Weather like pre su7 before 8 AM UK time, photos and videos provided!2

[Moach6908]

Yes, they did say it shouldn’t interfere with clouds, and if that really were the case, then quite possibly we’d have no issue at all with the blending system.

But it does interfere. This is very likely a yet-unacknowledged bug. The system clearly does not act in accordance to what is being claimed. Clearly, this is indicative that something is wrong in there. A Bug.

And perhaps that is all that stands in between the metar blending system and the former glory that was the weather before its introduction. That would be a wonderful turnout, as it’d mean that a properly functioning metar blend might yet one day deliver those beautiful cloudscapes of pre-SU7.

Can we get a “Bug-Logged” status on this? There is plenty of evidence to support the conclusion that metar blending IS somehow interfering with cloud generation;

Given the statements that it should not do this, we have thus logically arrived at the presence of a bug.
i.e: The degradation into “All-Puffulus Weather” is incorrect and unintended behavior of the metar blending system.

[Perrry]

@Moach6908

The degradation into “All-Puffulus Weather” is incorrect and unintended behavior of the metar blending system.

Well, i’m not that sure it’s unintended. Remember the popcorn clouds everywhere? Those they removed. But i’m 100% sure those were placed to make the atmosphere feel a bit more varied than it were when su7 were released. But it turned out to feel even more generic.

Can we hope to see pre su7 weather back someday? Or is the battle lost. I bet it’s lost.

A question for those that want METAR to form weather in the sim. Why is METAR even reported? I know the answere. Because the weather constantly changes.

When we ask to have a METAR form weather it means we want the weather to be fixed fpr 30 minutes.

[Moach6908]

Well, they did fix that one. (One wouldn’t assume they had “intended” it, if they later made sure to make it go away. Nobody plans for bugs.)

But perhaps there was more to that story than originally thought. There’s even the chance that some relation might exist between this old issue and the yet-unsolved problem of cloudscape degradation. It might just be that popcorn-clouds wasn’t really as solved as all that after all.

Or it could be something else.

In my experience in software, bugs tend to have causes that are silly in direct proportion to the strangeness of their symptoms. That is: The weirder a bug appears, the more one goes “D’Oh!” when the reason for it is discovered. No amount of experience can make a programmer immune to this law.

I wouldn’t say the “battle is lost” - not at all. Quite possibly, there really was never a battle.
We’ve just been observing things without enough information to decide on whether they should be seen as consequences of a poor tradeoff, (the original “why did you do this to us?!” theory) or simply as something that should be reported as a bug (a more likely explanation with far less hubris/blame to go around)

Which is why I’d like to have this thread reassessed by community managers and reinterpreted as a bug report. - It should read:
Metar Injection is causing adverse effects of severe degradation to the quality of cloud depiction.

We were informed (albeit not sufficiently) that Metar blending should NOT have any effect on cloud formations in the sim.

It does.

Assuming no critical points were lost in poor communication, we can then infer that a bug does indeed exist. And therefore, that the sub-par cloudscapes since SU7 are the result of undesired/unintentional behavior, aka: a Bug.

[Perrry]

But the METAR disturb weather thread has been around since su7 were released. Why isn’t that stated as bug logged if it’s not intended. There is so many bug reporting threads that appeard right after su7 were released. haze appear suddenly, transitions are not smooth, Metar disturb weather, clouds stops injecting after some distance flown. To name a few.

To me all of this feels more like they completely broken the weather system but when they talk about it in the q&a it sounds like there is no issues at all.

[sydicek]

I believe that when there’s not a bug logged tag it means the devs weren’t able to repro the issue.

I’m really starting to believe what someone said recently. That the devs (those that we see at the Q&As) rarely if ever fly with live weather enabled.

Just notice when they’re releasing some videos on their channels. They always use weather presets.

[Moach6908]

It does FEEL like that. But that is an observation coupled to an interpretation of a chain of events. Not necessarily it points to the core of the matter.

Yes, clouds DO look significantly less impressive than they were before SU7. Something must be broken.

Was it a deliberate choice that did this? Because devs assumed players cared more about weather matching VatSim than they do about the game looking good? – I find that conclusion rather unlikely.

If nothing else, it is contrary to every other criticism put towards Asobo by the louder players, the much more common theme being “they only care about eye candy” and similar. - Any merit to such claims is entirely beyond the point here. This is only to point out a trend which in itself provides contrary arguments to the premature conclusion that “weather broke because of a bad idea”

A more straightforward explanation would go “weather broke because something isn’t working as expected”.

This is a more plausible scenario that completely eschews hubris as a causing factor. It also fits the claims of “clouds should not be affected” in a way that does not require additional theories of “lies and defensive misdirection” to make sense.

Quite possibly, people have been so distraught by the degradation that nobody had yet thought of lodging the matter in the form of a bug report. I’m doing this now:

Bug Report:

Assumed Intended behavior: Clouds like of Pre-SU7, with metar blending used to ensure accuracy with live real-world reports.

Observed behavior: Clouds look markedly less authentic than Pre-SU7, with repetitive patterns and very visible circles of discontinuity around weather stations. Also, accuracy with real-world reports remains unsure.

That is how a bug should be put forth. And this is the bug we’re reporting now. We would like to see a “Bug-Logged” tag on this thread to indicate that the matter has been correctly reassessed under these assumptions and will be treated as an item that requires fixing.

Additionally, we suggest an alternative/workaround that might serve satisfactorily in the meantime:
To install a toggle option to disable Metar Blending by user preference.
This option would provide much desired relief, greatly improving conditions to unearth the true cause of the problem. It would favor much clearer feedback from non-bereaved players, whilst devs may work under far less negativity and pressure than otherwise.

[Perrry]

Thank you for reporting it Hope to see a fix soon.

I have observed same unrealistic behavior after su7.

[AndyXPO]

I thought this bug had disappeared, but I’ve done 3 flights over the weekend and this has started happening again. Had some lovely pre-su7 style stratus cloud layers which suddenly dissolved and turned into puffy cumulus clouds. Did not correspond to any change of metar or anything like that. It’s as if the server goes “oh we weren’t supposed to show you those clouds, hold on a minute while we retrieve the dumbed down versions… bang… there you are.” So disheartening as you get a glimpse of what live weather is capable of.

[hobanagerik]

Next time it happens, or rather before then after it happens, any chance of sharing some screenshots of it?

[AndyXPO]

Yes I’ll do my best next time I see some what I consider to be pre-SU7 clouds, I’ll take some shots and then if it changes I’ll get the comparison shots. Cheers.

[francobfort]

Cloud change in sim update 13 @ 1pm UK time

Cloud change in sim update 13 @ 2pm UK time

Cloud change in sim update 13 @ 3pm UK time

Cloud change in sim update 13 @ 4pm UK time

Four times in a row the same sudden shift, followed by a gradual decay into blue skies. Each hour I restarted the sim and for a while it was matching real weather pretty accurately. Then something forced it into a lesser state. I couldn’t make the opposite happen without restarting.

[AndyXPO]

Perfect evidence - the sad thing is I don’t think Asobo either know, or care, that this is happening. Look at that beautiful high overcast cloudscape… then bang… fluffy generic clouds. It is so sad to see what is possible in the sim, but is destroyed by the “improvements” they brought in. I would give anything to go back to the weather we had at launch.

[JustEmperor4724]

In the 2 and a half years I’ve had MSFS the weather degradation introduced with SU7 is the single largest disappointment I’ve had with the sim (maybe the colour-banding introduced with SU5 was worse but at least it was fixed, after 6 months). For a product that sold itself and prides itself on its visuals I really don’t understand why this was never been fixed. The most frustrating thing is that it was once fantastic and it has some moments now when it is fantastic. It feels so close but yet so far. Around 2 years now of undynamic preset weather after a wonderful 6 months of live weather. Hope died several months ago that this would be fixed. It is the one key element that will make me take the MSFS24 plunge or not.

[Perrry]

@AndyXPO

the sad thing is I don’t think Asobo either know, or care, that this is happening

I could try help them figure it out. There were several things changed in the weather in su7. I try to point them out one by one.

The things they did in su7:

1. Changed the MB model that is postprocessed by Meteoblue with satellite and METAR data with more frequent updates (What i think is the cause of the generic clouds and sudden transitions of the weather we see after su7)

2. Added the visibility from METAR that is never postprocessed with the MB-model. It’s always matches the METAR. For example if a METAR says 100 visibility we get that into the sim. (What i don’t think causes the rendering issue of the clouds) If you wonder how i know it’s not postprocessed? When su7 were released there were 1 hour of MB-data that were missing at midnight time i think it were between 2300z-0000z but the visibility from METAR were still in the sim. Thats why i know it’s not postprocessed/bleneded with MB-data.

3. They mentioned in the release notes of su7 that they changed the function of density sliders and may also have affected the live-weather density, and that may be the cause of less density in clouds after su7 as many have noticed and reported since su7. (Not the cause of those sudden changes of weather though)

Hope this helps. I would really like to have the pre su7 MB model back as well, at least as an optional feature

[CharlieFox00]

Perfect evidence of what? There’s absolutely no accompanying data. Time, location, local observations, comparing and contrasting with model runs would be nice.

All we see here is that the rendering changed from overcast to broken. If y’all want to make a point, you need to provide accompanying data.

[Perrry]

@CharlieFox00

If y’all want to make a point, you need to provide accompanying data.

I think what they want to mention is that weather never changes that fast IRL. I never had that sudden transitions of weather at release of this sim.

[CharlieFox00]

I will grant that. It was way too fast. But to say “lesser state” and all the other stuff is thin.

[RarefiedSwing22]

One other degradation I would note, since release, is the massive reduction in draw distance of the clouds around the player.

In the Feature Discovery Series (Episode 2) about weather, it was stated that for live weather the clouds are drawn to 600km around the aircraft. This was important, it was said, because it ensured that clouds would extend all the way to the horizon when at altitude. And it was true - it was glorious! Now, clouds never extend to the horizon unless you are at low level.

Yesterday I planned to write a new topic on this (I may still do) complete with screenshots, where I noticed that the cloud draw distance appears to change based on latitude.

• Over Florida the cloud draw distance is only around 185km,
• While further north over New England the cloud draw distance is reduced to just 140km.

(Test method: I slewed to max height and applied a single overcast cloud level to sea level. I then used ForeFlight to measure the distance from my aircraft to the edge of the cloud circle - again, I realise screenshots would help, but I’m not at my PC right now I’m afraid!)

If this is an oversight then I hope it’s something that could be fixed as part of 2020. If it’s deliberate for performance then I hope they can find a way to revert it for 2024.

[francobfort]

It was overcast the whole afternoon. The sim did show the overcast weather accurately every hour but then changed into blue skies a few minutes later. I repeated this four times to make sure.

[CharlieFox00]

Where was this?