@HalberQuacky The cirrus clouds in my shot above is fully volumetric, they are not textures.
it’s not perfect by any means but it’s working.
easier to see like this, just right click and open in new tab.
there are many papers on how to create cloud types, several volumes papers exist on this subject.
I have personally studied several of them in my free time.
I’m fairly certain asobo is no stranger to those.
@ToniGsxr600 I am not here to start a sim war, so stop the agitated attitude please.
It was an example of what is possible, and frankly surprising a small team and company can pull it off while Asobo somehow can’t is my point here.
We are 4 years in now, it’s just strange we have not seen any improvement yet.
I still use MSFS as my main simulator and I just want to see it improve.
yeah same, it used to be easier tho.
something changed with the layering in recent updates, could make them a tad more convincing with manual weather sliders.
but ofc it could never be as proper cirrus ones.
There are many types of cirrus. Volumetric techniques can do a reasonable emulation of cirrus that are broad sheets with little structure. What it can’t do are cirrocumulus, which have a degree of vertical development broken up into many individual cells.
Since using MSFS since Alpha, I’ve seen it go through many transformations, both good and bad. When it comes to weather, sadly the current state is the least enjoyable it’s ever been for me, mainly for all of the reasons other users have mentioned above, so I won’t repeat them.
Despite the beautiful scenery and some fantastic aircraft adorns, more often than not now I load up with all the intention to conduct a flight, see the state of the sky around me and then turn it off again.
I remember way back when we identified the issue with live weather, when it would get to around 1800z and then the sim would inject the weather for 0600z the next day, meaning a whole 12 hours worth of weather data was not only getting missed, but being replaced by the wrong data. This contributed hugely to the insurances we saw, sadly by the time this was finally listened to and fixed, the decision to change the live weather to what we have now had already been made.
And finally, one thing I will saw about METAR, from someone who works in the ops room for one of the biggest airlines in Europe, is that the data they provide 1) is extremely limited and 2) becomes outdated the second it’s published. Let’s say I look out the window at my local airport and it’s a thick layer of cloud as far as the eye can see, with no blue sky. The METAR report however, says it’s SCT035. Is the meter wrong? No, it’s reporting what it can detect within the limitations. The overcast layer is at 6,000ft and the meter cannot detect it. So what would we see in MSFS, more often now it would be a patchy mess of clouds. Meteoblue forecast says overcast, METAR says scattered, how should it know what to do? Allegedly cloud cover does not come from METAR, but I don’t believe this to be true.
Anyway, like many, I sincerely hope we see some significant improvements in 2024.
A ceilometer can see through multiple cloud layers, including overcast ones, and report on coverage (in oktas) of each of those layers. The ones I work with can see up to 8 distinct layers between the sensor and 10,000 metres altitude. It would report your scattered layer at 3500 as 3 or 4 oktas, and your overcast layer at 6000 as 8 oktas. It would also tell you how thick each of those layers are and if there were any other layers above the overcast at 6000 (including coverage and thicknesses of those hidden layers).
I see we’re still stuck in thinking METAR is directly changing the cloud depiction. The conversation always circles back to that, despite there being little evidence that has been the case for quite a long time. We can talk outdated data all day, but it’s a moot point and just leads the conversation astray
If METARs weren’t changing cloud depiction, wouldn’t we have a lot of “raining at the airport but no clouds” reports? But we don’t. Instead we have reports of clouds making annoyingly fast transitions – which is probably due to some nearby METAR. Besides, either Seb or Martial said that Meteoblue was injecting METAR data so even if Asobo do nothing, METARs will still influence clouds.
Here’s my latest cirrus-like clouds attempt. They’ve added texture in there so flat smooth fading high clouds don’t seem possible right now. But if you make them high enough altitude, the results are good to me:
Here’s Live Weather from a long time ago and, to me, it looks like the same texture (or very similar) but live weather wont use it the same way anymore
METAR inject data in the sim because METAR inject data into real-world modeling. They are part of the entire process that goes into modeling and forecasting. You cannot have modeling without surface observations. Period.
Yet, cloud depiction in the sim, while influenced at some level (fed into the overall picture?) is often independent of the metars the sim spits out. I have yet to see an overly abrupt cloud transition lately, at least anything that is well outside of transitions you’d see in real life. Visibility, yes, that’s problematic, but clouds are just doing their thing. Maybe you’re seeing a model refresh, same as it would in any refresh situation at any granularity of time, anywhere in the world.
Sadly the majority of them here in the UK and much of Europe don’t detect cloud above 5,000ft, another one of the main limitations of METAR reporting. So here you could have an overcast layer at 6,000ft and a METAR will report No Cloud Detected
I literally just did an IFR flight where I was in a large gap in the clouds, and over the course of about 5 seconds I was clouded in suddenly. It was quite jarring. I don’t really mind, but just thought I’d point it out. It probably depends on how many weather stations and METARS it has to gather data from too. Less = more spread out = more jarring transitions, maybe?
I’m not saying clouds match the METARs - far from it. The clouds don’t seem to match anything sometimes. Many of the METARs at airports I fly to will include overcast at 25000 feet if its there but the clouds pretty much never do that. The last time I saw a high level overcast was during a major hurricane that went up the eastern US coast last year.
When I say that clouds are influenced by METARs, I’m referring to the fact that when the sim started, clouds were purely the forecast model and we got a lot more variety and better looking clouds. I know you say it wasn’t realistic but I can’t call what we have now realistic either.
They are pretty much right where I expect them to be, based not simply on surface observations, but radar and satellite. As if all of those observational inputs are being fed into a rapid refresh model, as they should.
But there’s part of me that wonders if there isn’t a regional disconnect. I fly primarily in the US and I don’t see these issues with clouds anymore. Whereas folks who fly in Europe still see it all the time. We need data, but all we have is a bunch of loose observations and hunches.
It was not. It’s a simple fact. It was not behaving as real world weather behaves. It was just a pretty, animated scene stuck on a loop. It is much more realistic now (still not perfect, but it’s closer to matching the time and space in which weather exists). But again, it almost seems like the experience varies region to region.
I’ve seen it snowing, while landing in Nordic countries many times these winters. I’ve seen poor visibility in areas which actually have poor visibility at that point in time.
Things can be abrupt at times but I have had a very different experience than what you say.
