Visual Scaling, Spotting

[Tippis]

Forget math.

Just because reality proves you wrong doesn't mean we should forget about the mathematical and scientific proof for why it does so.

 

I had a 28” 4K screen and at normal viewing distance

…and what distance was that?

[SharpeXB]

…and what distance was that?

My eyes are 32” away from a 32” 4K screen and I can perceive jaggies on objects in DCS. Things like straight railings or wing edges, especially if I switched off antialiasing. So this display is not at a “retina” level.

[Mars Exulte]

How many times do you two need to fight the same circular battle?

[Tippis]

My eyes are 32” away from a 32” 4K screen and I can perceive jaggies on objects in DCS.

You should probably hand yourself in for some experimentation, then, or perhaps question how DCS' object rendering works and ask for it to be improved. :D

 

Either way, science and maths trumps your subjective and unproven assertions — doubly so since they're yours, specifically. Oh, and it would probably help if you actually tried responding to the right question with some kind of consistent information rather than suddenly shifting to a completely different scenario than the one that was actually asked for.

 

So this display is not at a “retina” level.

Depends on the intended viewing distance.

 

How many times do you two need to fight the same circular battle?

Probably until Sharpe starts getting a basic grasp of maths and science, or just manages to cobble together a coherent argument. Whichever comes first, which will clearly take a while regardless.

 

I mean, there's a very obvious thing he could go after in the stats I've used, but I have a sneaking suspicion that he hasn't even bothered to read it — same as always — and thus will just try to use his opinion as an attempted counter-argument to science.

Edited June 22, 2020 by Tippis

[SharpeXB]

or perhaps question how DCS' object rendering works and ask for it to be improved.

I would see the same thing in any game, it’s simply a matter of resolution. Maybe you should have your eyes checked. Oh and recheck your math, maybe you missed a decimal point. :book:

[Tippis]

I would see the same thing in any game, it’s simply a matter of resolution.

Nope. It's still a matter of distance — you are still confused about what properties matter. You always were, and as long as you are focusing on inches and pixels, you always will be. I'm not trying to trick you when I repeatedly point you towards the measurements you need to care about in spite of your obstinate refusal to actually read up on the subject.

 

Oh and recheck your math, maybe you missed a decimal point.

Nope. Maybe you should actually read the sources and do the maths yourself.

 

I'm actually curious now — let's see if you understand the issue and can actually form an argument from it. There is one very clear and very obvious thing that you could go after, but let me just warn you ahead of time: unlike the thing about which units and measurements matter, this suggestion is a trap because it will not actually change the answer to the initial challenge you posed. The reason that answer will not change lies within those units and measurements I'm not trying to trick you with. Indeed, if and when you figure out why those measurements matter rather than the ones you are focusing on, you will also figure out what the trap is and can thus avoid falling into it.

 

I'm being very generous here, so let's see how well you do with all these hints…

Edited June 22, 2020 by Tippis

[SharpeXB]

Nope. It's still a matter of distance

Ok so from my viewing distance, which is fairly normal, a 4K screen does not have invisible pixels, ie a “retina” display. It doesn’t replicate a real life 20/20 visual acuity view of the game.

[Tippis]

Ok so from my viewing distance, which is fairly normal, a 4K screen does not have invisible pixels, ie a “retina” display.

Your chosen viewing distance does not affect whether it's a “retina display” or not. Nothing is ever a “retina display” if you allow for that variable.

 

Not to mention that it's a marketing term that doesn't actually mean anything to begin with, but never mind that unfortunate detail. :D

 

It doesn’t replicate a real life 20/20 visual acuity view of the game.

And again, that's just your choice, not a because of any kind of limitation in the display. As mentioned, that kind of acuity can be achieved with any kind of monitor or light mixing system you'd ever want to suggest.

 

 

Also, I take it this means that you, as usual, don't want to actually read up on the maths and science behind the concept and understand what this whole “visual acuity” thing you're referring to really means? Instead you just leave it and go back to plugging the same unfounded and disproven statements you've had no success with before and hope that this will magically turn them into a coherent argument for no reason…

Edited June 22, 2020 by Tippis

[SharpeXB]

Your chosen viewing distance does not affect whether it's a “retina display” or not. Nothing is ever a “retina display” if you allow for that variable.

Well yes I’m using the Apple term “retina” to mean a display without discernible pixels at its intended viewing distance. Such a display would arguably simulate 20/20 vision. Fair enough?

 

In order to make my 32” screen a “retina” level display I would have to view it from farther away than would be practical for gaming. So for the purpose of playing flight sims it does not provide the equivalent of 20/20 vision.

 

How many different ways do you want the obvious to be explained?

[Tippis]

In order to make my 32” screen a “retina” level display I would have to view it from farther away than would be practical for gaming.

Not really, no, as the maths has shown.

 

How many ways do you want the obvious to be explained?

Your unfounded opinions and assumptions are neither obvious, nor explanations.

 

You might have stood a chance of making that kind of complaint if you had ever tried to use any kind of actual proof to support your stance — perhaps a bit of research, maybe a dollop of science, possibly some simple trigonometry… but alas.

[Gladman]

Lmao, the same arguments on the same topic can be found for the last 3 years at least. Its a game, get over it.

 

Happy flying.

[Fri13]

Your real eyesight is better than 1080p or even 2160p. One of the reasons there is a zoom view is to make up for lack of resolution. Since you can’t increase the resolution of your screen, the only solution is to magnify the image. VR needs a zoom view for this reason as well. The zoom view feature is not unique to DCS. All flight sims have this and for the same reason. Why it’s so confusing for DCS players in particular I have no idea.

 

Tell me something I Don't know....

 

But your problem is that you don't really understand that to spot something is extremely difficult in reality.

 

DCS is unrealistically easy to spot things now. It should be far more difficult, harder and almost impossible for most cases. But it ain't. And people like you are thinking that Zooming is required so that you would get things "more realistic", without understanding that already without zooming and even on VR you have superior spotting capability and detection in DCS.

 

And all you want is more unrealistic capabilities because you think that display resolution is holding you back..... What you truly want is labels. So simple as that.

[Fri13]

My eyes are 32” away from a 32” 4K screen and I can perceive jaggies on objects in DCS. Things like straight railings or wing edges, especially if I switched off antialiasing. So this display is not at a “retina” level.

 

That is not "seeing individual pixels". That is spotting "something off" as straight line is jagged line.

 

You to see individual pixels would mean you see the each pixel individually on the screen.

 

3840 x 2160 28" becomes retina limited with normal 20/20 vision at 56 cm (22").

 

I can make that exact same thing in real world that you are claiming "is not retina" and claim then that my eye is "not retina" because I can see "jagginess".

But it is nothing more than method how I present "a line".

[Taz1004]

Lmao, the same arguments on the same topic can be found for the last 3 years at least. Its a game, get over it.

 

Happy flying.

 

Same arguments on the same topic can be found on the last 30 posts of this thread. :megalol:

[Mars Exulte]

Same arguments on the same topic can be found on the last 30 posts of this thread. :megalol:

 

If only we could harness this energy for something productive.......

[SharpeXB]

That is not "seeing individual pixels". That is spotting "something off" as straight line is jagged line.

When you see a jagged edge you are seeing pixels. Do you see jagged edges on objects with your eyesight in real life? No. So even a 4K monitor is not giving you the equivalent of 20/20 eyesight.

Here's an example, especially when the scene is moving the jagged edges and crawlies on the masts and rigging are quite apparent. This is the native image in 3840x2160 without antialiasing. I would have to move my viewing distance across the room in order to not perceive this.

https://www.dropbox.com/s/s9f62i3iz74hiso/Screen_200622_071938.png?dl=0

 

Not really,

Snip...

Ok I’m done. You’re back on the ignore list. Your problems with this game just aren’t my concern.

Edited June 22, 2020 by SharpeXB

[Tippis]

So even a 4K monitor is not giving you the equivalent of 20/20 eyesight.

Again — still — that depends on viewing distance

 

And here's the thing that you failed to pick up on that would have saved you from repeating the same mistake over and over again: 20/20 acuity is not about seeing pixels.

 

The eye doesn't operate on pixels, same as how it doesn't operate on “frames per second”. It operates on angular differences, but also on differences in brightness and hue; on contrast; on changes in angles, over angles, and over time. You can definitely see jagged edges on things in real life if the contrast is high enough to expose that jaggedness. If the contrast isn't there, the eye instead sees a straight line. Coincidentally, this is why anti-aliasing works: it takes advantage of a limitation of distinction in the eye.

 

This is the native image in 3840x2160 without antialiasing. I would have to move my viewing distance across the room in order to not perceive this.

That depends on the pixel density.

 

On a 100ppi screen (just below where most regular gaming screens sit), you'd have to move it to just under 1m away for it to be below the threshold of angular resolution. Hell, let's put it at 2m just to make sure separate contours most definitely have begun blending together.

On a 130ppi screen (just below what a small:ish 4k screen would have), unsurprisingly, it will happen 23% closer: just over half a meter to be below the threshold; just over 1m to for blending to have set in good.

On a 160ppi screen (what your tiny 28":er would have), it equally unsurprisingly gets another 20% closer — half a meter for threshold; one meter for blending.

On a 226ppi screen (your average “retina” desktop/laptop display), we're now down to 40cm for the threshold; 80cm for blending. Hell, let's triple it to just over 1m to really wipe out any ability to pick out contours.

On a 326ppi screen (like no an older cellphone), we're down to less than a meter even for triple the threshold distance.

On a 458ppi screen (like on a silly cellphone), we're getting close to half a meter for triple the threshold.

 

Unless you're living in a closet, at no point does the image have to be “across the room” to be below the angular resolution of the human eye. Perhaps you mean “across the table”? Perhaps you start to understand why your constant attempts at simplifying this to just one measurement, be it pixels or inches, fails every time: because in isolation it can never tell the thing you want it to — in isolation it is meaningless. You always need to include those annoying angles you have yet to wrap your head around.

 

Ok I’m done.

That's the problem with your insistence on trying to argue against research, science, and maths: it doesn't work. You get beaten every time because you refuse to put in the work and as long as you do that and only have opinion and assumption on your side, you will never be able to formulate a cogent argument and you'll end up in this spot over and over again.

 

Top tip for next time: when I post a link about vision that talks about stellar observation, check what it says about the different measures of acuity actually mean, and what those are to the human eye under every-day circumstances. Again, though, this will not change the fact that any display can recreate normal visual acuity — it will just change at what distance that happens.

 

But of course, we both know that you're just bluffing. You're far too bound up in getting the last word in, and you have far too vested an interest in keeping your artificial advantages going to allow this conversation of game improvements go on unchallenged.

Edited June 22, 2020 by Tippis

[draconus]

So one guy is fixed on some old reasearch that was only suitable to checking target aspect in old sim with no other objects presented on screen and the other guy doesn't want it and has problems discussing why exactly. Meanwhile ED prepares to improve spotting by adding new realistic effects and tweaking pixels. Hopefully spotting discussion ends by the time all of us use VR with high enough resolution and lifelike gfx rendering.

[Mr. Big.Biggs]

Hey fellas. Anybody heard of this really cool flight sim DCS?

[LowRider88]

It’s great to hear that there is something coming from DCS for spotting.

I was starting to give up on this thread since it seemed to be hijacked for broader topics that I don’t really care about.

 

I don’t really care so much about the jump from dot to 3D in mid visual range.

That to me is esthetics, and is not an immersion killer for me because I’ve gotten used to that playing flight sims since 1994.

I play DCS to relive historical battles, not to watch a movie.

 

My main concern was the head on visibility of the small jet fighters, which only Fri13 and Draconus seemed to pick up on.

 

That detail may scare some multiplayer users who fear they can’t see anything. But for me it adds the next level of challenge I am looking for.

 

And it is supported by reputable military reports, and not commentary from youtubers.

[Taz1004]

Since there's all these disagreement and discrepancy regarding at what range something should be visible or not. If anyone thinks visibility is unrealistic, they can tweak the labels so it compensates only in the range they think they should be visible. They can make it so the 1 pixel colorless dot fades in and out at certain range.

[Fri13]

Meanwhile ED prepares to improve spotting by adding new realistic effects and tweaking pixels. Hopefully spotting discussion ends by the time all of us use VR with high enough resolution and lifelike gfx rendering.

 

Again, spotting in DCS is unrealistically good. You can spot ground and air units far further distance than you should.

 

So to "improve spotting" should really be "decreasing spotting distances" to make it more realistic.

 

Already Rift S renders with same quality as 4K display does.

Rift S has 2560×1440 (1280×1440 per eye) resolution, and I can replay the track on 4K monitor and I can as easily spot things from further I should. I can even set the DCS on desktop to run at 1280x1440 size and get a perfect comparison without optics etc.

 

We need new effects etc, that should start blending units to ground. Simplest is just transparency. You make units 50-75% transparent when immobile and color them actually with camouflage colors instead just darker 3D model like.

 

As it is just completely unrealistic to spot things as now in DCS used in VR.

 

So what really need is to make units hide better, much better....

[Tippis]

It’s great to hear that there is something coming from DCS for spotting.

I was starting to give up on this thread since it seemed to be hijacked for broader topics that I don’t really care about.

The problem is that your issues need to be considered within the context of the broader topic, especially when you start out trying to connect to that broader topic but gets it wrong what those concerns and discussions are all about.

 

My main concern was the head on visibility of the small jet fighters, which only Fri13 and Draconus seemed to pick up on.

[…]

And it is supported by reputable military reports, and not commentary from youtubers.

You'll see that a lot. Some posters like to throw anything they can in there, no matter how unsourced and subjective, simply because they have a hard time arguing actual research and maths. Coincidentally, the rationale for the scaling solution also comes from reputable military reports: an attempt to make differently-aspected aircraft be identified as easily as they are in real life. Because it turns out that the purely mathematical solution made it too hard.

 

Don't confuse “real” with “hard”.

 

They are not synonymous. There are plenty of things in DCS that are harder to do than they should be because they're not done all that realistically; where a touch more reality would make them easier. Some aspects of the visibility issues fall within that category. Other aspects fall at the very opposite end of the spectrum: they're still not handled realistically and thus become too easy. Both need to ba addressed, but it can't be a single solution and it will not universally give you any “next level of challenge”.

 

You have found some interesting data to support a specific problem, but you still need to consider the greater scope of that problem: does it hold true for other aspect angles — i.e. is it a matter of “overdraw” where pixels get filled in from smaller shapes than would be realistic, but the same pixels get equally filled in from larger shapes where it would be more sensible? How does it slot into the full spectrum of visibilities — i.e. if this front-aspect target is drawn as less visible, how does that affect its visibility when it's even farther away or even closer? How much of this visibility is down to things like texture filtering, resolutions, aliasing, and other rendering software and hardware issues? Now consider that it needs to hold up anywhere from 80km/h biplanes to mach 2.0 jets… And let's not even begin to discuss how trivial night spotting has become now and what happens when we throw VR into the mix. :D

 

This is why a “quick fix” will not work, even for “some people”: because it is not a single, simple problem for a single simple scenario. It's not that only some people have picked up on it — it's that most people have been in the visibility discussion long enough to have come to understand that larger scope.

[LowRider88]

The problem is that your issues need to be considered within the context of the broader topic, especially when you start out trying to connect to that broader topic but gets it wrong what those concerns and discussions are all about.

 

 

You'll see that a lot. Some posters like to throw anything they can in there, no matter how unsourced and subjective, simply because they have a hard time arguing actual research and maths. Coincidentally, the rationale for the scaling solution also comes from reputable military reports: an attempt to make differently-aspected aircraft be identified as easily as they are in real life. Because it turns out that the purely mathematical solution made it too hard.

 

Don't confuse “real” with “hard”.

 

They are not synonymous. There are plenty of things in DCS that are harder to do than they should be because they're not done all that realistically; where a touch more reality would make them easier. Some aspects of the visibility issues fall within that category. Other aspects fall at the very opposite end of the spectrum: they're still not handled realistically and thus become too easy. Both need to ba addressed, but it can't be a single solution and it will not universally give you any “next level of challenge”.

 

You have found some interesting data to support a specific problem, but you still need to consider the greater scope of that problem: does it hold true for other aspect angles — i.e. is it a matter of “overdraw” where pixels get filled in from smaller shapes than would be realistic, but the same pixels get equally filled in from larger shapes where it would be more sensible? How does it slot into the full spectrum of visibilities — i.e. if this front-aspect target is drawn as less visible, how does that affect its visibility when it's even farther away or even closer? How much of this visibility is down to things like texture filtering, resolutions, aliasing, and other rendering software and hardware issues? Now consider that it needs to hold up anywhere from 80km/h biplanes to mach 2.0 jets… And let's not even begin to discuss how trivial night spotting has become now and what happens when we throw VR into the mix. :D

 

This is why a “quick fix” will not work, even for “some people”: because it is not a single, simple problem for a single simple scenario. It's not that only some people have picked up on it — it's that most people have been in the visibility discussion long enough to have come to understand that larger scope.

 

By all means, if you want to broaden the scope and feature creep, you can open a brand new thread, rather than confuse what I am asking for. Please stop telling me what I am asking for.

[LowRider88]

Again, spotting in DCS is unrealistically good. You can spot ground and air units far further distance than you should.

 

So to "improve spotting" should really be "decreasing spotting distances" to make it more realistic.

 

Already Rift S renders with same quality as 4K display does.

Rift S has 2560×1440 (1280×1440 per eye) resolution, and I can replay the track on 4K monitor and I can as easily spot things from further I should. I can even set the DCS on desktop to run at 1280x1440 size and get a perfect comparison without optics etc.

 

We need new effects etc, that should start blending units to ground. Simplest is just transparency. You make units 50-75% transparent when immobile and color them actually with camouflage colors instead just darker 3D model like.

 

As it is just completely unrealistic to spot things as now in DCS used in VR.

 

So what really need is to make units hide better, much better....

 

Just to add to your comments Fri13,

 

30 nm seems way too far for visual spotting.

30 nm is about 55 km.

I drive from one end of a major city to the other for my work commute, and that takes 42 km.

Is it really possible to see a tiny BF-109, beyond the opposite end of a major city?

Seems quite far fetched to me.

 

From my tests in DCS, the roughly 2 nm range when an F-5 or Mig-21 appears head on Is roughly when the AWACS or GCI calls “merge”. That makes sense to me.

I should need their help to tell me they are coming in visual range, not see him coming from 15 nm out. If that’s the case, why bother call merge?

 

Still not sure why people are refuting your accurately sourced data.