit was set to 0.0 then it was definately a misconfig. But I have torn a wing off the stock PA46 with surprisingly little maneuvering. I was flying at the top of the yellow, a few pounds under max, I dropped a chart down by my feet. When I bent over to get it, my shoulder hit the yoke, and snap, off it went and I got to ride a spiral all the way to the ground somewhere in northern Arizona.

Note to self: Don't do that.

[Keith Smith]

Hi Chris,

I'm so glad you're enjoying the flights on the network. There's a growing cadre of student pilots, instrument students, and certificated pilots who are using the system the same way you are, to work on their procedures and practice flying 'in the system'. We also have a good number of online pilots who are doing the same thing. I've now seen at least 7-8 cases where pilots would've handled an offline flight perfectly , but the additional work of doing it online caused some kind of an issue (of varying severity). In all cases, the pilots have reported that it was beneficial to encounter that situation here, rather than in the air. In fact, it's why they're here to begin with. So, all in all, you're on the right track, and I heartily encourage you to keep at it.

The system will only get busier in the future, which will be good for everyone. Just tonight, in fact, we had someone put into a hold at the IAF for the VOR-A into EMT while we awaited the cancellation of the aircraft ahead of him. It's just not something that you have to deal with when flying offline. Then, just as he was about to enter the hold, the previous aircraft canceled, and I was able to clear him for the approach there and then. Even if the pilot was creative enough to give himself the hold in an offline flight, it's unlikely he'd change things up on himself and then simulated receiving the app clearance at the last minute. The online environment is just so dynamic (just like the real world), it really is a great proxy for doing it in an airplane for real.

Regarding the weather, I'm still not entirely sure how XP9's real world weather engine works (when you tell it to download r/w weather). I'll write to Austin to try to get the scoop. I do know for sure that it can only load one weather system at any given time (in terms of the actual weather you see when you look out the window). As to what the ATIS/AWOS reports when you dial it in from some distance, I'm not sure. That's the part I'll find out from Austin.

Mercifully, this should all become very natural when V10 comes out with its vastly-improved weather engine.

One small note, though, I don't remember scolding you (if this was regarding the initial heading) But in any case, you're right, it's much better for it to happen here rather than in the r/w, without question. For that reason, I've made sure that our controllers are sticklers for detail, particularly when working with those who we know to be r/w pilots.

Keith

[Daddy O]

The weather thing is kinda wierd. During the demo today at LGB the reported winds were light or 5mph. But on my system the windsock was indicating a strong crosswind, and I kept some steady rudder pressure throughout the day. Maybe you just need to synchronize the weather being downloaded between the aircraft and the tower. Maybe add a note that pilots should have their weather set to refresh every X minutes so everyone is on the same sheet of paper.

[c.b.powell]

I've been looking into X-Plane's "real-world weather", trying to find the source of this curious disconnect. I learned that X-Plane gets its data from http://weather.noaa.gov/pub/data/observ ... ar/cycles/ . But it seems that X-Plane isn't properly pulling the most current metar data.

For example, I'm writing this at 19:06 PST / 03:06 UTC. At present moment the newest file on the above noaa.gov site above is timestamped 03:02 UTC. X-Plane is set to the correct zulu time in Environment > Date & Time. It *should* be pulling the 03Z.txt file but I don't think it is, because the ATIS at KPAE is announcing "01:00 Zulu weather". I think that while PilotEdge controllers correctly use the most current time, X-Plane's download mechanism is flawed and pulling down one- or two-hour old data.

(This may be common knowledge...sorry if I'm just restating a known thing.)

[c.b.powell]

OK, here's a quick hack that I think may neatly solve this - at least if you're on a Mac or Linux box. The hack leverages the cron facility on these machines to bypass X-Plane's metar download and do it ourselves at the OS level, ensuring that we get the newest data.

Requirements: a Unix-like system with cron daemon; curl (or wget); date.

Step 1:
Edit your crontab file ("crontab -e") to set up a line as follows:

*/5 * * * * curl `date -u "+http://weather.noaa.gov/pub/data/observ ... s/\%HZ.TXT"` -o /Applications/X-Plane\ 9/METAR.rwx

You'll note that I employ curl but you could also make this work with wget. It uses the 'date' command to figure out what the UTC hour is right now, and pulls down that file. It writes that file out to the X-Plane directory where X-Plane looks for real-world weather. This command runs with a frequency of every 5 minutes.

Obviously, adjust the X-Plane path to match your system.

Step 2:
In X-Plane > Environment > Weather, turn OFF "Download Real-Weather" but turn ON "Scan Real-Weather file". I set the granularity to every 1 minute. This configures X-Plane to evaluate the data we regularly pull down in Step 1.

Step 3:
Hopefully, this means that pilots will have WX that closely matches what the PilotEdge controllers have. I've not verified that fact yet in-flight but I'm eager to hear if this does the trick.

By the way, an interesting point: I observe that X-Plane *does* correct the METAR wind direction data from true (as delivered in the metar) to magnetic (as the tower would tell you in an ATIS recording). That's rather nice.

Another observation: in a given hourly NOAA metar file you'll find a number of entries for a given station. For example, KPAE has 20 entries in the file I just pulled down. It seems that X-Plane sifts through these and tries to grab the most recent one. At least, I hope it does.

[Keith Smith]

This is a copy of the message that I sent to the Laminar team earlier this evening. I'll let you know what I hear back:

If you enable real weather download in V9, as well as the option to scan the file periodically, can you explain the following:
1) if you tune the published AWOS/ATIS frequency for a given airport what weather will be reported to you, is it the weather for that airport as reported by the downloaded weather, OR is it the weather that's currently loaded in the sim for the local area?

2) what mechanism does XP use to determine the 'current' weather conditions? I know it can only load one weather system at a time (in V9), so the question is, which weather is it loading at any given time? Is it using the nearest airport with weather reporting, or is it blending several nearby airports?

END OF MESSAGE.

I have a feeling that the issue is not the fact that the weather may be 1-2 hours old, but is an artifact of something else.

I've eluded to this in a few other messages, but PE was designed with XP V10 in mind, which is going to have a radically different weather engine. V10 was delayed, however, so we're somewhat stuck with V9's weather capability for now. I'm trying to better understand exactly how V9 works.

The PE server pulls weather down from the same weather source as X-Plane. I recall reading the documentation on the site explaining how those files are built. My recollection is hazy, but I think their recommendation was, in fact, to pull down weather that was 1-2 hours old because the 'most recent' data might be incomplete (due to the nature of weather collection/reporting around the world). The only way to get a complete data set was to use something other than the 'current' weather.

Based on pilot reports, however, people are experiencing weather conditions that are not likely to be the result of data that's 1-2 hours old, but rather, the result of xplane using weather from a different airport, or perhaps a blending of several nearby airports.

Once we have answers to some of these questions, it should clear this up.