Suggestion for Visibility Enhancement System.

It is something I've discussed before long ago.

I also found the newest discussion thread about a visibility model here:

https://forums.eagle.ru/showthread.php?t=209219

However,

since there is a suggestion about a new equation for a visibility model,

and instead of letting this equation be buried in a long thread,

I will set up a new thread.

This thread deals with the method I introduce here,

and it is better to do any other general visibility model discussion with the thread introduced earlier.

Let's explain the problem of the current visibility model first:

1.

In DCS 2.5.2, An aircraft drawn by dividing one pixel is always drawn as a point of one pixel.

2.

This is the only view correction that DCS 2.5.2 has.

3.

Since 1. and 2. In DCS, While aircraft can be drawn with invisible distance in real life,

you can not get any visibility correction for aircraft within a few miles that require visibility correction.

4.

It is expected that a combination of the next resolution and in-game FOV is required for the current monitor to provide real human eyesight.

4.1.

The visual acuity 20/20 can distinguish 1/60 angle of a degree(is called 1 minute).

4.2.

For a rough calculation, the combination of resolution and viewing angle

that 1 degree of in-game FOV has an information amount of 60 pixels is as follows:

1920 [width px] / 60 [px/hFOV] = 32 [hFOV] for FullHD

2560 [width px] / 60 [px/hFOV] = 43 [hFOV] for WQHD

3840 [width px] / 60 [px/hFOV] = 64 [hFOV] for 4K

5.

FOVs in 4.2. are not suitable for a constant visibility to fly or look around to get tally.

Users rather apply wider FOV with less visual information.

6.

The lack of visual information affects the two essential elements of the air combat and formation flying,

to get tally (or keep visual) and to distinguish the other's attitude from the contours.

7.

Since 5. and 6. what we need is a visibility correction that works with any FOV, any resolution, any distance.

8.

The visibility correction system that considers FOV, resolution, distance, would be a model enlargement equation that contains all of those three variables.

8.1.

Plane size continues to decrease by distance so pilots will not lose their sense of closing/leaving of opponent aircraft and distance relation to visual plane size.

8.2.

As an original 3D model is to be used, visibility will have a relation to each plane wingspan/length/height so the increase/decrease visibility made by opponent orientation will be more realistic.

8.3.

As an original 3D model is to be used, each planes coloring will affect its visibility. low vis coloring will work like in RL.

8.4.

As it is effective in WVR range (1.0nm-3.0nm) BFM/Formation experience will be a much more realistic thing.

8.5.

There will be no visual anomaly while model transiting to imposter/dot (becomes visible-unvisible-visible while zooming in/out)

8.6.

Helps VR users who can't zoom in.

I suggested such an equation last year.

https://forums.eagle.ru/showthread.php?p=3022334#post3022334

The Original idea is Coming From Serfoss 2003.

http://www.dtic.mil/docs/citations/ADA414893

If with Serfoss 2003 enlargement method, DCS visibility would be like following:

Since an equation from Serfoss 2003 does not have a FOV/Resolution consideration,

I added those variables in the equation I suggested last year.

A new equation was proposed by @sanpats at BMSForum a few months ago.

https://www.benchmarksims.org/forum/showthread.php?28660-About-SmartScaling-and-future-high-res-monitors&highlight=smart+scaling

As the method seems to correct equations accurately for differences in FOV and resolution, I strongly recommend his new equation here.

There were 2 equations @sanpats suggested.

Serfoss 2003 Enlargement Factor:

R = range(ft)/1000

Factor = 1.0 + 0.09226 * R - 0.00148 * R^2

@sanpats Factor 1:

adjustedMagnifyingFactor =

((referenceSerfossFactor - 1) * (((hFOV * 60) / hResolution) - 1)) + 1

//if adjustedMagnifyingFactor is below 1, use 1 instead

@sanpats Factor 2:

If (hFOV*60)/hResolution > 2 and range < 6 then //range = 6 means 6000ft

dampeningFactor = range/6

adjustedMagnifyingFactor =

1 + ((adjustedMagnifyingFactor - 1) * dampeningFactor)

The result of the first equation slowly matches the number of pixels of the aircraft at any FOV any resolution to the results of an appropriate number of aircraft drawn pixels obtained from Serfoss's experiment result.

https://www.benchmarksims.org/forum/showthread.php?28660-About-SmartScaling-and-future-high-res-monitors&p=463870&viewfull=1#post463870

The result of the second equation would be more aggressively.

https://www.benchmarksims.org/forum/showthread.php?28660-About-SmartScaling-and-future-high-res-monitors/page5&p=463905#post463905

DCS cannot scale up the 3D models. See Wags explanation at 1:45

Is that completely unchangeable core of the code?

Can't DCS calculate RCS from the 3D model before enlargement and draw enlarged 3D model?

Only Wags and ED knows...

Yeah I kinda knew visibility thing is hard, so I kinda avoided thinking about for all these years, it goes to the core of the engine and computer software graphics in general, maybe DCS can spearhead the campaign into something special which may even need special engine or driver and even API or GPU HW support, what some people seek may even be impossible without some changes in the full ... display monitors, resolutions, the way things get rendered (rasterization it self even?), anti aliasing, FOV ...etc

If future solutions need extra support, at least Vulkan API is extensible and actively updated. Even if ED is just one company and DCS just one game, if ED needs some kind of special API support as well as HW support to achieve some of the stuff needed for good simulation versus arcade games, with the then it may not be so far fetched especially with help of the open community and Khronos, to get such thing's in the Vulkan API via extensions which would hopefully get HW support and I think it should be much easier than in the past, where you had to be as high caliber as John Carmack or similar and twist GPU vendors arms for years to do something. For example Adaptive VSYNC i think is one of those things John Carmack pushed for as he talked about it in interviews but I'm sure there's tons of no-brainer things that software devs push on GPU vendors that aren't publicly known, if FPS goes below monitor refresh it simply disables VSYNC.

Yes this sounds weird why would simulation have to do with graphics ... I am a bit rambling without really knowning in detail, but I can see a bit the connection, the simulation it self may not be a problem and has nothing to do with Vulkan API or GPUs, but that's all behind the scenes anyway, you need the geometry and graphics to MATCH that what the simulation has calculated and accurately represent it.

I asked for clarification on the topic a while ago. Nineline said he'd put it before ED but I've never heard anything.

Spotting is one of the Achilles Heel’s that I find with DCS.

It’ll be amazing indeed to have it improved vastly somehow.

IIRC curvature is applied to NTTR as well.

Oh it is, oh I thought I was speculating too much so I deleted it.

Anyway the post was something like: This visibility thing may even need map tech support not just engine, as I sense it probably spans to the geometry field as well, not just rendering of draw distance (far clip), LODs, Pixels, Anti-Aliasing and FOV. Maybe because this isn't about model-geometry that you have in stuff like trees, but what do I know at this point.

I think I heard something about terrain curvature but maybe only on the newest tech maps like Persian Gulf? Not sure if it would make a big difference but at least in ground-air cases, as visibility is about angles as well and a little bit can be a big difference up far away.

Is that completely unchangeable core of the code?

Can't DCS calculate RCS from the 3D model before enlargement and draw enlarged 3D model?

 

Only Wags and ED knows...

The only sensible answer is yes, irrespective of what Wags says.

Telling the graphics card to rescale a model for the purpose of drawing it is trivial and does not affect the actual 3D model in any way. It's wholly self-contained process that happens on every frame using utterly trivial and primitive function calls — not something that could, should, or ever would possibly or sensibly be used for any outside algorithm to figure out cross-section at any range.

At no point does the LoD mesh itself need to be scaled in a way that would affect such calculations.

This conversation has been had multiple times, you're saying nothing new, and we're all just arguing in circles.

That said, if you can't keep track of somebody within 5 miles that is on you, not the game. The aircraft are quite visible if you know where to look. If they're otherwise hard to see, it's probably because even a 60ft plane is tiny at 5 miles and would be hard to see. You bird watching near an airbase with heavy traffic at regular intervals is not comparable to 'real life scenarios' where STUDIES BY THE MILITARY proved that what forum warriors THINK they should see and what the military deteemined people would ACTUALLY see are very different. Minimal effort on Google will yield these studies of you care to read them. All this napkin math and supposing is irrelevant.

As for formation flying or close range combat, model size isn't your problem, poor situational awareness is. The solution is practice, not making the planes appear so big that even a complete ditz can't possibly miss them.

I thought I missed the points so I deleted a reply. Nevermind.

That said, if you can't keep track of somebody within 5 miles that is on you, not the game. The aircraft are quite visible if you know where to look. If they're otherwise hard to see, it's probably because even a 60ft plane is tiny at 5 miles and would be hard to see.

I can't spot 2.0nm beyond 20 ft plane in DCS 1920*1080 70FOV in the best visibility weather I can set in the simulator. Is this same in RL?

You bird watching near an airbase with heavy traffic at regular intervals is not comparable to 'real life scenarios' where STUDIES BY THE MILITARY proved that what forum warriors THINK they should see and what the military deteemined people would ACTUALLY see are very different. Minimal effort on Google will yield these studies of you care to read them. All this napkin math and supposing is irrelevant.

So what actual STUDIES BY THE MILITARY you are referring to?

Serfoss2003 is also a study by the military and says model enlargement is needed to provide realistic orientation discrimination. What do you think about it? you can find it with minimal effort.

As for formation flying or close range combat, model size isn't your problem, poor situational awareness is. The solution is practice, not making the planes appear so big that even a complete ditz can't possibly miss them.

Will you think that as long as it's rendered no matter what color or resolution or pixel size, spot probability or orientation discrimination can be as same as in real life?

It's not for a complete ditz to not possibly miss them. It is for more realistic possibility of spot and information in any settings.

Honestly, I admit calculations here are just guessed. Those spots or attitude determine possibility can't be matched without subjective experiment results with the cooperation of pilots. Serfoss 2003 was only the study I could find that tried figuring out a proper visual aid for a simulator from such kind of experiment, but it still has many unclear points. like in what FOV did they render 3D models image for the test, He only says its projecting 20/40 eye vision image. @sanpats tried guessing what FOV of image Serfoss showed subject testees and made a formula to match the result of pixel size to the image showed in the experiment at any FOV any resolution.

If you have any interesting study, quote them and describe your opinion about possible visual aid (or maybe conclusion that no visual aid is needed). That's welcome. Not trolling like its easy to find but you just don't know you just not practicing enough.

Or you could just read over the last few threads started on this same topic since they removed model enlargement, where people already went over this same stuff. And yes, I've read the study you're referencing... the last time it was cited a few weeks ago in a thread that went on for I believe 20+ pages as many people argued for/against the notion. Like I said, you're not having any new thoughts here.

Ultimately it doesn't matter, as ED have indicated they don't want to use that sort of thing, and that it possibly causes issues with how they're coding radar. Whether that second point is good or bad or reasonable end result is the same : it's unlikely to come back.

Spotting is an issue that has existed since flight sims came out, and not all of them used 'model enlargement'. It is not game breaking any more now than it was ten years ago.

Spotting is an issue that has existed since flight sims came out, and not all of them used 'model enlargement'. It is not game breaking any more now than it was ten years ago.

That's not really true, though. It's far more gamebreaking now that we have a much wider range of resolutions, and also VR on top of that, and where the lack of resolution correction means that you get a significant disadvantage from having more advancer or capable hardware — it's not just gamebreaking, but outright silly in how it affects the game balance.

This is results of @sanpats formula

Result for 1920 * 1080 monitor

Result for 2560* 1440 monitor

Result for 3840 * 2160 monitor

The resolution would never be a problem as his calculation corrects those hardware differences.

you guys are not reading my posts carefully and just though already existing model enlargement but its more improved formula.

F-15 will be rendered about 9 pixels at 3.0nm for any resolution hardware. This will equal the visibility between different hardware.

WITHOUT THIS FORMULA lack of resolution means significant disadvantage from having more advancer or capable hardware.

Again:

@sanpats Factor 1:

adjustedMagnifyingFactor =

((referenceSerfossFactor - 1) * (((hFOV * 60) / hResolution) - 1)) + 1

//if adjustedMagnifyingFactor is below 1, use 1 instead

Of course, DCS cannot recognize monitor size so that can be a problem but at least for resolution...

This can be the answer for one reason why people refused model enlargement in a former thread.

That's not really true, though. It's far more gamebreaking now that we have a much wider range of resolutions, and also VR on top of that, and where the lack of resolution correction means that you get a significant disadvantage from having more advancer or capable hardware — it's not just gamebreaking, but outright silly in how it affects the game balance.

Me having high quality HOTAS, head tracking, triple monitors, and high end liquid cooled processor and GPU all mounted in a customised cockpit gives me a significant advantage from having more capable hardware. I have a wider FoV, higher framerates, longer view range, more precise control, and faster response times. The same applied twenty years ago man with potato was at severe disadvantage to man with super computer. Don't even get me started on min-maxers who tweak their graphic settings for improved visibility and FPS.

Welcome to PC gaming. Life is not fair, and you are not the top of the food chain. When people say "Ohhhh this hardware disparity is crazy it makes so much unfair". Yeah. Get used to it. It's always going to be that way. If you want equal footing where everybody is using the same shitty hardware and same shitty controls, may I introduce you to Xbox, Playstation, and Wii. Have fun over there.

Me having high quality HOTAS, head tracking, triple monitors, and high end liquid cooled processor and GPU all mounted in a customised cockpit gives me a significant advantage from having more capable hardware. I have a wider FoV, higher framerates, longer view range, more precise control, and faster response times. The same applied twenty years ago man with potato was at severe disadvantage to man with super computer. Don't even get me started on min-maxers who tweak their graphic settings for improved visibility and FPS.

 

 

Welcome to PC gaming. Life is not fair, and you are not the top of the food chain. When people say "Ohhhh this hardware disparity is crazy it makes so much unfair". Yeah. Get used to it. It's always going to be that way. If you want equal footing where everybody is using the same shitty hardware and same shitty controls, may I introduce you to Xbox, Playstation, and Wii. Have fun over there.

This NEW enlargement allows you using 4K monitor but FHD monitor user never has a disadvantage at the same time. Any user will fly more realistic simulation experience.

you can notice that only reading the graph I have posted.

Me having high quality HOTAS, head tracking, triple monitors, and high end liquid cooled processor and GPU all mounted in a customised cockpit gives me a significant advantage from having more capable hardware. I have a wider FoV, higher framerates, longer view range, more precise control, and faster response times.

That's just it: no you don't.

You have less view range and less time to respond. And if even if what you said were true, that would still just be an argument in favour of having correct scaling.

Also, whataboutism isn't really a strong argument for keeping an unbalancing factor around — especially one that has a known and trivial solution.

Wait, so now you guys are saying 1080 users are at a disadvantage? Last megathread on this topic, it was the other way around, with people claiming 4k users were at a disadvantage because when the game shifts to rendering aircraft as single pixels, a single pixel for a 4k user is smaller than a single pixel for a 1080 user. Be nice if you folks would make up your mind @@

@tippis

No, I have more view range, because I'm running a 1080ti and can crank the view range to maximum, whereas a guy running a lesser machine with the view range turned down cannot. Resolution has nothing to do with my example, especially since I never mentioned my resolution.

My point, since you guys seem to have missed it : you're never going to achieve anything even remotely resembling "equality" in a PC sphere. Hardware differences are incredible and confer equally enormous advantage/disadvantage depending on whether you are giving or receiving.

@chihiro

Any game can recognise monitor resolution, if instructed to do so. For proof look at the options where you can change your resolution :P And yes, I've seen those charts before. The last time they were posted, the last time this topic came up.

Wait, so now you guys are saying 1080 users are at a disadvantage?

No. People with high resolutions are, because your targets end up much smaller for much longer. Cranking the view range does not change this because it's an inherent defect in the rendering pipeline. You have less view range and reaction time because you are forced to try to spot much smaller targets than they have to. They will see them easier, sooner.

My point, since you guys seem to have missed it : you're never going to achieve anything even remotely resembling "equality" in a PC sphere.

Your point is not particularly relevant since as how you trivially can achieve equality in this particular case. Just because it's hard elsewhere does not mean we must at all cost avoid doing it here.

the graphs are right there senpai, the equation achieves equality.

you are just utilizing a fallacy anyway, just because perfection isn't possible doesn't mean we shouldn't attempt to move towards it.

not perfect, but more perfect than before.

also you seem really defensive for some reason

Wait, so now you guys are saying 1080 users are at a disadvantage? Last megathread on this topic, it was the other way around, with people claiming 4k users were at a disadvantage because when the game shifts to rendering aircraft as single pixels, a single pixel for a 4k user is smaller than a single pixel for a 1080 user. Be nice if you folks would make up your mind @@

That's because of dot system current DCS has.

It is not working within 5nm but works beyond where plane renders below 1 pixel.

Lower resolution wakes this system earlier which becomes an advantage.

Anyway, it will not help anything within WVR Combat or Combat Formation distance.

It also draws planes that can't be spottable in RL.

It has to be removed.

@Cik

No, it means I disagree with the suggested approach to resolving this "issue", just like I did the last time it came up and people suggested variations of the same thing. I strongly disagree with using "make it bigger-ism technology" because it's stupid and I consider the argument "but it's not faaaairrr" to be ridiculous. It is in exactly the same vein as complaining somebody has a better GPU or HOTAS.

Also, monitor size factors in on this, too. A guy running 4k on a 55" TV does not have this issue, even if the guy on a 27" monitor does. Meanwhile the guy playing on a 55" 1080 screen can reach out and stick his thumb through the other guy's pixels. End result is the same, one guy still ends up with a bigger target than the other, even with this "enlarge-o-vision" technology, somebody is still going to be easier depending on the tech you're using. So, it comes back around to my original observation of "you're suggesting to make it so big that any goof can see it no matter his resolution or observational acuity, meanwhile people who are capable of using the squishy white things in their face get to deal with gerbil sized aircraft zipping around their screen cause it's apparently "more realistic" according to the visually and .... uh... hardware...ally.... impaired among us."

Also, monitor size factors in on this, too. A guy running 4k on a 55" TV does not have this issue, even if the guy on a 27" monitor does. Meanwhile the guy playing on a 55" 1080 screen can reach out and stick his thumb through the other guy's pixels. End result is the same, one guy still ends up with a bigger target than the other, even with this "enlarge-o-vision" technology, somebody is still going to be easier depending on the tech you're using.

It's not the main purpose but 27-inch FHD will see the same size plane picture to 55 inch 4K monitor users because a model will be enlarged for FHD but not so much for 4K. 55 inch with FHD may get a larger model picture with an equal pixel size.

Whatever. Done debating it. I don't care for enlarge-o-vision, the core precept is backasswards and is trying to account for people's hardware. Nobody has to run a 4k screen or a 1080 screen, or at those resolutions or screen sizes if they don't want to. Lower end 4k screens aren't any more expensive than any other screen these days, having come down dramatically from their early highs, and 1080 screens are practically given away regardless of size. If somebody is running one or the other, it's because they chose to, and if they feel disadvantaged because they can't afford to match the other guy's hardware or whatever they want to right this grave injustice @@

ED are unlikely to implement any form of this nonsense, based on their past statements and indicated preferences, so there's probably not much point in arguing about it anyway. They killed it, and hopefully it stays dead :)