If that is indeed how Rex is doing it, then yes, the Rex program will probably stop working on Tuesday when the next patch comes out if the base flight simulator executable file is recompiled in the patch, (which is very likely).
If that happens, Rex should be able to update their program to work with the new version quite quickly, but there would be at least some delay.
Well I am an outsider to this whole topic and what I learned so far is METAR is just a reporting format that is universally accepted for aviation purpose and the actual METAR is only reported in an hourly interval or so. Therefore, the so called âliveâ METAR is a forecast based on different models depending on what source you subscribe to.
So it looks like REX uses GFS forecast and MSFS uses MB forecast. It is a bit funny people here think GFS is the truth and disrespect MBâs forecast. To me, as long as MSFS depicts whatâs projected by MBâs model correctly, I have absolute no problem with it and no intention to buy a different model.
Btw, I like to track the hurricane forecast by ECMWF/GFS, etc. and to be honest, I donât see superiority of one model over the other. The only superiority we may have here is granularity of the modeling grid, etc. of which I have no knowledge.
Edit: I see where you come from and if REX only injects snapshotted actual METAR in the sim (could be outdated by 1 hour or so), then itâs definitely a step-back from what MSFS offers.
Thatâs exactly what I was afraid of going to a METAR-only solution. It may give you exact conditions, but at the cost of sudden weather changes during a longer flights. But that for me isnât even the worst part. Itâs having this âbubbleâ of the exact same condition around the aircraft. Even if they did manage a smooth transition⊠itâs still a bubble of the same weather condition around the aircraft.
MSFS default weather may not be 100% accurate and has itâs fair share of issues (which will surely be improved and worked on), but at least I can see weather ahead of me. I can see a huge cloud front from a distance and be able to fly towards it. And the best part of all⊠when flying online with others we all can see the exact same cloud formation. I can see the plane shake a little of someone in my 12 oâclock just before experiencing that bump myself. A guy was commenting on a rainbow left of an aircraft⊠I turn look left and see the exact same rainbow. Same goes with any clouds and even the shape of the cloud. To sum it up I was streaming what my sky looks like on discord and my fellow online pilots were able to see the exact 1:1 same image of the sky. And THAT already makes a huge difference. Thatâs much more important for me than seeing the same conditions I see out my window, as long as itâs true to how the weather works and what I have learnt during my meteorology class during my PPL training I am fine with it. And as there are no airliners right now to fly on VATSIM anyway it doesnât really matter that the winds might be off a little in accordance to whatâs reported live in the real world.
The main thing on VATSIM is that everyone is flying within the same weather anyway, it doesnât really matter if itâs whatâs reported in real or not.
Long story short. I prefer ârealâ weather conditions, rather than âaccurate to whatâs outside nowâ weather conditions. MSFS default in my opinion appears to be more on the former side, REX appears to be more on the latter side.
There are definitely advantages and disadvantages to each model. Iâm sure that Halber will swoop in soon, but in the meantime, for purposes of a flight simulator game with the best possible âlive weather,â you want the:
How accurate a model is at like 96 hours is highly relevant in the real world, but isnât relevant for us here. Weâre not looking for forecasting, weâre looking for current conditions.
I am not a big fan of REX as a company. They tend to not update often and bug reports are treated as if you are doing something wrong by reporting them. I bought Sky Force from them in 2019 and it never worked properly in FSX or P3D as advertised. The issue was weather would never update it was always the same weather as the departure location even though the METAR reports said otherwise. Support said it was my fault and refused to help though I have been working with flight simulator in a design role since 1995. They then reversed their position and said an update would be coming soon to fix it. I waited 2 months with no update to correct the issue. They then refused to refund the purchase stating a no refund policy and I ended up having to go through my bank to dispute the charge since the product did not work properly to get my money back and remove the software which my bank granted after submitting all the documentation of my communication with REX.
Keep in mind the real world weather not updating and being current for your location has been an issue since it was implemented with FS2000. You can take off to overcast skies in Chicago but when you land in Dallas you will have the same weather even though it is clear skies in Dallas. This issue has never been fixed as I stated above.
I have used Active Sky for sometime with P3D, FSX and FS2004 and I love it. I love the interaction with the developers etc. With all that being said I think I will wait until Active Sky comes up with something which will more than likely be a free upgrade for existing users rather than an additional purchase. That is if Microsoft does not address and fix the issue themselves.
I am not saying not to buy Weather Force just giving you my experience with REX, their software and their company.
Default weather engine > REX by FAR ! Please guys hold on to your money. Donât make the same mistake I did. REX weather engine is a DOWNGRADE. I wish there was a refund option.
Asobo/Microsoft have given us this insane dynamic weather engine. Albeit a little inaccurate but is way better than REX. Just give them time to fix the little issues. It will get better. I mean heck it already is a lot better than it was 2 weeks back. Microsoft ainât stupid. They didnât went into a partnership with meteoblue just for the giggles. Good stuff is coming. Just wait and donât throw money at stuff that will be fixed by Asobo. Donât let history repeat itself please. You got problem with the default weather engine? Tell Asobo ! Make threads here. They are listening. Its a 10 year product boys.
For the US only, the ideal model would be the HRRR (High Resolution Rapid Refresh). It is updated every hour on the hour, and each run includes not only the latest airport METAR reports as initialization conditions, but also real-time satellite and radar data. The HRRR has a very small grid, and a wealth of forecast data for low levels, but not much info for higher altitudes. (It only includes upper wind data up to the 500 mb level).
And, of course, the HRRR only covers the continental US, so it would be of no use anywhere else in the world.
BUT if we had a complete understanding of how each weather parameter is injected into the sim, it would probably be possible for a 3PD to create an HRRR-based âUS Onlyâ weather injector that would have all the advantages of a horizontally and vertically unified model (like MeteoBlue provides), but more timely, and more accurate for one specific geographic area. The complete HRRR model is readily available for download from the NCEP.
I assume that the MeteoBlue model is based on their worldwide 30 km grid NEMS product that they run for time + 0 out to seven days, which is why it takes 5 hours to generate and distribute to end users. Since MSFS is only one of many end-user clients that MeteoBlue serves, they probably do not, and cannot make a âspecialâ model run just for MS/Asobo.
Since MSFS really only needs forecast data for the current time out to 24 hours, a forecast model that only contains data for one day could be generated and distributed much more quickly - but again, I doubt that MeteoBlue is in any position to provide a âspecialâ model just for one end-user.
I just found this in another thread and it seems to work well. Havenât tested it on long flights but loading etc is nice. It takes a bit of an expert user to install it but seems good.
This makes a lot of sense.
Does this mean there would be ~5 hour delay for each forecast and MSFSâs live weather is what was forecasted 5 hours ago?
I tried to match the live weather of 5 airports in MSFS to MBâs map data and found out itâs synced quite well (maybe about ~10min delay). It seems plausible that thereâs a hybrid mechanism to mix TAF from 5 hours ago with current METAR reading to produce we we have in MSFS.
Thanks!
I am glad that you liked it 
There should not be any delay. A model run goes from 0 out into the future. The future (forecast) is simply overlaid onto real-time in the game, and you get that forecast until the next model run gets uploaded.
Most of these global models go out for like 150+ hours. So in theory, Asobo could just load up a new one once a week. That would not be ideal since the further out you go, the higher the risk that the forecast is wildly wrong (>72 hours things can start getting whacky). What would be ideal is loading up the new model each day, and using the 0-24 hour data until the next one comes out. You should still have pretty accurate weather in that timeframe.
That said, Asoboâs current, actual implementation is unknown. There have been multiple reports of fairly out-of-date weather so who knows at this point. That is a rhetorical question since obviously Asobo knows, they just arenât telling.
I don 't understand this business about âgarbageâ from all this developers. It seems to me that the problem is that none of then are going to be able to charge 50, 80 or even a hundred dollars for an addon like they do on the other sims. I still believe MSFS is the best hope for us simmers as long as we donât fall into the same endless pit we had before, were we ended up paying ridiculous amounts of money for an addon. I keep on remembering how much devs like, to give a decent example, Captain sim, charges so much money for any addon plus whatever addon for that addon (sorry for the redundancy) they make, and to top it all, with really â â â â â â FRS. I would have thought that by now we would have wised up enough to take and enjoy this sim the way it is which really, regardless of the flaws, is a hell of a lot better. We all know is just a matter of time before they get it right but weâre just to impatience.
Issue with the Unreal Weather is that it can not be installed to the MS Store version of the game only the steam version. Microsoft locks you out of the file permissions to the folder so you canât edit the files etc. I am trying to find a work around now but it did install into my Steam version fine.
I keep both versions for development purposes and I have an Xbox so I can play it on my Series X when it is released for that.
i agree with you, i will throw money only for flight school airliners
Out of curiosity: launch Command Line as Administrator (thatâs important) and navigate to that folder from there. Now try to create some simple file there, for example via the following command:
dir > _test.txt
The above will list the current directory contents and redirect the results into _test.txt
I also added a link to the blog post that deals with WindowsApp folder permissions.
Lastly, from my experience with Windows, there has never been a case where I could not unlock some folder.
Apologies and I think I should clarify my initial statement here since it is a bit confusing.
What I meant is that the weather we have in MSFS, would be a t+5h forecast from 5 hours ago, which is theoretically worse than t+4h from 4 hours ago, etc.
However, METAR observations (at least from MBâs map) seem to be updated every hour and based on my limited investigation, the real weather in MSFS seems fairly close to that report. So itâs either MBâs model is accurate enough for t+n usage or thereâs some overlay/correction mechanism leveraging most recent METAR report.
Now thinking about this, I donât think Asobo has any thing to do with the live weather much if every bit of information is broadcasted by MB. What Asobo needs to do are:
If thereâs a way to examine what weather information is being streamed in MSFS, it would be much easier to determine what went wrong in some of the cases.
Yeah I took full ownership of the WindowsApp folder but will not let me take ownership of the folders I need to copy the files etc. There is another thread on this as well.
No there should NOT be a delay. The forecast model contains (in advance) the forecasted weather for any location on earth for at least the next 24 hours. So, if you are in Dallas at 1300 local time on Sunday Sep 27th, live weather should give you the model forecast for that specific hour, day and location.
And sometimes it does. Sometimes not. When it fails, it appears you often get the forecast for the right location, but either exactly 12 hours previous, or 24 hours previous. This happened to me last night in Las Vegas at 0200 local time (in KLAS). The temperature was 37C in the game. The actual temperature at that hour was 25C. The temperature was 37C - 12 hour previously at 1400 local time.
I think this is a problem with the Azure servers, not processing new updated model data from MeteoBlue quickly enough, or some kind of conversion error between the local time in the sim, and the corresponding Zulu time.
For me, (in the US East time zone), the chance of getting old data is increased in the late afternoon or early evening in my own local time zone, and less likely to be offset in the morning.
For what itâs worth, MeteoBlue distributes a new updated model forecast to all clients (including Microsoft) at approximately 05:30Z and 17:30Z each day.
It seems that you forgot to âcascade downâ your permission changes, and only applied them to the folder itself, but not to its children.
no I did the Cascade down as well