Finding the range the hard way.

[Telemon]

Trying to do it the hard way I’m reading the method of determining range from The Submarine Torpedo Fire Control Manual (http://hnsa.org/doc/attack/index.htm) found in Sub Skipper's Bag of Tricks-Techniques, tactics, tutorials, videos, in the forum and am somewhat puzzled.
The manual is reputed to be an official document and I am surprised that it seemingly contains errors. These errors are confusing me and I hope some one will clarify things for me.
From the Manual, page 5-2. The details of the periscopes states that the optical length of the Type II ‘scope is 40ft and that of the type IV 36ft. No other lengths are given.
The next paragraph states’ …examination of the tables reveals that …we have had to sacrifice about 6ft of periscope depth’. I have examined the tables but can only see a change of depth of 4ft! Where is the other 2ft?

Page 5-3. Reading the first paragraph on the page we are told in the figure (Plate II?) the target subtends 5 divisions (of the reticule) in high power and 1¼ divisions in low power. The Manual goes on to state ‘That it is known that at 1000yds 17 ½ yards or 52.5 ft subtends an angle of 1 degree’. It then says that ‘Using this information we can deduce the following formulas(sic):
The formulae given are. R(range) = (19.1h)/n
R(range) = (76.2h)/N

Where R = range in yards.
h = height in feet.
n = number scale divisions in low power.
N = number scale divisions in high power.

From the information given can some one explain to me how the figures 19.1 and 76.2 are derived and how they are related to 17.5 and 52.5?

Finally the figures referred to in Plate II.
The two worked examples.

(a) For the upper diagram; Range = 76.2 x 120 /5 = 1840 yards.

(b) For the lower diagram; Range = 120 x 19.1h/1.25 = 1840 yards.

Both answers are incorrect!

Upper dia. Range = 76.2 x 120 /5 = 1828.8 yards (rounded up would give 1830 yards)
Lower dia. Range = 120 x 19.1/1.25 = 1833.6 yards (rounded down would give 1830 yards)

Again can some one explain the discrepancy please or point out where I am misreading?

[WernherVonTrapp]

With my math, I certainly cannot provide the assistance that I think you need. Bearing that in mind, please bear with me as I try to understand why you are trying to figure this out, which if your math answers are correct, you already have.
The Fire Control Manual that you reference seems to have been distributed in the 1950s, after WWII. Is it possible that some postwar modifications had been made? Additionally, according to the official USS Cobia website (the sub from which that Fire Control Manual belongs), that particular Gato Class sub was a product of the Electric Boat Company of Groton, as opposed to the Manitowoc shipbuilders which also built Gato Class subs.
Might some of these variables account for the discrepancies that you point out? Again, my math is about as sharp as a soup spoon so I cannot offer assistance in that manner but, I am trying to aid you in an alternative manner.

[commandosolo2009]

why even complicate things? you got SONAR and now prefer doing it with a periscope? is this like hey Germans, we Americans dont need sonar anyways challenge? stop wasting time, head to the boat, order a fresh pair of cob pipes and use the sonar, the American way :yeah:

[razark]

why even complicate things?
Because sonar emits noise, and there's more than one way of doing things.

[commandosolo2009]

Because sonar emits noise, and there's more than one way of doing things.

Speaking of which, if I ping after the DD sonar, would they still discover me? If you seen ["Below" - 2002 ]?

[Rockin Robbins]

Okay, let me take the questions one at a time. In the first instance it isn't apparent in the table why the sub would have to give up six feet in periscope depth for a 4' difference in periscope length. It possibly could be that the radar had to be further out of the water to operate satisfactorily. Then they could figure that 4 is approximately equal to 6. They don't explain.

The second question is related to trigonometry. We're just taking the ajacent side of a very long right triangle and dividing it by the opposite side. That ratio is called the cosecant of the angle. To subtend one degree and object is 57.2986885... let's call it 57.3 times further away than it is tall. We can use that on a galaxy. If we think it is 100,000 light years across and it is one degree wide, then that galaxy has to be 5,793,000 light years away.

Notice that all our measurements can be in any unit at all, miles, feet, yards, light-years, it doesn't matter as long as the units are the same. Here they toss a curve in by measuring the masthead height in feet and the range in yards. So for our 1º angle, We have to take our computed distance and divide by three, or we could just divide our cosecant, 57.3 by 3 in our calculation. 57.3/3 equals 19.1. So now we know that for an object subtending 1º, the formula is Range=19.1 times height in feet. Looks suspiciously like part of the above formula!

So if each division is one degree, how much closer would an object be if it subtended two degrees? You can easily see that it would be the range at 1 degree divided by two. Now n is the number of divisions (degrees at low power). So take your range at one degree, 19.1h and divide it by the number of degrees and you have calculated the range: 19.1h/n=R.

Now if you have the range and want to compute the height, then you have to change the formula. Since we know the range, we have to multiply it by the ratio of the opposite side over the ajacent side of the triangle. That is called the sine and the sine of 1º is .01745.... okay, we'll call it .0175. So if R=1000 our height is .0175 x 1000 or 17.5 yards. And that is how 17.5 yards compares to 19.1 times the height in feet. They are two completely separate numbers used for two different purposes.


The calculations in plate 2 are just incorrect as you have shown. Fortunately the mistakes are still accurate enough to maintain the integerity of the firing solution as they are only a half a percent off the real range, much closer than the accuracy of their measuring device. It is very probable that they used a slide rule to calculate the formulae, and so just reported what they saw, which would be a close approximation of the real answer. That would explain the inaccuracy. Any error from the slide rule would be much less consequential than a garden variety arithmetic mistake, which could be of any magnitude, where the error from the slide rule is, as I showed, of such a small degree as to be inconsequential.

[Joe S]

Is there a formula that works for periscope range finding in the game? I was under the impression that the magnification in the game did not match that of the real persicopes. At any rate, I have never been able to manually use the periscope range finder in the game. I have been able to do it with the split image device but not by observing the number of graduations and then doing the math. If it can be done I would like to try it. Thanks, Joe S

[Rockin Robbins]

The short answer is no because the angle subtended by one division is dependent on your screen resolution. Last I knew Capn Scurvy was working on that and had a solution in the works.

[Joe S]

Thanks! I like short answers! Joe S

[Joe S]

I have been thinking about the short answer. Is the problem related to the mastheight? If so, we could use the height of the stack. It is much easier to see and less likely to be distorted by the resolution. Joe S

[Telemon]

Thanks every one, for your comments particularly Rockin Robbins. This is a very interesting subject.

A point Rockin. You suggested, I think, that the formula is invalidated, or is dependant upon a particular screen resolutuion "...because the angle subtended by one division is dependent on your screen resolution". But surely screen resolution is irrelevant. If one knows the angular value of each division of the reticule then the angle subtended will always be the same.

My practice is to use the tangent rather than the cosec, this gives the range to the target at the waterline rather than to the masthead. My reasoning is that although, at longer ranges, the difference is small and can be accepted at shorter ranges i.e below 1000yds the error can be significant. I like to get in to c750 yds.

Joe S. I agree with you that trying to get a range using the stadimeter especially with 'no stabilised view' selected is a nightmare. Howver, why shouldn't you use the stack! If you look at the manual it is possible to estimate the height of the stack or any other part of the target, bridge, derrick et al. then do the trig calculation to get the range. Be aware though that because you are using a smaller height the error in range will be greater. On a pitch black night with rain though, it might be better than nothing.

Incidently I use an old Thorton PP221 slide rule fo my calculations.

[Hitman]

The short answer is no because the angle subtended by one division is dependent on your screen resolution. Last I knew Capn Scurvy was working on that and had a solution in the works.

IIRC there is a problem with one specific resolution, but in all others the image is correctly scaled.

The thread is here, for anyone interested in further reading: http://www.subsim.com/radioroom/showthread.php?t=175729

[Rockin Robbins]

There's a natural tendency to want to pick an easy to see aspect of the target: stack, cabin height, flight deck, but you probably wonder why the real guys liked to use the masthead height. By the way, their recognition manuals tended to give heights for stacks, masthead, cabin top and deck of most targets. Real sub skippers could pick whatever they wanted for their stadimeter range measurements.

But the accuracy of your distance calculation depends on three factors. The closer the target is, the more accurate your determination will be of course.:D The more accurate your stadimeter angle measurement is the more accurate your distance will be.

But also the higher the measurement point is on the target, the more accurate your distance determination will be. Your error for a masthead height of 50' will be half that of a distance determination based on a stack height of 25' for instance. Your accuracy is directly proportional to the height of the aspect being measured. So whenever they could the real sub skippers used masthead heights, the tallest part of any target.

Stadimeters are nasty inaccurate enough without you handicapping yourself!:yep:

[Joe S]

Thanks for the explanation about the potential for error. Would it not be true, however, that in a situation like ours, where there is a tendency for the image of the mastheight to be corrupted due to screen graphics issues, that the stack would be a more reliable measurement to use even though it might be theoretically less accurate? Joe S

[Rockin Robbins]

I don't know about graphics corruption issues. I don't have any. I'd have to say the jury is out on that one. However, if the mast is twice as tall as the stack, then your error tolerance is doubled: you can have twice the error using the mast and still have the same distance error resulting. It's difficult for me to imagine that you could have twice the error sighting on the masthead. You might get a little more error, but it would still result in a more accurate distance measurement.

It can be difficult to resolve the masthead at extreme range. However, at that range, small errors make huge differences in your result, so range measurements that far out should be viewed as exercises in futility anyway. They just give you a starting number that you know will be very different before you shoot, no matter what aspect of the target you are measuring.

[Joe S]

In my experience with this sim, the image of the mast is often times very inconsistent. It will apprear to be a certain height and then the next second its only about 2/3 as tall, which I always assumed was due to the inabilitly of the computer to clearly display the image. With a more substantial object such as a smokestack, this image on the screen remains stable, and it would seem to me that you would get a more reliable result for that reason. I havent tested this idea, but it makes sense, at least in theory. Joe S

[Rockin Robbins]

That's interesting. I've never seen that effect. I wonder which video cards are affected?

[TorpX]

In my experience with this sim, the image of the mast is often times very inconsistent. It will apprear to be a certain height and then the next second its only about 2/3 as tall, which I always assumed was due to the inabilitly of the computer to clearly display the image. With a more substantial object such as a smokestack, this image on the screen remains stable, and it would seem to me that you would get a more reliable result for that reason. I havent tested this idea, but it makes sense, at least in theory. Joe S

I've noticed the same thing. I assumed this was from the software simulating haze, mirage or any RL visibility factors. It definately makes accurate range estimates a lot harder.

However, as RR points out, the main purpose of long range plotting is to allow you to con the boat and reach a favorable attack position. There is no need to compute a final torpedo firing solution when the target is at long range, at least not ordinarily.

[CapnScurvy]

In my experience with this sim, the image of the mast is often times very inconsistent. It will apprear to be a certain height and then the next second its only about 2/3 as tall, which I always assumed was due to the inabilitly of the computer to clearly display the image. With a more substantial object such as a smokestack, this image on the screen remains stable, and it would seem to me that you would get a more reliable result for that reason. I havent tested this idea, but it makes sense, at least in theory.



I've noticed the same thing. I assumed this was from the software simulating haze, mirage or any RL visibility factors. It definately makes accurate range estimates a lot harder.

However, as RR points out, the main purpose of long range plotting is to allow you to con the boat and reach a favorable attack position. There is no need to compute a final torpedo firing solution when the target is at long range, at least not ordinarily.



The observation that the mast head "disappears" from sight at distance is very real in this game. I don't know if its just the game resolutions not giving the graphics card enough information to keep them from flickering, or some other graphic "setting" that may help to eliminate this problem (whether its a setting in the game options or through the graphics card itself). Possibly it's a problem specific to the hardware we have sitting on our desks that create this issue but the point is, small pixels of an image do seem to disappear from view at distances they shouldn't. The top of a mast head is a prime example. Without being able to "see" the mast head makes manual targeting with the Stadimeter very unreliable.

Unlike the real world sub commanders, the game only allows for one "height" reference point to be used when calculating range from the Stadimeter. The game chose the top of the mast head. Real world captains had any position on a target to use for the Stadimeter reading as long as a reasonably accurate height dimension could be used in the trigonometry equation. The fact that some of the mast heights in the game are so far off from being reasonably accurate, prompted me to correct them with SCAF. And yes, I took the liberty to use other height reference points besides the mast head for the very reason of the graphic disappearance.

Double R is quite right that the lower the height measurement the greater the error can be if your off a bit with the Stadimeter positioning. At close range the game expects the mast height to be towards the top of the periscopes view (towards the top of the area between the waterline and the top edge of the periscope lens). When a Stadimeter reading is obtained in this area, a misplaced stadimeter position (by just one or two pixel lines) will give an error of only a few (maybe up to 10 or so) yards/meters in range. On the other hand at longer distances from the target, the same height reference point will appear lower within this periscope view area. The error of just one or two pixel lines may create a 15 to 40 yard/meter difference in found range (it will increase even more the closer to the water line you go).

You may ask, How wide is a pixel line anyway? The stock game Telemeter division marks (those are the small marks on the periscope lens) have three pixel lines; one at the top, one in the middle to separate, one at the bottom of a two pixel width line. Add the fact that some mast heights are off by several yards/meters (at a 1500 yard distance a single yard/meter can throw off the calculation by as much as 50 yards/meters; the Hiryu CV should have a mast height of 37.4 meters not 31.0 !!) the stock game puts quite a handicap on manual targeting.

I'm almost ready to release a mod that will give reasonably accurate height and length measurements found in the Recognition Manual. It will allow you the use the periscope Telemeter divisions (the lens marks on the periscope) to calculate "Range" with the aid of a device called an Omnimeter. It will also allow you to obtain "Angle on Bow" of a target by knowing its actual length and measuring the "angled" length of the target counted by the Telemeter divisions. The Omnimeter will use this "difference" to calculate the AoB. An added "Range Dial" will be placed in the game to input a manually found range distance into the TDC for a torpedo firing solution. The American side will be able to use the Position Keeper with the inputted range from the Range Dial to "track" a target just like the Stadimeter or Sonar. These additions coupled with having the stock game optics corrected to what the world view should be sized to (which are well off too), will greatly improve game play when using manual targeting.


=============


Telemon, as Double R pointed out the actual quoted figures in the "Torpedo Fire Control Manual" are wrong. He's quite right that the errors are very minor, so the instruction is still helpful regardless of inaccuracy.

The corrected figures should read:

Page 5-3. Reading the first paragraph on the page we are told in the figure (Plate II?) the target subtends 5 divisions (of the reticule) in high power and 1¼ divisions in low power. The Manual goes on to state ‘That it is known that at 1000yds 17 ½ yards or 52.356 ft subtends an angle of 1 degree’. It then says that ‘Using this information we can deduce the following formulas(sic):
The formulae given are. R(range) = (19.1h)/n
R(range) = (76.4h)/N

Where R = range in yards.
h = height in feet.
n = number scale divisions in low power.
N = number scale divisions in high power.

From the information given can some one explain to me how the figures 19.1 and 76.4 are derived and how they are related to 17.5 and 52.356?

This is the beauty of the American (and German) periscope. The magnification difference between low power 1.5x and high power 6.0x, are multiples of 4. The formula 19.1 times 4=76.4 are directly related to 1.5x magnification times 4=6x magnification. This multiple is also carried over onto the periscope lens having Telemeter division marks of 8 large divisions marks and 32 smaller division sections. The lens itself was manufactured to have a 32 degree Field of View at low power, an 8 degree FoV at high power (you see how all the pieces fit together!!). It's a known fact that one degree of length is equal to 52.356 feet (or 17.452 yards) at a distance of 1000 yards (we're getting into longitude and latitude, curvature of the earth, degrees, minutes, seconds type stuff now), so using these known factors with the formula makes range finding possible.

Finally the figures referred to in Plate II.
The two worked examples.

(a) For the upper diagram; Range = 76.4 x 120 /5 = 1833.6 yards.

(b) For the lower diagram; Range = 120 x 19.1h/1.25 = 1833.6 yards.

It's hard to figure anything out if you don't get the right formula to start with!!! I wonder if these errors weren't some kind of diversion to keep us simpletons from figuring out this stuff. Or did the guy who proof read this material not know how to use a pencil (or think we wouldn't). :88)

[TorpX]

The observation that the mast head "disappears" from sight at distance is very real in this game. I don't know if its just the game resolutions not giving the graphics card enough information to keep them from flickering, or some other graphic "setting" that may help to eliminate this problem (whether its a setting in the game options or through the graphics card itself). Possibly it's a problem specific to the hardware we have sitting on our desks that create this issue but the point is, small pixels of an image do seem to disappear from view at distances they shouldn't.

Very interesting. Might someone be able to figure out the cause?




Unlike the real world sub commanders, the game only allows for one "height" reference point to be used when calculating range from the Stadimeter. The game chose the top of the mast head. Real world captains had any position on a target to use for the Stadimeter reading as long as a reasonably accurate height dimension could be used in the trigonometry equation. The fact that some of the mast heights in the game are so far off from being reasonably accurate, prompted me to correct them with SCAF. And yes, I took the liberty to use other height reference points besides the mast head for the very reason of the graphic disappearance.




I like the idea of having other reference points, but think the highest point is usually the best. At long range, part of the hull will be below the horizon further reducing the accuracy of any lower pt estimate. Also, its simpler to use mast height for all ships.

Is SCAF incorporated in RFB?




I'm almost ready to release a mod that will give reasonably accurate height and length measurements found in the Recognition Manual. It will allow you the use the periscope Telemeter divisions (the lens marks on the periscope) to calculate "Range" with the aid of a device called an Omnimeter. It will also allow you to obtain "Angle on Bow" of a target by knowing its actual length and measuring the "angled" length of the target counted by the Telemeter divisions. The Omnimeter will use this "difference" to calculate the AoB. An added "Range Dial" will be placed in the game to input a manually found range distance into the TDC for a torpedo firing solution. The American side will be able to use the Position Keeper with the inputted range from the Range Dial to "track" a target just like the Stadimeter or Sonar. These additions coupled with having the stock game optics corrected to what the world view should be sized to (which are well off too), will greatly improve game play when using manual targeting.




Yes, I've dropped in to monitor your progress. It looks good. :up:


Am I right in assuming that we will still be able to use the stadimeter?

[Rockin Robbins]

Newcomers, if you don't know CapnScurvy he's spent more time with the stadimeter/periscope system than you've spent playing the game. Matter of fact, probably more than I have spent playing the game. He's by far the authority on this area of Silent Hunter 4 and y'all need to pay attention to his posts. Even including the supermodders, there's nobody who has contributed more to the community and more to the great state of SH4. Scurvy isn't a prima donna and he shoots from the hip. Accurately I might add.:salute:

Capn, I still bet the reason for the "errors" was rounding error coming from the use of slide rules to perform all the calculations. Try it yourself and what you read off the slide rule is what they use for an answer. Real geeks at the time would never perform an arithmetic calculation unless a gun were pointed at their head!

[joegrundman]

Unlike the real world sub commanders, the game only allows for one "height" reference point to be used when calculating range from the Stadimeter. The game chose the top of the mast head. Real world captains had any position on a target to use for the Stadimeter reading as long as a reasonably accurate height dimension could be used in the trigonometry equation. The fact that some of the mast heights in the game are so far off from being reasonably accurate, prompted me to correct them with SCAF. And yes, I took the liberty to use other height reference points besides the mast head for the very reason of the graphic disappearance.



you know i was under the impression that SH4, as opposed to SH3, did in fact allow any reference point to be used, simply by manually setting the stadimeter to whatever "mastheight" you desired. I understand that by clicking the checkbox in the recog manual it always chose the masthead height, but by not checking the box, and instead manually setting the masthead height, you could in fact set it to whatever height you fancied.

is this not your experience?

[TorpX]

I think in RFB you can use any point you want, but once you click on the rec manual icon, the stadimeter keeps using the default value. Normally, I like to use the rec manual, since I don't have a ready listing of all the possible reference points and it seems simpler.

[joegrundman]

there is a diagram and a scale in the recog manual, both of which can be used to determine the height of funnels or top of bridge, or indeed anything

- and this is a stock feature, not rfb

[TorpX]

Yes, I know there are the horizontal lines, but isn't the scale different for every ship? Without a known scale it isn't very useful.