I would like this thread to be very clean: please don't introduce any side issues.

 

Over the past couple of weeks, there has been a lot of discussion here about detection, and visual detection appears to relatively the least well understood component.

 

What do we know about visual detection?

 

1) Joe has provided a lot of reports, some anecdotal, some well-referenced, of how his units can't see the broad side of a barn. He has also provided a lot of anecdotal reports that the AI easily detects his units for unknown reasons. We cannot at this point say that those were definitely visual detections. He has provided some well-referenced reports that the AI easily detects his units, but there are only a few definite visual detections in there.

 

2) I can anecdotally say that visual detection doesn't seem to work very well.

 

The best explanation at this point seems to be that visual detection doesn't work very well, but it is possible to get lucky.

 

I believe that size factors strongly into the visual detection rules. I believe that these rules probably look very similar to the radar detection rules. It is a lot easier to track a detected unit than to first locate it, and I believe that Harpoon takes this into account.

 

I am looking for definite confirmation of any or all of these rules.

 

Let's look at some numbers. Consider an airplane that has a linear size of 50 feet. Two miles away, this aircraft subtends an arc of about 1/3 degree. That seems pretty small, but I can't say what it actually looks like. At 20 miles, however, that angle becomes .03 degrees, and I think that is very hard to see.

 

Let's try a measurement.

 

Below is a video of a low-level Phantom flyby. The Phantom comes in low from the background. Start the video, and when you first see the Phantom, pause it. Note the time, then unpause and note how long it takes for the Phantom to get right on top of you. Repeat, except this time you know exactly where to look. You might get a little more time. Knowing the speed of the Phantom, you can calculate the distance at which it is first seen.

 

My point is that if you are relying on visual detection, you can be fairly certain that the first evidence you get of an enemy fighter will be a missile up your tail.

 

 

I get 20 s, less than one Harpoon detection cycle. At 600 kn, that's about 3 nm.

I would like this thread to be very clean: please don't introduce any side issues.

 

Over the past couple of weeks, there has been a lot of discussion here about detection, and visual detection appears to relatively the least well understood component.

 

What do we know about visual detection?

 

1) Joe has provided a lot of reports, some anecdotal, some well-referenced, of how his units can't see the broad side of a barn. He has also provided a lot of anecdotal reports that the AI easily detects his units for unknown reasons. We cannot at this point say that those were definitely visual detections. He has provided some well-referenced reports that the AI easily detects his units, but there are only a few definite visual detections in there.

 

2) I can anecdotally say that visual detection doesn't seem to work very well.

 

The best explanation at this point seems to be that visual detection doesn't work very well, but it is possible to get lucky.

 

I believe that size factors strongly into the visual detection rules. I believe that these rules probably look very similar to the radar detection rules. It is a lot easier to track a detected unit than to first locate it, and I believe that Harpoon takes this into account.

...............................etection, you can be fairly certain that the first evidence you get of an enemy fighter will be a missile up your tail.

 

 

I get 20 s, less than one Harpoon detection cycle. At 600 kn, that's about 3 nm.

Good demo, something similar on the ocean with smallish boats if there is a bit of a swell too, amazing how close they get before you see them unless you know where to look - and that's in the day time with the sun behind you and high in the sky.

I'd have to say I haven't noticed problems that I'd put down to visual IDs, but while we are on that tack it sounds like its worth investigating.

I'd assume that like a lot of things in Harpoon its governed by the paper rules (that I know nothing about), so what do they say? and then we can test if we are getting that.

Good demo, something similar on the ocean with smallish boats if there is a bit of a swell too, amazing how close they get before you see them unless you know where to look - and that's in the day time with the sun behind you and high in the sky.

 

Play Dawn Patrol from either original GIUK, or the EC2003 version, both available on this site, and you'll pretty consistently get a visual detection on the Nanuchkas at 14 miles...

The visual detection and line of sight rules are also governed - at least to some extent - by H3 paper (as with the radar LOS).

 

The air to air visual LOS is related to the 'size' of the target, except at High altitude, where it is 50 nm.

 

Otherwise, at Low altitudes:

 

Vsmall - 0 nm

Small - 3 nm

Large - 6 nm

 

Recall that the 'size' of the target is determined by the Size Field in the Aircraft Annex of the PE.

 

The time of day will also affect chance of visual detection, i.e. 100% in daytime, 66% in twilight, 33% at night

 

Times of day are (roughly) broken down as follows:

 

2000 - 0400 hours, Night

0400 - 0900 hours, Morning (dawn/twilight)

0900 - 1600 hours, Day

1600 - 2000 hours, Evening (dusk/twilight)

 

Weather will also cut down on visibility, with modifiers as follows:

 

Clear weather, no precipitation = 1.0

Light rain/fog = 0.6

Moderate rain/fog = 0.2

Heavy rain/fog = 0.05

Thanks, Brad. I will eventually need to get the full rules, but this will do for now.

Play Dawn Patrol from either original GIUK, or the EC2003 version, both available on this site, and you'll pretty consistently get a visual detection on the Nanuchkas at 14 miles...

 

Namuchka is 40 ft. wide, 200 ft. long, geometrically more opaque than a plane, exists within a narrow slice of solid angle, has a wake, and in calm seas should be much easier to detect than a small fighter.

Some further notes for you on visuals and influencing factors, especially as regards ships and mobile land units (MLUs):

 

The 'Mast Height' and 'Detect Size' fields do not appear to matter when it comes to visually spotting ships and MLUs.

 

The 'low light level' flag in the Ships/MLU Annex gives that ship or MLU an enhanced visual capability. Essentially it makes it always 'daytime' for the platform as far as visuals go, but it is negatively affected by bad weather/precipitation.

 

It generally has a shorter range than the 'IR' flag, but can see submarine periscope wakes and gives more information upon detection (i.e. exact position, name, propulsion type).

 

The 'IR' flag is not influenced by weather and gives bearing, type and size of a detected target. It cannot, however, see submarine periscope wakes.

The 'Mast Height' and 'Detect Size' fields do not appear to matter when it comes to visually spotting ships and MLUs.

 

But we must remember the mast height matters in relation to the active/visually searching unit, at more mast height, more visual detection range. with range values similar to that in the paper rules.

The 'Mast Height' and 'Detect Size' fields do not appear to matter when it comes to visually spotting ships and MLUs.

 

But we must remember the mast height matters in relation to the active/visually searching unit, at more mast height, more visual detection range. with range values similar to that in the paper rules.

Please verify for us dummies,

for ship to ship visual ID, the mast height of the ship being searched for doesn't matter but the mast height of the ship doing the searching does.

Also does the mast height of the searching ship help to see low altitude aircraft, I looked at this in the manual but wasn't clear to me some of the detail.

Don

This is quite interesting. I've attached a scenario I am using to look at detecting ships from planes (westpac, HCDB110225). Also is a txt file that is data stolen from the manual regarding Maximum line of sight.

So far things seem to tie in re manual data, vis range from the game weather report (which takes time of day etc into account).

What really surprises me is the small difference that the size of the 'target' makes. Ships range from small (lifeboat) to very large (supertanker) but in good conditions at low and medium altitudes all sightings are generally being made in the range 7 - 10 nm.

Once conditions deteriorate, (eg night time) then the biggest difference seems to be the frequency of missing the smaller targets.

I'll try to put together a number of runs in different conditions (funny, but the weather seems to be always fine) and post up something more informative.

I'm interested in what others find and think, but I'm wondering if periscopes and life boats are to easy to see and supertankers a bit too difficult??

I'm assuming others are are looking at air to air first as Brad posted some stuff on that (which is why I thought I'd look at air to ship).

Don

SHPVIS.zip

What really surprises me is the small difference that the size of the 'target' makes. Ships range from small (lifeboat) to very large (supertanker) but in good conditions at low and medium altitudes all sightings are generally being made in the range 7 - 10 nm.

 

I have looked at this very crudely. It's a matter of the 2-D geometry of a tangent to a very large circle. It seems to be the case that the range is not very sensitive to the size of the smaller object. Probably I should work it out carefully and post the results.

What really surprises me is the small difference that the size of the 'target' makes. Ships range from small (lifeboat) to very large (supertanker) but in good conditions at low and medium altitudes all sightings are generally being made in the range 7 - 10 nm.

 

I have looked at this very crudely. It's a matter of the 2-D geometry of a tangent to a very large circle. It seems to be the case that the range is not very sensitive to the size of the smaller object. Probably I should work it out carefully and post the results.

I suppose one thing is how realistic we need to be with visual, take this example;

In real life observing from the bridge of a 20,000 ton oiler such as the old HMAS supply, a super tanker side on is visible in good conditions from very much further than from right front on (or stern on). From the side view such a very large ship is quite more visible than say a FFL, but viewed right front on, or stern on, the difference is no where near as much. As you say it is the tangent or angular size of the target.

There are many areas where visual detection is crucial, so I'd think that we need to have it working well.

Don

Very, very interesting issues, but I've a very small amount of spare time the next months.

But from early experimentations in another game iterations, I remember some doubtious aspects (some of it experimenting a WWI DB, without radar neither sonar) worth of verification in the recent iterations:

- I don't remember to have detect visually a periscope never (only the whole submarine at periscope depth).

- I add a sonar named "Periscope" to the SSs, as after repeated experimentation, because the SS without sonar detects visually nothing (and the range of the sonar named "Periscope" was only a minumum 0,1 nm). After added the "Periscope" sonar, the SS apparently detects visually targets at long range (range very higher than 0,1 nm.).

- I want to reproduce historical situations as the MAS 15 attacking Szent Istvan , but the apparent imposibility of approaching undetected (only visually, of course) the PT boat to the battleship determines the long-range destruction of the PT boat (the battleship detects EVER the PT boat at long range, she can never approach the battleship and fire the torpedoes), the failure of the scenario, and it dissapointed me as for forget all the episode.

 

I think we need a lot of comprobations and perhaps to reduce the visual detection range, but I'm not sure of it or about how to reach our aim. But it can be worst, I remember some years ago comments in the fora about the CGN Long Beach detecting visually in bad weather a PT boat at 70 nm or so (very far away of the horizon limit of 27 nm ), but it was in some iteration of ANW, not in a version of Harpoon Classic .....

Calculations for visual sight range. Not quite finished yet, but you can certainly fill in the blanks & check whether the numbers in the manual are correct. One thing you can see here is that the line of sight range between 2 objects mostly is determined by the height of the higher object.

 

Visual_Sight_Range.doc

 

Calculations for visual sight range. Not quite finished yet, but you can certainly fill in the blanks & check whether the numbers in the manual are correct. One thing you can see here is that the line of sight range between 2 objects mostly is determined by the height of the higher object.

 

Visual_Sight_Range.doc

 

From your calculations and comments I'm thinking hard about my (30 year old) memories.

I'm thinking in good conditions line of sight is the limiting factor while in poor weather actual visibility is the limiting factor. At night there is a reduction but the good/poor weather is more important.

In good conditions visual range of a ship (from ship or aircraft) depends on mast height of observed ship and 'eye' of observer (altitude of pilot, deck height (ship size) of sailor.

I think this is close to what is happening in Harpoon but the distances seem very short of your calculations which seem spot on, what are others finding?

In poor weather I think the size of the target becomes more important than line of sight distance. For testing I'm having trouble getting any bad weather, I have realistic weather box selected but the clouds don't come. Is weather location dependant at all or something else?

Don