Another great piece of work!

It isn't clear to my why pink noise and multilevel should be separate options: my naive expectation (knowing very little about wind) would be for the wind at each level to be pink noise. So you wouldn't need the two options, you'd just decide whether to add new levels, with mean, direction, and turbulence at each level.

If I'm using Imperial units, switch the wind model to multi-level, and create a new level, the wind speed is 16.4 ft/s (which is 5 m/s). Now when I click on the wind speed to change it, it jumps to 5065617 ft/s. I can now edit the speed as I wish. If I add a second new level and go to change the wind speed, I get no surprises.

It isn't clear to my why pink noise and multilevel should be separate options: my naive expectation (knowing very little about wind) would be for the wind at each level to be pink noise. So you wouldn't need the two options, you'd just decide whether to add new levels, with mean, direction, and turbulence at each level.

Good point! I would however still make a distinction between "Average wind" and "Multi-level wind" in the UI (and internally). I feel like it will be confusing for users if they only have the multi-level wind input. I can have a common section in the beginning to specify the standard deviation. I'm not sure about wind turbulence, since it's calculated using the average wind speed. Maybe I'll include that only in the average wind section.

This is a great thing that many have inquired about.

This is way outside my own domain, though, and I'm curious how it might actually be used. I'm guessing this would be used at the field, based on info gotten from some website (or app)? Or is there any reason to ever set this up in advance? I'm not sure how one would know the conditions to that level of detail in advance.

Does there need to be a way to copy these settings between simulations? Does there need to be a way to import that data from a website in order to be useful?

Or does this somewhat mimic the existing plug-ins that do this? If people are happy with those, and this works similarly, then I would imagine this is good for a first release. But I really have no idea.

It isn't clear to my why pink noise and multilevel should be separate options: my naive expectation (knowing very little about wind) would be for the wind at each level to be pink noise. So you wouldn't need the two options, you'd just decide whether to add new levels, with mean, direction, and turbulence at each level.

Good point! I would however still make a distinction between "Average wind" and "Multi-level wind" in the UI (and internally). I feel like it will be confusing for users if they only have the multi-level wind input. I can have a common section in the beginning to specify the standard deviation. I'm not sure about wind turbulence, since it's calculated using the average wind speed. Maybe I'll include that only in the average wind section.

It seems (to me, anyway; like I said I know very little about wind) that it would make the most sense to have the values at each level include the turbulence; just as currently pink noise comes up with a default wind speed and turbulence, the new model would come up with default values for level 0. A user could just adjust those values, or could add new levels.

I'm not quite sure why we have both turbulence and standard deviation, since they're calculated from each other.

This is a great thing that many have inquired about.

This is way outside my own domain, though, and I'm curious how it might actually be used. I'm guessing this would be used at the field, based on info gotten from some website (or app)? Or is there any reason to ever set this up in advance? I'm not sure how one would know the conditions to that level of detail in advance.

Does there need to be a way to copy these settings between simulations? Does there need to be a way to import that data from a website in order to be useful?

Or does this somewhat mimic the existing plug-ins that do this? If people are happy with those, and this works similarly, then I would imagine this is good for a first release. But I really have no idea.

I frequently see people launching over around 10K feet checking the winds aloft before deciding whether they actually want to launch that day. The way I see this being used is somebody, morning of launch, before heading to the field, going to a site like windy.com to pick up winds at several levels to run a simulation.

What would be really cool would be going to a weather site with an API (I'm told there are many of them) and grabbing wind conditions at multiple altitudes from there. But no need to do that on this release.

I frequently see people launching over around 10K feet checking the winds aloft before deciding whether they actually want to launch that day. The way I see this being used is somebody, morning of launch, before heading to the field, going to a site like windy.com to pick up winds at several levels to run a simulation.

What would be really cool would be going to a weather site with an API (I'm told there are many of them) and grabbing wind conditions at multiple altitudes from there. But no need to do that on this release.

Fair enough.

It seems (to me, anyway; like I said I know very little about wind) that it would make the most sense to have the values at each level include the turbulence; just as currently pink noise comes up with a default wind speed and turbulence, the new model would come up with default values for level 0. A user could just adjust those values, or could add new levels.

Sure, I can make it configurable for each entry. I hope I can get it all to fit in the UI though :)

What would be really cool would be going to a weather site with an API (I'm told there are many of them) and grabbing wind conditions at multiple altitudes from there. But no need to do that on this release.

Yup, this PR was sort of an entry point for that feature. Once we have this back-end, it's "just" a matter of fetching the right values and plugging them in. A lot of users have asked for that feature, would be nice to have it included.

Or does this somewhat mimic the existing plug-ins that do this? If people are happy with those, and this works similarly, then I would imagine this is good for a first release. But I really have no idea.

My initial intent was to have native support for the existing MultiLevelWind plugin, since a lot of people seem to like it. And I'm tired of users stating that the plugin does not work with OR > 15.03, because the plugin was never officially updated.

Updated behavior:

Very nice. I still think the "average" model should be replaced with just a multi-level model with a single level, but that's not crucial.