2 hours ago, psolk said:

Why are you so fixated on screen shots??? 

picture says a thousand words

2 hours ago, psolk said:

It's what happens with all of those pixels

exactly.

2 hours ago, psolk said:

even ones that are too small to be seen individually that when you add motion creates the shimmering effect

which shimmering effect are you talking about. I count about 7 different, some much worse than others, but non of them are related to AA afaict.

2 hours ago, psolk said:

as a result of AA not being properly applied.

no no no no no no and again no.

I already posted you screenshots demonstrating that AA is being applied exactly as AA should be applied.

2 hours ago, psolk said:

fact you think that "can't" happen on a 4K monitor because of pixel size

NOT what I said.

What I said was, seeing artifacts in 4K is proof that the problem is not AA, which even when it is working perfectly only removes artifacts smaller than a single pixel.

You’ve got to expect these minor problems when your light years ahead….

24 minutes ago, DD_Arthur said:

You’ve got to expect these minor problems when your light years ahead….

some of them arent even problems. 

The last one I found was shimmering that happens in real life, I just hadn't noticed it before.

Only way I can tolerate this forum anymore is with the ignore list.  Suddenly with one ignore everyone in this thread agrees...  5 pages later ONE person is still arguing though.  Amazing.

Edited February 11, 20233 yr by psolk

8 minutes ago, psolk said:

Only way I can tolerate this forum anymore is with the ignore list.  Suddenly with one ignore everyone in this thread agrees...  5 pages later ONE person is still arguing though.  Amazing.

So you want LR to impliment 64x AA?

Not gonna happen imho, and Imma gonna keep laughing at all of you who thinks it will. laugh the most when LR finally fixes the issues you are talking about with no changes to AA, especially not 64x AA.

Edited February 11, 20233 yr by mSparks

9 hours ago, UKflyer said:

Very suggestive image, not at all realistic, but suggestive. 👍 

Please don't be offended, it's certainly not your fault... that dark panel is terrible, unwatchable, it highlights another XP defect generated by the lighting system: the unrealistic fake adaptation of the eye - fake sensor camera imitation, and other such cinematic gimmicks typical of Battlefield video games or Hollywood movies.

Because of these absurd XP12 special effects that poor Airbus has become a 2D paper sheet, completely black, leaning against the screen, with some instrument stuck inside. 😢
Not even FS98 had ever descended to such a low level of human eye simulation unrealism.
I'm not surprised at all: XP has no idea how to simulate the human eye!
He only knows how to imitate cameras.
If any of the Laminars don't stop this madness XP12 risks becoming black and white sim.
All white (overexposed) the sky.
All black (underexposed) the cockpit.

The other sim who can't be named was also affected by the same problem and the same visual insanity.
But the customers fought hard to get the developers to fix the problem.
And they (in part) succeeded.
Now the other sim has very nice lighting inside the plane.
Who knows if XP12 will follow the wise example, or will continue to sleep and drag the "dark side of the force" for a long time as in the previous XP11.

As for the AA instead..... well 🤔.... they are working on it 🧐 (better to answer like this, at least they all agree) 🙃.

Edited February 12, 20233 yr by efis007

7 hours ago, efis007 said:

And they (in part) succeeded.

was that the one where they fixed distant objects shimmering by just not drawing distant objects anymore? 

Not sure that counts as succeeding in any way shape or form, more like nuking a key feature of flight sim to cater to the noisey crowd who have never sat in a cockpit.

10 hours ago, efis007 said:

Very suggestive image, not at all realistic, but suggestive. 👍 

Please don't be offended, it's certainly not your fault... that dark panel is terrible, unwatchable, it highlights another XP defect generated by the lighting system: the unrealistic fake adaptation of the eye - fake sensor camera imitation, and other such cinematic gimmicks typical of Battlefield video games or Hollywood movies.

Because of these absurd XP12 special effects that poor Airbus has become a 2D paper sheet, completely black, leaning against the screen, with some instrument stuck inside. 😢
Not even FS98 had ever descended to such a low level of human eye simulation unrealism.
I'm not surprised at all: XP has no idea how to simulate the human eye!
He only knows how to imitate cameras.
If any of the Laminars don't stop this madness XP12 risks becoming black and white sim.
All white (overexposed) the sky.
All black (underexposed) the cockpit.

The other sim who can't be named was also affected by the same problem and the same visual insanity.
But the customers fought hard to get the developers to fix the problem.
And they (in part) succeeded.
Now the other sim has very nice lighting inside the plane.
Who knows if XP12 will follow the wise example, or will continue to sleep and drag the "dark side of the force" for a long time as in the previous XP11.

As for the AA instead..... well 🤔.... they are working on it 🧐 (better to answer like this, at least they all agree) 🙃.

it's not too bad actually, i would expect the cockpit to be dark with no panel lighting during sunset.

 

the night lighting does need some improvement, especially in and out of clouds but volumetric lighting when implemented will help greatly. I expect good things from LR in this area.

The daytime lighting is the best i've seen in any sim.

 

and one more of the amazing lighting near Tromso 🙂

Edited February 12, 20233 yr by UKflyer

9 hours ago, efis007 said:

Very suggestive image, not at all realistic, but suggestive. 👍

Looks very realistic to me, except some known issues with transparent clouds.

1 hour ago, UKflyer said:

the night lighting does need some improvement

My suggestion to LR there, is that maybe the source of the problem is using RGB lighting.

for night vision our eyes rely on rods which are black and white but 500 to 1000 times more sensitive to light than the RGB cones.

RGB+rods lighting model would be another first for the industry, but I doubt they have had time to even think about it yet.

RGB+rods lighting model - Yes, which means night lighting should have higher dynamic range compression than day and saturation should be turned way down at night.  In another application of this, to see colors in Saturn through a telescope, your eyes need to be adjusted to dim lighting rather than dark.

1 hour ago, skipph said:

This guy is the smoke and mirrors expert

It should be clear.

If you complain about anti aliasing when there is nothing wrong with anti aliasing.

What you were actually complaining about wont get fixed.

simple as that.

anti aliasing removes artifacts less than a single pixel in size, 2 or 3 pixels if you are talking 8 or 16x AA

I 100% understand that artifacts significantly bigger than that are being described as a problem with anti aliasing. Im just trying to explain its not, and if you want them fixed you will need to be more specific.

Like if you call out the engineer to fix your microwave, because its not melting your rock collection properly , they are just gonna laugh at you.

or maybe you are just holding it wrong.

 

 

Actually no, the term aliasing (which anti-aliasing is the avoidance of..) is not limited to pixel resolution.  

The actual term "aliasing" refers to a signal sampling phenomena where you have a "signal" - in this case the image which you are trying to represent - that is attempted to be reproduced using "samples".  It is called 'aliasing' when your actual signal has a much higher frequency than your sample rate and the resulting frequency of the signal rebuilt from the samples "appears" to be a lower frequency.  The somewhat awkward application of the word 'alias' (taking on a false identity) means that the high frequency signal you intended to show "takes on the identity" of a lower frequency signal.   In a simple example, imagine a super hi res checkerboard where the space between the white and black squares is far less than a pixel.  If you imagine your sampled pixel grid, some samples will hit black some white but the pattern will not look like the original checkerboard.  In fact the final result will look a bit jumbled and will tend to look like a checkerboard with a lot less squares.. this is aliasing.

For a single image in computer graphics there are multiple places where sampling occurs, not just pixels.  There is also texture sampling, shadow ray sampling, etc. all of which are capable of, and correctly termed as 'aliasing' when the information they reproduce changes a lot between samples.  The task of "ANTI-aliasing" is solved in several places in the rendering engine.  Only ONE of those is the pixel sampling process that is addressed by things like MSAA.  M Sparks is correct in one way in that techniques like MSAA will never fix texture aliasing on its own. To address that you need to incorporate things like "anisotropic texture sampling", "texture resolution/quality settings", "shadow map resolution" and so on.  Those are not "always" (see below) handled by hardware settings on the graphics card, but they still are correctly said to combat aliasing.

On top of that, with moving imagery like a flight simulation, there is a temporal component to the signal - how the image changes over time is also "sampled" but in this case it is by "frames" not pixels.  "Shimmering" effects caused by sampling differences frame to frame can ALSO create aliasing.  This is what hardware techniques like TAA and FLSS work to address.  This last sentence IS important because here we have a case of hardware graphics card features that are addressing OTHER areas of the aliasing problem besides screen space pixel sampling.  AI techniques like FLSS attempt to address "multiple" aspects of causes for aliasing (geometry, textural, and temportal) in a non brute force way.  While not perfect, in practice it can produce some high quality results in a processor efficient manor.

All this is to say that to address the issue that we all call "aliasing" it is not sufficient to limit yourself to discussion of monitor pixel resolution.  Im completely confident LR knows this as well and are working on solutions.

Edited February 14, 20233 yr by VFXSimmer

1 hour ago, VFXSimmer said:

Actually no, the term aliasing (which anti-aliasing is the avoidance of..) is not limited to pixel resolution

take it up with the authors of e.g.

https://www.mdpi.com/2673-4001/3/3/29

which explains exactly which artifacts AA fixes.

XP12 AA fixes those artifacts exactly as described.  e.g.

becomes

waiting for it to fix something else is going to be a long wait.

 

3 minutes ago, mSparks said:

take it up with the authors of e.g.

https://www.mdpi.com/2673-4001/3/3/29

which explains exactly which artifacts AA fixes.

XP12 AA fixes those artifacts exactly as described.  e.g.

becomes

waiting for it to fix something else is going to be a long wait.

 

The author is wrong ... well more specifically .. what is wrong is taking that article as the complete discussion of what anti-aliasing is.  That describes ONE algorithm to address one specific aspect of it.  Says so in the text.

Edited February 14, 20233 yr by VFXSimmer