I’ve been on a personal quest to figure out what’s going on with the weather in career mode. If you don’t want to read this whole thing, it’s safe to say the current career weather system is bad for so many reasons.
First of all, the time selector in career mode is pretty clunky UI, as are the infamous red dots that appear on the mission icons. The only time they are meaningful is when the time slider is set to the mission time. This will only happen if it is daytime in the area in which you are flying, because there are no night flights (I assume because many airports don’t have lights and it’d be a major hassle). UPDATE: this was edited as I found some meaning to the time slider/red-dot UI, explained in a later post below.
I have flown dozens of missions and written and taken screenshots of my observations. In every case I’ve observed, no matter where the time slider is, no matter if a red dot is present or not, it is loading the historic weather that matches the set-place mission departure time, sometime in the past 24 hours. I found that the in-sim weather matched the historical weather with greater than 90% accuracy, and I’d expect some give and take for interpolation, interference, or lack of resolution (which is fine because weather is dynamic, anyway). When the slider is set to “live” and has a red-dot, it’s only because the preset mission time is also happening right now, so the weather will coincidentally be real-time. The live/red-dot combination is basically only an indicator that it will use live, current weather, not a determinant.
The weird thing about the slider/red dot is that the dot will appear whenever the slider is set at the mission start time out to about 2 hours later, so it will modulate the mission start time to the slider, within an hour or so window, as long as the dot is there. But this seems to be pretty meaningless for weather - it’s still going to load it within that window. And sometimes the dot randomly disappears as you move the slider. But other than that, the entire presentation of the dot is useless. (Update - slightly less useless, as explained in a later post).
In an operational sense, the sim is not giving you enough weather information. Here’s what they give you in the briefing screen:
- The categorical conditions (VFR/MVFR/IFR/LIFR) at the airport pair
- Departure and arrival times in local time (not UTC - we’ll come back to that)
- A weather icon for departure and arrive that tells you if it’s sunny or not
- The visibility in NM (should be SM)
- The temperature in °F (should be °C in aviation, always)
What’s missing:
- Everything else, most notably forecast wind at the airports, and the entirety of the enroute weather.
The implications of this are obvious at face value - people are launching into very poor weather (see wind) without being informed. But the kicker is that it’s really hard to know what you’re getting into, especially in some parts of the world - it takes a lot of work and understanding that the sim isn’t providing, during the boreal winter, which is a really tough time of year to fly for many of us. (Update - it actually is providing a big picture as discussed below (though still kind of clunky and not enough info, but good for a start)).
Preflight planning is critical. It involves aircraft performance, hazardous weather and terrain, usability of airports, fuel and payload to name a few. We’ve established the sim alone doesn’t give you enough information. Categorical conditions and temps, when reliable, are only a small part of the picture. What’s the number one complaint right now? Wind. And suffice it to say there’s a lot more that you’re not getting in that quick glance, at least not enough to make a good “go” decision.
But more broadly, when faced with using realistic weather, when a person starts to look at missions, a reasonable approach is to choose an area to “work.” By this, I mean taking a broad overview look at a weather map and deciding “looks pretty good in this area, I’ll fly here.”
But remember above, where I said the missions only use the historic weather at the scheduled mission start time? This means you might have to check the historic weather for an hour to 24 prior, which is obviously dynamic, always changing for each individual mission. And the mission planner is not filterable and sortable, so when you do finally find a mission that looks decent, you have to look at the mission start time, convert it to UTC (which takes time and understanding), then go into archive mode at your weather resource, if it has it (I use aviationweather.gov and look at the historic weather for that time.
This might be a fairly easy task for a singular mission, especially if both airports have a nearby METAR and TAF (they often do not). You may get lucky and it happens to work. But say you get 15 minutes into checking the weather and discover it’s a no-go. Maybe you discover that the winds aloft are crazy headwinds and the already-scant fuel you’re given just isn’t going to cut it, or there are low ceilings at a non-instrument airport, etc. If that happens, you have now have wasted that entire time and have to pick another airport pair and start planning from scratch, likely at a completely different time. It completely negates the idea of getting one weather overview based on one set time (be it live or otherwise) and making a decision. Getting multiple “big-picture” overviews and drilling down to multiple specific historic weather instances for multiple airport pairs, for different airplanes, is absolutely frustrating for any user.
This leads to one of three decision-making scenarios:
- A pilot gathers some historical weather information for the scheduled mission time using outside resources, it looks decent, and it’s a go - they fly the mission, hopefully successfully
- A pilot jumps into the mission fairly blindly - just using what the sim gave them. Maybe they see the weather is awful during preflight and choose to abort. Maybe they takeoff - that’s up to them. If they do fly, maybe it works out all the way to the end, maybe it doesn’t. This requires a lot of blind luck to pull off successfully.
- A pilot gathers some historical weather information for that time, but it looks poor and they decide not to go.
Scenario 1 will lead to the best-case outcomes, but it takes understanding, work, and a lot of luck to not spend a ton of time searching and (re)planning.
Scenario 2 has the biggest opportunity for worst-case outcomes and is by far what most people seem to experience and get frustrated with. They’re rarely getting the current weather, especially if they’re flying at night - they’re getting what it was. Finding a complete, useable resource for that is difficult enough in the US, but nearly impossible in some other places. But understanding all the things that are important to understand, well, the sim isn’t teaching anybody that.
Scenario 3 is a cluster. After aborting and going back out to the menu, the pilot has to select and vet an entirely different mission that is most likely specified at an entirely different time and place. So all the weather they just looked at in the no-go mission is moot - they are starting over from scratch. And they haven’t even flown yet. This is incredibly time-consuming and, to reiterate, very frustrating.
In the end, it would be better to give us the choice to use one of three options:
- Using current, live weather, no matter the scheduled mission time
- Using the historical weather within the past 24h at the mission start time, as it is now, but again - completely optional (this so you’re not stuck with only poor “live” weather)
- Using a weather preset (including clear or custom), which could arguably incur an easy-mode normalization of credits earned
Again, I have a feeling we’re locked into using specific time ranges because they don’t want to let us fly at night for several reasons. So many of us will never truly use real-time “live” weather when flying in our home regions. But since “live” weather is actually historical weather locked to the mission time anyway, we’re stuck using it and trying to figure out what it is giving us.
I would love if somebody would replicate the results of my findings. I’d hate for you to just take my word for it and I want to know if I missed something. Here’s a snippet of my methodology if you want to replicate:
- Choose a career-mode mission within the US, preferably between an airport pair that has an AWOS/ATIS or uses METAR. It also is nice if you use a plane with a G1000 or similar, or have a tool you can use to read winds aloft.
- Take note of the mission start time and convert it to UTC
- In a browser window, set up at least two instances of aviationweather.gov (avwx) - one set to live time, the other set to the archive view at the scheduled mission start time, converted to UTC, within the past 24h. If you set the sim time slider to anything other than “live” (far right), open a third instance of avwx set to the archive at that UTC time.
- Launch into the mission and pull up the AWOS (or whatever other tool) to get a snapshot of the current in-sim weather. Take note of the in-sim winds, altimeter, temp/dewpoint should be enough. Compare this to “observations” on the METARs at the various avwx instances
- When you’re ready to depart, set the avwx instances to wind aloft forecasts - make sure you set the slider to the nearest altitude above the airport. As you climb through those altitudes, note the winds aloft in the sim (using the cockpit indicator or whatever tool) and compare to each of the avwx instances. Do this again at 3,000’ intervals until you get to cruise.
- Do this again at 3000’ intervals during descent. Make sure you change the time slider in the avwx instance of historical weather to reflect the current mission time (derived from the aircraft clock, make sure to use UTC) along the way. If you don’t, you’re looking at only the weather at departure time. The “live” avwx instance will continue to update automatically as you go.
- Then compare the weather at the arrival airport to the latest METAR in each avwx instance (remember to set the time slider in the “historical” avwx instance to the current mission time)
Notes: this works for the US, it works better in more isolated areas where multiple METARs don’t interfere with another, causing interpolation, and it does to an extent rely on those METAR for verifiability - it’s hard to pin down other kinds of forecasts with reliability, but comparing METAR made it very clear. It also doesn’t take into account any bugs, it doesn’t account for whether this existed prior to patch 3 or if it was introduced then, and cloud cover was not part of the study.
A couple other things I may have discovered:
- The winds aloft may not be converting properly to/from magnetic and true. The headings shown in the sim, as verified by the in-sim indicator and LittleNavMap are always off by about the amount of local magnetic variation compared to the winds aloft headings on avwx (given in true, and I’m doing the conversion - it’s almost like it’s applied backwards). This could just be spatial/temporal interpolation, and again, a little variation is okay. But it’s pretty consistently off the same amplitude and direction, which is raising a red flag. Maybe I’m just lucky. More testing is needed.
- Possibly related: the wind indicator in some (maybe all) aircraft is not displaying wind properly. For instance, last night I was flying a mission in the X-Cub and the wind indicator said winds at 144° magnetic, but when I put the nose of the aircraft to 144 magnetic, the indicator arrow showed a crosswind component. It did not show straight winds until I flew a heading of 133, a difference of 11° (it was a little gusty, so hard to lock on exactly). The local mag variation was about 10°E. More testing also needed.
- There’s also one more thing that needs more testing: I think the ATC/AI may be making routing and runway decisions based on real-time live weather rather than historical, no matter what anything is set at. It’s a hypothesis at this point, but I’ve noticed a few times where the historical weather in the sim correctly has the wind howling down the runway for the mission time, but it wants me to take off with that as a tailwind. Current real-world live weather showed conditions that favor the runway the ATC/AI expects (the storm moved on, for instance), but of course did not match historical weather.
Anyway, good luck. Hopefully we can affect some change for the better!
UPDATE: I found some meaning to the time slider/red-dot UI, explained in a later post below.
in the system that need to be reported in the “Bug Reporting” thread. Perhaps the excellent moderators will pick up on this and move it to there, but I’m not holding my breath for that to happen… 
