I like the clouds I see but :
Anybody notice an aircraft performance drop at altitude compared with MSFS?
With AS the CJ4 cruises at considerably lower speed than with an MSFS live weather atmosphere.
Indicated, true, or groundspeed?
What’s the OAT, altitude, and nearest altimeter setting?
This is IAS. With AS the CJ4 will not attain M.74 at F400+ even at climb power settings. In MSFS it with over speed if you don’t come back into cruise power.
I’ll have a look at the atmospherics the next time I fly.
OK, with a bit of more investigation it seems that ActiveSky is giving significantly higher temps at altitude than expected - which would explain the performance drop I’ve noticed in the CJ4.
Testing with a flight from KTEB to Chicago Executtve. SimBrief was giving an OAT of -52 at FL430 (ISA +5)… At that level/waypoints
Active Sky was giving OAT -41, ISA +16
MSFS Live Weather was OAT -49, ISA +8
Looks like the ActiveSky temperature calculation is off. I’m wondering if AS takes humidity into consideration as that obviously affects the lapse rate? Whatever , it’s a pretty serious impact on aircraft performance. Ive flagged on the HiFi user forums and note that other users have also reported this.
All of the winds/temps aloft data are published and it should be granular enough to get an sim-useable environmental lapse rate from that.
You mean in AS or IRL? I know the MSFS atmospheric model takes humidity into account.
I’ll be interested to see what the dev says on the forums, but for now I’ll be using MSFS live weather when flying jets. I’m interested now to see what the impact on the ATR is as well.
Humidity (well, dewpoint really) works in layers anyway, it isn’t static all the way up - you can change the dewpoint just by advecting air (see what happens around a front or a dryline). Either way, you take humidity into account to estimate what a parcel will do from its initial state if it rises adiabatically (and eventually condenses, provided there’s enough instability). However, there are so many compounding factors that go into that, like lift, sink, mixing, as well as non-adiabatic processes like evaporation.
For the sim it’s better to work off a model output or an actual sounding and interpolate rather than just say “once the parcel is saturated, the rest of the column of air above must follow the saturated/moist adiabatic lapse rate all the way up,” because if you apply that universally, you’re going to get some really out of whack results.
Yes, complicated thing, the atmosphere. Still MSFS does a pretty good job of producing the right winds and temps aloft when using a simbrief flight plan as the yardstick. I wonder if this is down to the lack of API access and the use of presets.
It comes down to what data they’re using and injecting into the sim, which we don’t exactly know. As I said, winds and temps aloft data are publicly available, but aren’t super granular. That said, they can usually be interpolated and the result is useable enough for the sim. They don’t need to and shouldn’t model the actual environmental lapse rate under the hood of the sim because it doesn’t have the power to do so and attempting it would probably end in wild results.
Instead, they can just use the short-term synoptic-scale forecast wind and temps aloft data and inject it - it’ll be close enough for 98% of sim usage (except surface winds, which will open up a can of worms here). Thats what SimBrief is doing and what any pilot would do when planning a flight, with the understanding it’s not going to be 100% perfect (either in sim or reality). I imagine the base live weather of the sim is doing roughly the same.
I do wonder why AS is so far off, though…
I would strongly advise bringing this up on their forums, they are very responsive and if identified as a bug pretty quick to put it in a beta.
I added a post with the same data to an existing thread (from the last beta) reporting the same thing.
Same again today. LIRA-LFPO. Was reading +14°C higher than forecast/MSFS live weather. Seems pretty consistent with the +14.
Damian at HiFi has confirmed that changes to how MSFS calculates the lapse rate were causing an issue with OAT. There is a beta build available. Here was his reply to my issue report on the forums.
Thanks for investigating and fixing!
So they were modeling lapse rates and injecting them as environmental? I don’t understand why they’d do that when the data are readily available.
Perhaps related to other aspects of how the atmosphere is modelled such as the changes to light scattering/moisture. I don’t know how Asobo do it exactly, but I do know that it’s caused issues before (remember vatsim altitude problems?)
I’m having some weird problem.
I see lightnings in small cumulus or even in clear sky, where there should not have been.
When I put the game in active pause, the weather depiction dramatically changes (not always but often) and get big black cumulus where lightnings could be.
The cloud depiction was not correct when I started the fly and it needed active pause to switch to the correct depiction.
Without lightnings, I can’t say if the depiction is correct or not…
I have submitted a ticket.
I’m interested to know what settings people are using to try and add realism in terms of transitions.
Ideally I would not like to see the clouds morphing in front of me, but also don’t want the transitions to be so slow that the weather doesn’t change soon enough in order to represent the conditions in my locality.
This is for airliner flying by the way.
I bought this a few days back, and so far I am really impressed, on a number of fronts.
Cloud formations are closer to what we had at release than what we have now.
Wind is more consistent with reality, with the option to double them at ground level, to counter the Asobo halving of weather. Much discussed elsewhere, what they appear to have done is take the winds aloft, and halved them at ground level, instead of starting from the ground, and doubling them above.
This was a test I ran last night, parked on the runway at KOSH. The flat part was Asobo live weather, and then I enabled AS.
Wind gusts are much improved, and don’t have that baked in repeating pattern I found back in January.
In Asobo live weather, or preset for that matter, gusts are extremely frequent, leading to very aggressive changes in speeds. In AS these changes in velocity are still there, but are now far more gradual, which the thread I linked to above illustrates.
Obviously the program is a compromise, but a step in the right direction, and I’d really love the Asobo designers to get AS, and spend some serious time with it, then revisit what they are doing with their wind simulation. Plus, there is always the passive mode so you get the nice transition-less(?) Live Asobo weather, with the AS wind simulation injected.
I’m planning on running some more tests tonight at airports with reported gusts, so I’ll have some graphs up comparing how Asobo, and AS do it differently.
It’s pretty superior in most respects to MSFS live weather. The transitions are the main issue, even when flying low and slow.
I have started to find the AS cloud depiction a little predictable also - they are always flatter, less CU like than in MSFS (which has the converse problem). More CU would be welcome.
