Are clouds the best ever now? I'm impressed.

[mrueedi 702]

11 hours ago, ThomseN_inc said:

They dont look that realistic imho. The lighting and colours is just off....

Lightning and colours would only be off, if the replicated weather would require different lighting and colors. But there is no doubt, that there are real clouds with about this color and lighting.

There is such a tremendous variety how real clouds can look, that the lighting and color can vary between pretty extreme endpoints and the produced image is still realistic.

[ThomseN_inc 1,128]

7 hours ago, MrBitstFlyer said:

Look very good to me in SU10!

 I was talking about the clouds depicted in that particular screenshot.

Edited July 31, 2022 by ThomseN_inc

[Tuskin38 6,525]

Low clouds over KSFO, seen from KOAK

Edited July 31, 2022 by Tuskin38

[Noel 4,503]

Clouds and weather from a distance seem to be improving, and I'm not in the SU10 Beta.  But what is very easy to see is just how very low resolution these clouds are when viewed from up close and certain angles it's more apparent.  Call them voxels or pixels or whatever they are a grid-based rendering that fails because of its low resolution and until that is addressed we will not have skies that impress when up close especially, and this must affect how light interacts with them and so forth.  I hope Asobo can address this issue in particular and I see no good way to have distinct identifiable cloud morphologies until this is addressed.  As hardware improves and even now I'd like see resolution improve and turn over this adjustment in density/resolution to users who can dial-in a level their hardware can cope with.

You can see that I'm zoomed in some on this, but not all that much, to capture this very close up cloud:

Just horrid!  I wouldn't harp on this so much if there weren't already hardware resources to exploit which there most certainly are.  Give us the slider to impact cloud resolution in particular such that in the screenshot above you could dial out this low-density rendering, and fully, as the hardware permits.  Then continue to work on subtleties and cloud morphology, lighting and so forth. 

[March Hare 720]

58 minutes ago, Noel said:

I wouldn't harp on this so much if there weren't already hardware resources to exploit which there most certainly are.

Are there? I think you'd need to be developing the simulator to make such an absolute statement.

Don't forget Asobo are leaving headroom for other features users have pretty much demanded, too. I expected this, then Seb confirmed it himself in one of their live streams.

Development is a balancing act...

 

Edited July 31, 2022 by March Hare

[Noel 4,503]

1 hour ago, March Hare said:

Are there? I think you'd need to be developing the simulator to make such an absolute statement.

I'm a user, even I have resources to exploit.  And my hardware is now 2.5y/o.  I would happily dial back T-LOD and or other variables to improve cloud density/resolution.  Tell me March Hare--you'd rather NOT have a slider to allow YOU to decide what type of complexity you'd like to bias?  Why, pray tell?  With your logic that essentially says, "Um, Asobo knows where they want to go, so there is no need to suggest giving the user control over something like cloud density/quality."  By that very same logic you could say Asobo should remove T-LOD, O-LOD and multiple other controls that afford the end user to tune the experience to their liking.  Weird, hard to explain POV.

Edited July 31, 2022 by Noel

[Spike54 34]

All I know is that I spent hundreds of Euros over the years with P3D, FSX and X-Plane on the various versions of Active Sky, Rex, Sky Maxx etc, trying to get a decent depiction of live weather and clouds.

MSFS gave me beautiful clouds and live weather for free right out of the box and as far as I am concerned they've done an excellent job.

[HiFlyer 9,360]

25 minutes ago, Noel said:

I'm a user, even I have resources to exploit.  And my hardware is now 2.5y/o.  I would happily dial back T-LOD and or other variables to improve cloud density/resolution.  Tell me March Hare--you'd rather NOT have a slider to allow YOU to decide what type of complexity you'd like to bias?  Why, pray tell?

There are many many things in this universe that I want. 

I'm also told (at least that's what my mom said) that many, if not most of them I can't have.

*Shrugs*

I get it. People want perfection in all things in their sim/game. But even with that said, I have to think that a dial for additional cloud complexity in a program that is by almost all accounts already first in class, is going to be way way way down the list, somewhere at the back of volume 575 of first world problems, last page, last paragraph.

Obviously, an individuals mileage may vary on that, but at 30+ pages in, I think most people have pretty much made their individual stands.

Edited July 31, 2022 by HiFlyer

[March Hare 720]

44 minutes ago, Noel said:

I'm a user, even I have resources to exploit.  And my hardware is now 2.5y/o.  I would happily dial back T-LOD and or other variables to improve cloud density/resolution.  Tell me March Hare--you'd rather NOT have a slider to allow YOU to decide what type of complexity you'd like to bias?  Why, pray tell?  With your logic that essentially says, "Um, Asobo knows where they want to go, so there is no need to suggest giving the user control over something like cloud density/quality."  By that very same logic you could say Asobo should remove T-LOD, O-LOD and multiple other controls that afford the end user to tune the experience to their liking.  Weird, hard to explain POV.

I'd like to tell you your ideals are realistic, but I can't. I'd like to tell you why they aren't realistic or practical (from a development POV), but you've already reached an immutable conclusion.

I didn't suggest that some kind of improvement can't be made to clouds, either, and neither did Seb. He suggested they could try certain things, but he looked doubtful, and they have much more to get working (and stable) first.

Nothing I said implied that I don't want sliders to set preferences, either.

[Noel 4,503]

30 minutes ago, HiFlyer said:

There are many many things in this universe that I want. 

I'm also told (at least that's what my mom said) that many, if not most of them I can't have.

I get it. People want perfection in all things in their sim/game. But even with that said, I have to think that a dial for additional cloud complexity in a program that is by almost all accounts already first in class...

Amazing.  Perfection?  I'm talking DENSITY slider, not COMPLEXITY, and there is a big difference.  Density is just one attribute of several that go into cloud depiction. 

You see the grid layout upon which these clouds are built.  Have a close look at this--notice the grid and how one could describe this grid a s set of identical cubes all touching each other on a side.  Or +'s, etc.  There are so many of them per square area.  My sense is this is the basis of cloud resolution and therefore depiction--very visible the closer you get.   From a distance--no problem, as is the case with low resolution imagery.  Which is why I say clouds/weather improves, from a distance, and overall effects.  That doesn't have any impact on what we're talking about here.

Can you envision going to cell(s) in the logic table from whence this comes and allowing the variable that controls the number of cubes or pluses (+) per square area.  Suppose its default value at this moment is "100", and this value represents a grid density of 100 cubes per square inch or what have you.  Change this to 200, and now you have increased the number of cubes per same square area.  This doesn't define the shape, color, light reflectance and shadowing of clouds--just changes the number of cubes per square area.   I have a huge amount of GPU processing power that is not tapped, at all.  I chose to limit frames to 30 by Vsync, and my system always leans towards main thread limiting.  I adjust T-LOD to grapple with that, and all's well.   My primary point is until this issue is addressed we will have near-views of clouds looking like this, which is GROSSLY out of line with that "first in class" you allude to.  And, as I originally said, how hard can it be?  Now, of course I don't know what increasing cube density could do to anything else but just looking at this grid-matrix architecture there is absolutely a variable controlling density.  Asobo has decided, at the moment, the value will stay at X.  I'd like to see what changing X does, within a range of reasonability.   

It's weird these reactions I don't understand where it comes from at all.  Yes, I generally like what's been happening overall with weather.  But right now clouds up close are just very off looking to not just me--it's off to anyone with eyes.  For some reason when ideas are proposed like that the counter responses have to do with "be happy you have what you have".  I am, but I'm not blind either.  

 

[Cpt_Piett 4,664]

12 minutes ago, Noel said:

You see the grid layout upon which these clouds are built.  Have a close look at this--notice the grid and how one could describe this grid a s set of identical cubes all touching each other on a side.

I guess we all perceive things differently. Or have different expectations, is probably more correct to say. I very rarely notice the pixelation of the clouds. Only when I zoom in very close. Which I just did, as I wanted to grasp what you are talking about. And I agree - at that close range there might be some perceived "grid layout". I guess. I've never thought of it that way. Most of the time I'm very happy with the way the clouds are depicted. Before they do any work on the cloud density, I'd rather they look into adding more cloud variation. As well as continue tuning the dynamic weather, making sure to have smooth changes between METAR and Meteoblue weather. Also fine tune wind gusts which were introduced in the beta.  

[mrueedi 702]

35 minutes ago, Noel said:

But right now clouds up close are just very off looking to not just me--it's off to anyone with eyes.

But not on anyones system. Mine don't look like that. What setting do you have for the volumetric clouds in the graphics setting?

[HiFlyer 9,360]

33 minutes ago, Noel said:

Amazing.  Perfection?  I'm talking DENSITY slider, not COMPLEXITY, and there is a big difference.  Density is just one attribute of several that go into cloud depiction. 

Yes perfection. I chose the word very carefully.

Because once you get your cloud slider, there will be somebody standing behind you that wants a hue and saturation and intensity slider. And standing behind that person will be somebody who wants something else, and something else past that, and something else past that.

The goal can't be anything but perfection eventually, for all people under all circumstances at all times.

Which, to me, is part of our ongoing attempts to attain that perfection, for years, and in every subforum, for every sim, to the point where we can't tell the difference between our sims and a real airplane; a goal I believe is impossible until Star Trek Holodecks become a real thing.

In reality, as soon as Asobo makes one change in something as complex as this, then something else will most likely go off kilter according to someone else's eyes, and off we'll all go to the races once more, chasing perfection. (And yes, the word, I believe, ultimately, is perfection)

Even professional multi-billion dollar simulators can't match what we want.

Everybody wants their turn in the line with Asobo! 😃

 

 

[Stearmandriver 2,296]

If you guys could see cloud depictions in the most advanced visuals ever designed for level D sims...

... you'd choke.  😁

Look at all the screenshots in this thread, and then realize they're all happening on your little home computer. 

Expectations might need some managing lol. 

[Noel 4,503]

34 minutes ago, HiFlyer said:

Everybody wants their turn in the line with Asobo! 😃

 

It's just a suggestion, seems plausible (change grid density only), and see what it brings.  I can envision sharpening that up by 30% would be huge and well within today's higher end hardware to process.  You've changed the discussion from that, to "Does making a change suddenly mean you're open to all other changes coming from everywhere?"  Nonsense.  You evaluate the idea and see if it merits testing.  Very plausible grid density could be modified slightly without all hell breaking loose.  Just a suggestion, and there IS headroom to exploit right now.