The problem is that there are no real-time, real-world observational data sources that specify the altitudes, number of layers, or coverage of upper level clouds, or specific cloud types. There are also no real-time sources of upper wind data. All upper winds used in r/w aviation flight planning come from computer models, not actual observations.
Most airport METAR observations do not report any clouds that may exist above 12,000 feet. Some do not report any higher than 6000 feet. It’s quite possible for there to be a solid overcast at 7000 feet above a given airport and the corresponding METAR will report “CLR”.
There is no question that the sim does not have the ability to properly set visibility - even when setting custom weather directly (without Live Weather). The only current visibility parameter in the sim is “particle density” which is not sufficient, and does not directly correspond to the visibility measurements used in r/w aviation of miles/km or yards/meters. Until the core atmospheric model in MSFS gains the ability to generate true visibility restrictions (ranging from light haze to dense fog), it will not be able to make use of the visibility parameter in airport METAR reports.
Even there, real world METAR reports are not necessarily the “last and definitive word” on visibility either. Many METARs will simply indicate that the visibility is 6 statute miles or better. The actual visibility might be 7 miles - or it might be 70 miles - there is no way to know.
There is also no question that there is a big problem now in Live Weather, where the system apparently generates lightning anywhere that the model parameter for CAPE (Convective Available Potential Energy) is above a certain value. While CAPE is a good indicator that thunderstorms might form in a particular location, that parameter alone does not indicate that thunderstorms actually exist.
So please explain me how this is in line with the moderators remark of not attacking other’s posts! The moderator is talking about experiences of live weather issues, not about your scientific data of live weather issues. And trust me… we are not doing science at all here. Unless you are planning to rewrite this thread for something like Nature
Btw: I will stop commenting on your post from now on
Correct. Your post presents the needed truth of the matter to get to the heart of the matter.
So the question is, why did the feature presentation make the claim that the real weather would be the same in sim? The claim of exactly recreating real world weather is there for all to see, so they must be able to do this, eventually.
Look and hear the presentation starting at about 8:40, the “exact” claim is there:
Computer models and forecasts are getting better and better. They actually do a pretty good job. The Live Weather in MSFS still has a long way to go.
The claims in the video are typical marketing hype. Even if the sim was able to use all METAR parameters, beyond the current wind, temperature and pressure, (i.e. cloud coverage, precipitation and visibility), Live Weather will never be a “perfect match” for actual weather conditions.
Even though METARS are taken at least one each hour, they can be quite vague when it comes to higher level clouds, (and visibilty when conditions are VFR).
The current NEMS GLOBAL weather model used by MeteoBlue has a rather coarse grid (30 km), There are other models with a higher resolution grid, and which are updated more often than every 12 hours, but they are regional in coverage, not worldwide.
There are two sources of real-time data that potentially could greatly enhance the LiveWeather system, especially where thunderstorms are concerned: NEXRAD radar, and real-time lightning strike data.
The problem with NEXRAD is that it is not available everywhere. It covers the US, southern Canada and northern Mexico, Western Europe, Japan and South Korea, and parts (but not all of) Australia. NEXRAD data is freely available in the US from the National Weather Service, but that may not be the case in the other areas mentioned above.
Real time lightning data from the NLDN is horrendously expensive, and I can see no way that Microsoft could ever come to an agreement with the provider to use it in the sim, but there is an independent crowd-sourced lightning detection network called Blitzortung that potentially might be adaptable.
But taking this approach, Live Weather will never even be close, may has well stop trying. Would it not be better to think that the MSFS development team chose their words and claims carefully, so they would not appear to be plain liars?
Would it also not be better to think that since Active Sky does a very good job simulating real world weather, with more limitations in FSX/P3D/XP, that the MSFS team could have the potential to achieve even better results?
Since currently the MSFS development team is obviously very committed to realism in many aspects, why should it be thought that there may be a need for a submission of failure in the weather aspect?
I think this topic and others like it concerning Live Weather would be closed if the MSFS team considered Live Weather completed.
I suppose its possible that the current approach to Live Weather may be thrown out for another
idea. The main thing is to keep the votes, issue reports, and ideas coming. Something very simple may spark something big.
But to allow (or submit to) marketing hype as an excuse to fail is among the worst things to do.
I think Live weather is current, however I struggle to recognize it at first glance, which was possible in the past on most occasions. Here’s what I mean: I marked all the ares I think I recognised from the satellite imagery today around 1630Z. All areas outside these markings are hard to recognize or don’t resemble the actual situation at all:
Yes, one can clearly see that the sim depicts some thin cloud coverage up the west coast, but it doesn’t come near what looks like low stratus in the satellite image. And the situation within the sim (KSAN) is as expected, whereas webcams and the current METAR show a low overcast layer overhead KSAN and hazy conditions (visibility 9sm).
Long story short - something’s not as it should be, but it is not the “advanced weather bug” or “stuck weather”, as I didn’t find any fitting satellite images within the past two months.
Agree. However once again both communication in this thread and also via PM with the usual moderator has broken down like it always does. So we’re left just guessing.
Here’s what I understood from the recent Dev’s Q&A:
ASOBO is aware of discrepancies resulting from mixing METAR data with the weather system and are now delving deeper into this mixing process. One major change will be that the METAR data will be delivered to Meteoblue to get integrated into the weather model. Then there’s need for better integration of the Meteoblue data with the rendering engine, and that seems to be a tricky one they are still working on and should solve the visibility descrepancies by better taking clouds, fog and aerosols into account. The timeline for those seems to be the end of this year, and then there seems to be a far better resolution of the weather model in the works with a timeline of roughly end of 2022.
It was said many times - for other topics - that they are heavily relying on feedback and reports in these forums, so I guess it’s worth to continue reporting issues with Live weather regardless of the outlayed plans.
I don’t get how this will work. METARs are something which are given live throughout the day, this data could never be sent to meteoblue in advance. The only data they could send in advance would be a TAF report but even then that will still leave to inaccuracies because let’s say a TAF has a PROB30 of TS between a certain time, but those thunderstorms never happen. It would mean surely they they would happen in the sim even if not in real life.
Thanks for your feedback. Ok so you really have a lot of real experiences as i see. Perhaps i didn’t understood that your comment wasn’t to take too seriously, hence my reaction. Of course a sim will never replace real life but having accurate weather is something that other desktop-sims have since years. And when we have a “real live weather” option available that renders complete wrong or outdated weather conditions in 30% of the cases you will agree with me that we have a small problem? If people want the challenge to fly with random weather, i have no problem with that. I only have a problem when an important function like accurate weather is clearly buggy and people try to make it sound as being not an issue at all. Yes i want to train my real flights with the same conditions as real and i will choose my destination according the actual metars. When these are completey wrong (like rendering dust and 200feet overcast when the metars says CAVOK and i am suppose to fly VFR) you will hopefully understand why it can be a big issue for many of us
As I understand, Meteoblue’s model would do the weather generation in real time while taking into account the latest METAR data. Weather depiction with real time METARs is not something new.
I know it’s not, it’s been the standard in many flight sims for a long time now. But Meteoblue’s model is formulated and sent to MSFS ahead of time, it isn’t something which is constantly being streamed 24 hours a day in real time. So it’ll be interesting to see how they merge their model with metar data. I know they already do it for the winds temps and pressure, so it’ll be interesting to see how they further integrate clouds, precip etc
i don’t know how this real at all, after take off in the heavy rain in tokyo yesterday, i expected to go up under clouds maximum at 5000m, but no, look at this altitude, overcast was all the time i climbed around
Still boggles my mind that visibility parameters are still completely missing from the sim. Such a huge immersion breaker. The atmosphere is always cavok and crystal clear as if in space.
