Planes visibility and smooth online gameplay in DCS 2.0?

[klem]

Ewww...

 

The resolution of that chart is off. Here is the original report so you can see that relative aircraft sizes as they should be.

 

The chart is from the Human Factors in Aviation textbook.

 

The report is on a mid collision between a Cessna 152 and a motor glider. Unfortunately, it was fatal for all involved.

 

Problem there Crump is FOV again. You have to shrink the .pdf on your screen to the physical size of the paper it was printed on (assuming the author thought that far), presumably about what we used to call 'Quarto' in the UK, and then view it at normal reading distance. Here in Europe that's about A4. Then the 1/8nm (250yd) might look about right. You can consider what a fighter might look like on a motorway slip lane when you're at the 300yd/metre marker from it.

 

There's no way USAStarkey's "two miles" is correct as shown on screen. Perhaps if it's a 4"x6" photo at arms length? I live 2.2 miles from my local airport and there's no way a, say, Harvard fuselage subtends that kind of angle to my eyes at two miles. Even so I think there is a visibility problem and the LODs are incorrect.

 

The physical sizes and the distance images are viewed from make a huge difference. Perception is everything.

 

Attached are photos of a Spitfire at estimated ranges of 400yds and 200yds (argument #1, I estimated the distances myself). At its 55mm focal length the lens has a FOV of 22deg 20'. That's 1/3 of what we have in DCS 'normal' view (60 - 80?). The photo looks right for distance because we adjust our perception to suit our knowledge of the image but if we triple the canvas size to FOV66 and slap it onto a screen without changing the image size the result would be ridiculous. So how should that be viewed on-screen to give a true impression of what we see in RL? What it seems to come down to is 'what angle does an aircaft subtend to the human eyeball at a certain distance' and then 'how do you represent that on a screen'.

 

Consider this: a 37ft Mustang wingspan at 2 miles subtends (is covered by) an angle of ~0.0035 radians or 0.200 degrees. On a typical human FOV to screen of ~70 degrees that's just 0.0029% which on a 1920 screen = 5.5 pixels. Now look at the attached DCS image from the Mustang. The circled (centre) dot (a Mustang) is 3km (~1.9miles) away. Without splitting hairs it should be about 5 pixels but is in fact only 3. The grey line I added at the bottom of the circle is 5 pixels. I expect there's some technical argument why 3 pixels is used (perhaps it was almost 4 and rounded down). What I can say is that what I see in RL at 2 miles is nearer those 5 pixels than 3. Also, if you zoom right in, a/c at 5,6,7,8 and 9 km are each one pixel. And that's the other problem (apart from pixel count), the screen resolutions are too coarse. Whilst my three screens deliver about 6MPixels across my full field of view for about 40% of my visual height range the human eyeball has about 576MPx covering the same area. Even 4k screens can't compete with that.

 

So much for the 'size' problem.

 

How about rendering?

 

Even if 3 pixels is correct, the graphics engine does not render it clearly and at 2 miles it should be very clear if small. Remember, I had to move my view (Trackir) around just to get the pixels to show as they do because the pixels come and go, the attached image is the best I could get. Also, zooming in you can see the jumble of coloured pixels used in an attempt to stay faithful to the aircraft colours in just three pixel blocks instead of more consistent averaged colouring across the pixels to overcome the problem. The LODs simply do not deliver.

 

Conclusion? There has to be a compromise to compensate for screen resolution shortcomings of size and colouration. LODs have to be rendered more consistently and, at nearer ranges, darker to compensate for Graphics shortcomings. What we really want is to see that something is there at a range where we would be able to see it in RL. If that means consistent colouring and darker pixels for nearer objects then that's what it needs to be. Put simply it's a conjouring trick. You cannot have reality.

Edited June 15, 2015 by klem
Forgot the attachments!

[gavagai]

Even if 3 pixels is correct, the graphics engine does not render it clearly and at 2 miles it should be very clear if small. Remember, I had to move my view (Trackir) around just to get the pixels to show as they do because the pixels come and go, the attached image is the best I could get. Also, zooming in you can see the jumble of coloured pixels used in an attempt to stay faithful to the aircraft colours in just three pixel blocks instead of more consistent averaged colouring across the pixels to overcome the problem. The LODs simply do not deliver.

 

I keep trying to explain that smart scaling does not increase the size of aircraft in your FoV. Instead, it prevents the pixels from blinking in and out, as you describe. Thanks for explaining it so well, but some will always fall back on the straw-man argument that you're asking for aircraft to appear larger than they should.

[shagrat]

I recently modded a bit with the Infantry units. Now when you look into the LOD definition, you see that infantry shows until 1000m(!).

At 1001m the soldier plainly visible as a "dot", or whatever " vanishes!

There simply is no LOD defined in the file. Now in the EDM folder we actually have an infantry far LOD model (a simple 1.9m high rectangular block).

For my mod I've set the LOD to 1900m. That corresponds to the visibility on a 4k monitor where you can actually spot a single soldier, or MANPAD for that matter.

Same with vehicles... Not all of them but, most of the few I checked, they don't use the low detail LOD, but simply get switched off, at a defined distance, which doesn't match what I can spot on a monitor...

I didn't check with plane LODs, but I can imagine some LODs have a similar issue here.

Now smartscaling or not, if a model is switched off at a distance of 2nm or whatever, rather than calculated with a low detail LOD, it pretty much explains, why we can't spot trucks, infantry and maybe even planes.

If you pass the distance of the LOD, an object "blinks" into existence. If you go farther than defined in the LOD, it simply vanishes... No matter what resolution etc.

As the LOD and models are likely part of the integrity check it is up to ED to make use of the existing low detail EDM models and add a proper LOD progression.

 

We can do it as a mod, but that would brake a whole lot of multiplayer servers, I guess.

So before discussing: smart scaling, we should fix the LOD first, maybe it already fixes a lot?

[Sierra99]

The spotting ranges in DCS remind me very much of what I see almost everyday spotting traffic alerts from ATC.

 

There are lots of people who seem to disagree with this statement.

 

I agree with it whole heartedly since I have lived it. I have been in the cockpit of a KC-10A with 4 highly experienced crewmembers, 3,000 feet AGL at 12:30 in the afternoon looking for "Traffic your 11 O'Clock, 1 mile, 1500 feet below you, VFR". No matter how hard we looked, nobody was able to find it until they saw US and took evasive action.

 

Sometimes that's how it works in the real world.

 

I sometimes wonder if people have realistic expectations regarding things of this nature and the simulator world. It would seem since some people are unable to see micron size objects on their monitors that represent a lifesized object more than 5-10 miles away on their monitors, ED needs to fix something.

 

My question is; who is ED supposed to design the sim for? The person with 10K in videocards and a ultra fine HD, LED, 1080P 480Mhz monitor...

 

Or do they design for the majority of the players using what is considered mainstream equipment.? Just curious.

 

Like I said. I agree with this comment completly.

[klem]

........................................"Traffic your 11 O'Clock, 1 mile, 1500 feet below you, VFR". No matter how hard we looked, nobody was able to find it until they saw US and took evasive action..............................

 

If only the AI were this blind! Unfortunately they seem to have radar not eyeballs.

 

EDIT: I meant to say that I am not even thinking about see aircraft below me. In DCS, against a clear blue sky the black-skinned Hawk just 1nm ahead of me is reduced to a near-invisible flickering ghost. At 2.0nm it's gone. That just isn't any where near realistic, I can see clearly an a/c, even a small one, at over 2nm against a clear sky. It's an argument about how they are rendered. I would expect it to be more difficult when looking down against terrain.

Edited June 16, 2015 by klem

[Oydoron]

Consider this: a 37ft Mustang wingspan at 2 miles subtends (is covered by) an angle of ~0.0035 radians or 0.200 degrees. On a typical human FOV to screen of ~70 degrees that's just 0.0029% which on a 1920 screen = 5.5 pixels. Now look at the attached DCS image from the Mustang. The circled (centre) dot (a Mustang) is 3km (~1.9miles) away. Without splitting hairs it should be about 5 pixels but is in fact only 3.

 

Your argument would be great if you assumed that the P-51 was a sphere. It of course isn't. If the Mustang is flying away from you or towards you, you are looking at the wings edge on. The wings are only a few inches thick, good luck spotting those at such a distance. And the fuselage is less than 3 feet wide and about 4 feet tall. If anything, the game is being generous. A better estimation would be to examine the total area presented to you and figure out how many pixels that should be, since that gives the total amount of proportional photons you are getting from the plane.

 

It's when the wings are angled that you have your best shot, but even then, they are only 8 feet wide at the widest, and 4 feet wide at their narrowest. Good pilots of course know this and will present the smallest profile possible when escaping or approaching, they are basically invisible until they are right on top of you.

 

So the problem comes down to yes, it's hard, but at 2 miles a 3 foot fuselage only takes up less than half a pixel. Even at 1 mile, it's just barely a pixel wide.

[klem]

Your argument would be great if you assumed that the P-51 was a sphere. It of course isn't. If the Mustang is flying away from you or towards you, you are looking at the wings edge on. The wings are only a few inches thick, good luck spotting those at such a distance. And the fuselage is less than 3 feet wide and about 4 feet tall. If anything, the game is being generous. A better estimation would be to examine the total area presented to you and figure out how many pixels that should be, since that gives the total amount of proportional photons you are getting from the plane.

 

It's when the wings are angled that you have your best shot, but even then, they are only 8 feet wide at the widest, and 4 feet wide at their narrowest. Good pilots of course know this and will present the smallest profile possible when escaping or approaching, they are basically invisible until they are right on top of you.

 

So the problem comes down to yes, it's hard, but at 2 miles a 3 foot fuselage only takes up less than half a pixel. Even at 1 mile, it's just barely a pixel wide.

 

All true but at the end of the day, in RL, I can see a small aircraft against a clear sky when it's 2 miles away and if it's turning it's that much easier but in DCS that just does not happen. It's just vague almost invisible flickers. Yours and others' arguments about what can and can't be seen in RL are also not represented in DCS. There's a need for some kind of compromise.

[Sierra99]

If only the AI were this blind! Unfortunately they seem to have radar not eyeballs.

 

EDIT: I meant to say that I am not even thinking about see aircraft below me. In DCS, against a clear blue sky the black-skinned Hawk just 1nm ahead of me is reduced to a near-invisible flickering ghost. At 2.0nm it's gone. That just isn't any where near realistic, I can see clearly an a/c, even a small one, at over 2nm against a clear sky. It's an argument about how they are rendered. I would expect it to be more difficult when looking down against terrain.

 

That situation is even more troublesome in real life. Google "Open Field Myopia"

 

http://www.skybrary.aero/index.php/Empty_Field_Myopia

 

"Beyond the visual threshold the image of distant aircraft, whether seen in silhouette as a dark speck or relatively bright dot may thus be spread over a larger region of the retina and become an insufficient stimulus for the eye to focus on. Human factors studies show that a dot very close to the threshold size in an otherwise empty field could suddenly disappear because it was an insufficient stimulus to prevent the adjustment of the eye to assume its resting state."

 

From this paragraph...it would appear ED is actually doing a good job of simulating real life!

In the sim world it would seem we have it better than in real life because provided we have decent eyesight and our monitors are a reasonable distance away, we can usually see something as small as 1-2 pixels. In real Life, if the target your looking for blends in just a little... Your eyes basically "fold".

 

I've never tried some of the mitigation techniques we use for open field in the jet while playing.

Edited June 16, 2015 by Sierra99

[shagrat]

There are lots of people who seem to disagree with this statement.

 

I agree with it whole heartedly since I have lived it. I have been in the cockpit of a KC-10A with 4 highly experienced crewmembers, 3,000 feet AGL at 12:30 in the afternoon looking for "Traffic your 11 O'Clock, 1 mile, 1500 feet below you, VFR". No matter how hard we looked, nobody was able to find it until they saw US and took evasive action.

 

Sometimes that's how it works in the real world.

 

I sometimes wonder if people have realistic expectations regarding things of this nature and the simulator world. It would seem since some people are unable to see micron size objects on their monitors that represent a lifesized object more than 5-10 miles away on their monitors, ED needs to fix something.

 

My question is; who is ED supposed to design the sim for? The person with 10K in videocards and a ultra fine HD, LED, 1080P 480Mhz monitor...

 

Or do they design for the majority of the players using what is considered mainstream equipment.? Just curious.

 

Like I said. I agree with this comment completly.

Depends. If they just messed up the LOD progression like with the infantry and ground vehicles, then that should be fixed!

I guarantee you, spotting a plane that IS VISIBLE as a dot or two, on a 4k is hard enough, like IRL...

 

And to keep with your example, the crew of the other jet spotted you, right? So you had a chance to spot them, yet you were unlucky to not spot them, which is realistic and OK.

Still there are so many factors preventing you from spotting a plane in DCS, like obstructing cockpit elements, dirt on the window, head movement etc.

 

There is a difference between not spotting the plane and not being able to spot the plane, as it isn't rendered, thus invisible!

This becomes obvious, when a perfectly visible bunch of pixels simply disappears, because the view point moved another 10ft out.

I have no problems with "the planes being too small" I have a problem with objects blinking out of existence, that were perfectly visible a second earlier, just because somebody decided, they shouldn't be visible at more than x meters...

 

So ED should first fix the LOD (use the low poly block placeholder as far edm model, but render this for a good mile longer than now.

If this already fixes the spotting issue, fine.

If different monitor resolutions still make problems we may again look into resolution dependent smart rendering.

 

An alpha blending model with one pixel could be a good compromise here.

No "big black dots" yet a chance to spot something...

EDIT: just to mention, the AI does not suffer from any physical issues with their eyes or such. They spot you! At a fixed distance with a fixed chance, if they did they don't loose you...

Edited June 16, 2015 by shagrat

[jaydee]

Problem there Crump is FOV again. You have to shrink the .pdf on your screen to the physical size of the paper it was printed on (assuming the author thought that far), presumably about what we used to call 'Quarto' in the UK, and then view it at normal reading distance. Here in Europe that's about A4. Then the 1/8nm (250yd) might look about right. You can consider what a fighter might look like on a motorway slip lane when you're at the 300yd/metre marker from it.

 

There's no way USAStarkey's "two miles" is correct as shown on screen. Perhaps if it's a 4"x6" photo at arms length? I live 2.2 miles from my local airport and there's no way a, say, Harvard fuselage subtends that kind of angle to my eyes at two miles. Even so I think there is a visibility problem and the LODs are incorrect.

 

The physical sizes and the distance images are viewed from make a huge difference. Perception is everything.

 

Attached are photos of a Spitfire at estimated ranges of 400yds and 200yds (argument #1, I estimated the distances myself). At its 55mm focal length the lens has a FOV of 22deg 20'. That's 1/3 of what we have in DCS 'normal' view (60 - 80?). The photo looks right for distance because we adjust our perception to suit our knowledge of the image but if we triple the canvas size to FOV66 and slap it onto a screen without changing the image size the result would be ridiculous. So how should that be viewed on-screen to give a true impression of what we see in RL? What it seems to come down to is 'what angle does an aircaft subtend to the human eyeball at a certain distance' and then 'how do you represent that on a screen'.

 

Consider this: a 37ft Mustang wingspan at 2 miles subtends (is covered by) an angle of ~0.0035 radians or 0.200 degrees. On a typical human FOV to screen of ~70 degrees that's just 0.0029% which on a 1920 screen = 5.5 pixels. Now look at the attached DCS image from the Mustang. The circled (centre) dot (a Mustang) is 3km (~1.9miles) away. Without splitting hairs it should be about 5 pixels but is in fact only 3. The grey line I added at the bottom of the circle is 5 pixels. I expect there's some technical argument why 3 pixels is used (perhaps it was almost 4 and rounded down). What I can say is that what I see in RL at 2 miles is nearer those 5 pixels than 3. Also, if you zoom right in, a/c at 5,6,7,8 and 9 km are each one pixel. And that's the other problem (apart from pixel count), the screen resolutions are too coarse. Whilst my three screens deliver about 6MPixels across my full field of view for about 40% of my visual height range the human eyeball has about 576MPx covering the same area. Even 4k screens can't compete with that.

 

So much for the 'size' problem.

 

How about rendering?

 

Even if 3 pixels is correct, the graphics engine does not render it clearly and at 2 miles it should be very clear if small. Remember, I had to move my view (Trackir) around just to get the pixels to show as they do because the pixels come and go, the attached image is the best I could get. Also, zooming in you can see the jumble of coloured pixels used in an attempt to stay faithful to the aircraft colours in just three pixel blocks instead of more consistent averaged colouring across the pixels to overcome the problem. The LODs simply do not deliver.

 

Conclusion? There has to be a compromise to compensate for screen resolution shortcomings of size and colouration. LODs have to be rendered more consistently and, at nearer ranges, darker to compensate for Graphics shortcomings. What we really want is to see that something is there at a range where we would be able to see it in RL. If that means consistent colouring and darker pixels for nearer objects then that's what it needs to be. Put simply it's a conjouring trick. You cannot have reality.

Very informative Klem ~S~

This is a great discussion IMHO !

I fly 3 Sims (when I can). One of them had the worst possible "spotting" issues. Visual target would Disappear and "Warp" past you. It was very frustrating...But they fixed it and I can see Dots, Zoom in to Identify and see them "Materialise" as they get closer !....

I have been flyin DCS since Lock-On (flanker before it)....All jets with radar !...so I always used radar(Close in) to "Point" my eyes and "expect "the target to appear "!

I never saw any problem with "Spotting" in DCS !.....Until I flew planes without radar ! Namely the ww2 birds.

I know nothing of the technical aspects of Rendering, Lods etc !

I don't know what is a "Realistic" view !...But from a PC simmers point of view,(mine),DCS falls short... in this one area !..The rest of the DCS world is as accurate as can be !

A quick question guys : will the "Edge' make this topic Moot ?..Are Lod improvements factored into Edge ?

~S~ to all

[msalama]

I don't know what is a "Realistic" view! ...But from a PC simmers point of view,(mine),DCS falls short... in this one area!

 

Ah, but therein lies the contradiction exactly. You don't know what's realistic - me neither - but still feel this is wrong. But IS it or is it not, then? And if not, why fix anything?

 

There's stuff like missing LODs there, and those should of course be fixed. But as it comes to the main question, who's the one who _proves_ this is wrong? Because if no-one does there's nothing to fix actually...

[gavagai]

Ah, but therein lies the contradiction exactly. You don't know what's realistic - me neither - but still feel this is wrong. But IS it or is it not, then? And if not, why fix anything?

 

There's stuff like missing LODs there, and those should of course be fixed. But as it comes to the main question, who's the one who _proves_ this is wrong? Because if no-one does there's nothing to fix actually...

 

Proving a negative is often a futile task.

 

To me, the important thing is that many WW2 simmers are keeping DCS on the shelf until it is fixed (they say as much on other forums).

[[DBS]TH0R]

Great post Klem. :thumbup:

 

Proving a negative is often a futile task.

 

To me, the important thing is that many WW2 simmers are keeping DCS on the shelf until it is fixed (they say as much on other forums).

 

Seconded. DCS in WWII is in its current form, at best, prof of concept IMHO. From spotting problems to lack of proper map just to begin with.

[msalama]

Proving a negative is often a futile task.

 

Perhaps. But regardless, fix the missing LODs first and then see where we stand. If the visibility is still awful, fine, see what can be done, but DO NOT implement a gamey/arcadey big black blob in the distance kind of "fix" no matter what! Enough of that kinda BS out there already :chair:

[gavagai]

There is only one flight sim I can think of that uses the 4 pixel dot to render aircraft at long range, but long range spotting isn't even the thing that bothers people in DCS. As my 7 year old observed one day, "Look Dad! That 190 can turn invisible!" and he spots them better than I can.

 

My layman's perspective is this. With a typical display and strict realism an aircraft's wing is too thin to be rendered with a pixel at all but the closest ranges. Other sims render the wing thickness with a pixel anyway, but DCS does not. Notice that this small compromise does not increase the apparent size of the aircraft's dimensions. It only forces their dimensions to be drawn at certain angles where they otherwise would not be drawn at all.

[Sierra99]

Quick question...

 

How do you compensate for the capabilities of current monitors? Aren't pixels either On or Off? Is it possible the aircraft is being "rendered" but actually displaying (or not displaying) the image is a limitation of monitor technology?

[NineLine]

Quick question...

 

How do you compensate for the capabilities of current monitors? Aren't pixels either On or Off? Is it possible the aircraft is being "rendered" but actually displaying (or not displaying) the image is a limitation of monitor technology?

 

Dont even get me started on colour difference in monitors and peoples perception of that colour, probably one of the biggest challenges of this issue (and most ignored for the pixel thing)... because honestly this shouldn't be be about spotting a single fighter at 10+nm, but losing them during a fight, and that comes down mostly to contrast with the background, then there is identifying once you do spot... again dont get me started.. numerous stories of WWII pilots having to get right up on a target to insure it was an enemy plane...

[msalama]

"Look Dad! That 190 can turn invisible!"

As I said, fix the missing LOD problems first and see where we stand then.

[klem]

That situation is even more troublesome in real life. Google "Open Field Myopia"

 

http://www.skybrary.aero/index.php/Empty_Field_Myopia

 

"Beyond the visual threshold the image of distant aircraft, whether seen in silhouette as a dark speck or relatively bright dot may thus be spread over a larger region of the retina and become an insufficient stimulus for the eye to focus on. Human factors studies show that a dot very close to the threshold size in an otherwise empty field could suddenly disappear because it was an insufficient stimulus to prevent the adjustment of the eye to assume its resting state."

 

From this paragraph...it would appear ED is actually doing a good job of simulating real life!

In the sim world it would seem we have it better than in real life because provided we have decent eyesight and our monitors are a reasonable distance away, we can usually see something as small as 1-2 pixels. In real Life, if the target your looking for blends in just a little... Your eyes basically "fold".

 

I've never tried some of the mitigation techniques we use for open field in the jet while playing.

 

With respect, in a way you're coming at it from the opposite direction and accepting the worst of conditions as normal. Bearing in mind this is a combat sim and we're all supposed to be determined to come out on top - and just as determined not to get jumped - we aren't likely in RL to fly around with our eyes in a relaxed state. It's possible to focus your eyes out to infinity and, as the line goes, "search for the bastards". Looking at the 'Defences' in that link:

 

Defences

 

To counter the weakened ability of the eye to maintain a distant focus, good practices are to:

 

• Focus frequently on distant visible objects, e.g. on outlines of terrain at or near the horizon; this helps to stimulate the eyes to establish long-distance focal points;
  
• With flat screens we don't have that opportunity or capability. A compromise is necessary to give us "pretty good eyesight" (or we would not be 'fighter pilots').
  
• Stimulate the eyes by focusing at own aircraft wing tips;
  
• Same argument
  
• Consider flying above a haze/smoke layer if possible;
  
• OK if present
  
• It may be preferable when scanning the sky for other aircraft to use peripheral vision to detect movement. Peripheral vision responds better in terms of detection of weak stimuli associated with slight movement than the central vision.
  
• Peripheral vision is not available to most players unless they turn their heads away from the screen. Probably a daft expectation.
  
  

"In the sim world it would seem we have it better than in real life because provided we have decent eyesight and our monitors are a reasonable distance away, we can usually see something as small as 1-2 pixels"

 

Not if those 2 pixels are constantly swapping from a pale sky-blue to nothing.

 

One of the things you notice when you watch an a/c say 2nm away is that the atmosphere has merged the a/c colours into a grey or pale colour but it holds steady. It doesn't wink weakly in and out. I watched a Cessna 172 this afternoon about 1.7nm away approaching the local airport in a blue sky, it was a mid-grey and quite small but I had no problems following it or looking away and picking it up again. In contrast I follow my Hawk wingman until he is 1nm away, giving marginal broken pixel winks, almost invisible and certainly looking nothing like that Cessna.

 

There is the lesser argument about size/distance but a greater argument about distance with colour and constancy. It is impossible to see anything beyond a couple of nm in DCS. If we are going to have a compromise we may have to choose between the unrealistic near impossible and the unrealistic slightly exaggerated and we have to remember we are trying to entertain ourselves. Trust me, many years in these sims tells me people will not come and play with something that is impossible to deal with (it's already happened with DCS for several of my long-term sim friends). "Where'd he gooo?". "Sorry, he's over a mile away now, you won't see him". As far as non-radar a/c are concerned it could kill DCS.

 

 

 

With the Cold War jets it didn't matter so much with radar systems that give you bvr kills or the ability to track the target on radar until you are on top of it but in the WWII types and even the Hawk it's all eyeballing and there's no pleasure in not being able to see a reasonable representation of the other aircraft.

Edited June 16, 2015 by klem

[gavagai]

Quick question...

 

Is it possible the aircraft is being "rendered" but actually displaying (or not displaying) the image is a limitation of monitor technology?

 

Drawing pixels is algorithm dependent. That is why you don't have aircraft disappearing with the smart scaling algorithm, even though it was written in the 1990s when computer displays were primitive compared to today.

[shagrat]

Unfortunately, DCS uses an LOD modeling to determine, when to render a pixel or not, only then the screen resolution/pixel/distance calculation is deciding IF a pixel is rendered, e.g. the object at that distance appears currently larger than a pixel.

[Oydoron]

Unfortunately, DCS uses an LOD modeling to determine, when to render a pixel or not, only then the screen resolution/pixel/distance calculation is deciding IF a pixel is rendered, e.g. the object at that distance appears currently larger than a pixel.

 

I wish they would change to some blob based render. Calculate the visible area of the aircraft (eg, small for straight on, large for seeing the bottoms of the wings) in arcseconds² and then render that much brightness and size. Then if it's less then a pixel, you'd still get a dot that would be brighter or darker depending on aircraft orientation. With the current LOD code, it just renders a cube at a certain distance.

[WildBillKelsoe]

Wasn't there a poll about upscaling? I think its the most demanded feature. In F4AF I can easily spot bandits 10 miles away by pressing L. Why cant we have this in DCS? Is it a proprietary technology?

 

I really hope ED prioritises this thread on the todo list, really hope.

Edited June 19, 2015 by NineLine

[NineLine]

If its smart scaling you are looking for, I doubt you will see that in DCS. There are other things they can do, and they know there are spotting issues, but its tough, if not impossible for anyone to judge without being hands on with 2.0.

[Rangi]

If its smart scaling you are looking for, I doubt you will see that in DCS. There are other things they can do, and they know there are spotting issues, but its tough, if not impossible for anyone to judge without being hands on with 2.0.

Thanks Sith, it's good to know they are on the case and I trust ED to do the right (and most realistic) thing.