Stadimeter

[darksythe]

Im back been awhile since i been here but starting to get back in to the subsim scene. Got a ? for anyone who can help.

I am attempting to do manual TMA and ive got everything down to mathematical precision however anytime i i use the stadimeter i keep getting incorrect range. I know they're incorrect because i am trying it out on the torpedo training mission after a few failed attempts i turned on auto tma and got the correct ranging from the crew men reverted back into manual tma inputed the range manually and did the rest of my tma by the books so to speak and viola hit with all 3 fish.

I have found that if i manually input the mast hight into the stadimeter i get a closer match to the range.

[CapnScurvy]

The mast heights used in the game are inaccurate on almost every ship (allies or axis). The point is, you can hit a target using manually found range if your close enough. So stick your nose into the gut of the target and pull the trigger. You can do it with your eyes closed!!

An example of the inaccuracy is the Jap carrier Hiryu. If you use the games listed mast height for finding range in about a true 1200 yards distance, the manually found range will be about 500 yards off the 1200 yards distance. You'd better be real close to the Hiryu to get a hit on her!!

I know, there are those that say the inaccurate recognition manual mast heights are just "true reality" being displayed in the game. I say bull!! No sub Captain would have knowingly accepted the outcome of inaccurate range finding based on a manuals inaccurate estimate of mast heights. You can bet if the Captain could have figured out the correction he'd be "singing like a canary" about his findings when he returned to the Officers Club at the end of his first patrol so no other sub Captain would risk his life or his crew on inaccurate range finding.

I'd read the "Discussion" section in this (http://www.subsim.com/radioroom/showthread.php?t=126016) thread and try to use a version of SCAF that is compatible to what you play. Sorry, I don't have the time to keep up with every new ship some of the mods enter into the game. So making a compatible version for every mod out there is out of the question. Just rest assured you're not alone in seeing the inaccuracy of manually finding range.

[darksythe]

Thanks for the reply ill check that thread.

[dezertflyr]

Reading through the Submarine Torpedo Fire Control Manual (http://hnsa.org/doc/attack/index.htm) you will find several interesting tidbits in regard to the stadimeter:


In Plate III (http://hnsa.org/doc/attack/index.htm#plateiii), a picture of a stadimeter scale, a masthead height of 60 feet gives a range of 2300 yards. Note that the scale is constructed for high power observation. When ranges are measured in low power the computed value must be divided by four.


The following points should be kept in mind in height determination:
(1) Masthead heights may be purposely altered by the enemy to cause inaccuracies in periscope ranges.


(2) Tops of masts may be camouflaged in such a manner as to be invisible under average visibility conditions at any except short ranges.
And a personal fav...


Since the accuracy of the ranges is variable when using the stadimeter or telemeter the Approach Officer should inform the Fire Control Party when he considers the range accurate or doubtful.
Verrrrly interestink :03:

[DennyC]

When beyond 2000 yards, I commonly get stadimeter range errors of 100-500 yards. I can iterate the stadimeter measurement until it looks good on the plot, but it is frustrating to just replace one bad range with another until I get lucky. I've had better luck using either of these methods:

1: If the calculated range seems to be long, this means the angle was too small. Cheat the next stadimeter shot by aligning the masthead with the waterline minus a few feet. (Increase the angle.) Then check the plot again.

2. If you have time, you can tweak the PK's range by entering a temporary AOB that is "wrong", but one that will cause the PK's virtual range to more rapidly approach or recede from the target's actual range. In a few minutes, when you get better agreement, then return the AOB to the actual AOB as measured from your plot or estimated visually. This is a good time to update the bearing, as well. Then hide the scope, except for updates. I wish there were a way to enter ranges manually, as on a real sub.

My results have been improved to the point that I now use the stadimeter and PK more than I used to. For example, I can use a good PK solution to fire at target one, then get off an O'Kane shot at a second target that might be conveniently approaching a 90 degree AOB.

Target shifting checklist: Do it quickly.

0. switch to next tube and open it if necessary
1. turn off Pk
2. set speed to 0
3. turn scope or TBT to bearing that corresponds to 90 AOB (near 0 if you on a course that has target the crossing your T.)
4. click the update button and MAKE SURE the gyro angle does something reasonable. If not, either repeat or abort.
5. set the scope to the appropriate lead angle (per O'Kane method) and shoot as target crosses.
6. Beat it and go hide.


The Mk 10's seem to have a lower dud rate on the 90 shots, but as they are slower, your ships have more time to react to the first shots and evade.

[Fishbreath]

Get the TGT Dials to PK Fix mod--it unlocks the range wheel from the mast height thingy (or just replaces it; I'm not sure). Either way, it fixes the lack of manual range entering.

[Joe S]

You really do not need the mast height. You can get an accurate reading from any known object on the target for which you know the height, such as the top of the bridge, a turret, stack etc. The original target recognition showed the target against a scale by which you could determine the height of any part of the ship. Joe S

[Rockin Robbins]

Stadimeter ranges, even with unrealistically perfect information, are notoriously inaccurate. When Dick O'Kane lost his radar, the only reliable range finder available, he fired off a sarcastic message to Pearl amounting to "Drat! (use your own word here--he did!) Now I'll use twice as many torpedoes to sink a target."

And it's true. In the war, not only did they not have a recognition manual showing even half the Japanese shipping out there, but they consistently misidentified what they did "identify," frequently over or understating dimensions and tonnage by a factor of 2 to 3. Those errors were way outside of the error range in SH4 stadimeter lookup tables. We are way too efficient at hitting targets, even with the bad data CapnScurvy says captains wouldn't tolerate. In fact, our database would have been the envy of the fleet. Then CapnScurvy fixed it to not only contain every ship on the ocean (unhistorical), but to have perfect accuracy on every ship (also unhistorical).

Whether you think that is reasonable or not, you'll STILL get inaccurate stadimeter readings, depending on the ratio between the target distance and the height of the object being measured. Let's take an aircraft carrier, for instance. The flight deck is roughly half the height of the tallest mast. If you take the measurement from the flight deck because it is easiest, as opposed to said mastheight, your error envelope is doubled! That is, one pixel in measurement error equals twice the error distance when you use the flight deck. Unfortunately, in a digital world, where the size of a pixel is significant in our instance, a one pixel error can result in a big range error.

Your expert on that is CapnScurvy, who took months developing his technique, collecting and collating data and then consolidating it into SCAF. You'll find that an error of one pixel at a real distance of half to a full mile is appalling.

Actually the stadimeter is appalling and should be used as it was in the war: the last resort of a desparate captain, seeking any way to get a best guess shot off when otherwise he'll lose the target. Consider any hits garnered with the help of the stadimeter to be a bonus, with or without SCAF.

There's a reason the Germans, with a much less sophisticated TDC, had twice the hit percentage of the Americans with their primitive U-Boats. They used the stadimeter much less. Capitalize the word much.:up:

[Henson]

Adding to what RR said, a stadimeter range is not a linear function, it is exponential. I'm not sure if the game models this well, but in real life 1 division at a close range may eqaul just a hundred or so yards, while 1 division at a distant range may be equal to a couple thousand yards.

Most submarine TMA ranges work that way, which is why submariners habitually attach a confidence level to each range. My own rule of thumb is to instinctively distrust the accuracy of any range outside of a particular specified number of yards. Add that to the fact that many submarine officers subconsciously undercall ranges to leave a margin of safety, and you have a lot of doubt in the range/speed solution at periscope depth. The good news is that the arrival of digital homing torpedoes has greatly reduced the need for a gnats-ass target solution.

This problem with stadimeter ranges is not just a WWII thing.

[jerm138]

Actually the stadimeter is appalling and should be used as it was in the war: the last resort of a desparate captain, seeking any way to get a best guess shot off when otherwise he'll lose the target. Consider any hits garnered with the help of the stadimeter to be a bonus, with or without SCAF.

Hmm... I didn't know that. I must have been doing it wrong all along. :hmmm:

From the books I've read, it seemed pretty standard to use it. If not the Stadimeter, what are they using when they raise the scope and say "Range... Mark. Bearing... Mark." I always assumed they were having someone mark the range readout on the stadimeter dial.

Also, this is how I perform almost ALL of my attacks when not using automatic contact updates. I get a very respectable success rate too. My best was a 3000 yard shot at a 4000 ton freighter... blew the boilers and broke his back with the first fish... the next two missed forward, but only because he had stopped.

I'll usually use sonar to get rough initial data to intercept them, but once I'm in visual range, I use the stadimeter to send range and bearing data to the TDC. How else would you do get range? Eyeball it? I trust the stadimeter more than I trust my guesstimates. I only really trust it within a few thousand yards though. It's hard to get an accurate measurement on an object that is 5 pixels high.

[Joe S]

There were (are) a number of ways to estimate range : The target ID Books had a grid superimposed over the target which the "ship beyond the horizon" and "Horizon beyond the ship" by which one could estimate range. In the later years of the war, the attack scope had a radar unit in it which could give a very accurate range estimate. You can use the split image method as included in SHIV, for which you need to know the mastheight, or the Telemeter ranging by which the range is calculatd by measuring how many divisions on the verticle recticle are covered by an object, such as the mast or funnel of the target , for which you know the height. Unless you are doing a COMPLETELY manual firing solution, including drawing everything out by hand , I would not worry too much about range, it is the least important piece of data needed for an accurate solution. In fact, it is almost irrelevent if you have an accurate relaltive bearing, which is easy, and an accurate target speed, which is the tricky part. Joe S

[jerm138]

Unless you are doing a COMPLETELY manual firing solution, including drawing everything out by hand , I would not worry too much about range

This is how I've been playing lately... or at least trying to. I used to argue that plotting manually was unrealistic because the skipper wouldn't do that in real life. But once I tried it a few times, I saw the flaw in that argument... the guys plotting are just going off of the reports the skipper makes and so the plot is only as accurate as your estimations of range, speed, and AOB. Having contact updates off really tests your salt a little more.


In fact, it is almost irrelevent if you have an accurate relaltive bearing, which is easy, and an accurate target speed, which is the tricky part.

I don't see how range would be almost irrelevant. If a ship is doing 12 knots and he is 300 yards further away than the TDC thinks he is, I'm probably going to miss aft. I know that the constant bearing method eliminates some of the error, but the difference in target speed and torpedo speed still leaves some inaccuracy there.

Of course, the closer you are, the more likely you are to hit, regardless of minor errors.

I agree that getting the speed is the trickiest part. Sometimes, if I'm a little unsure about the speed, but pretty confident on range and AOB... I'll use the speed setting in the TDC to set my spread, aiming at the middle the whole time.

Let's say I estimate his speed to be between 8 and 12 knots. I'll set it at 8 for the first shot, 10 for the second, and 12 for the third. I have the PK running, so I don't have to redo the range/bearing every time. This will set up a stern > middle > bow type spread and boost my chances of a hit. It's a bit unorthodox, I know, but most of the best skippers had their own strategies too.

[Joe S]

For purposes of the firing solution, range is irrelevent. It has been proven many times. If you draw out an example on a piece of paper you will see why. Draw an example with a target on a course 90 degrees from your own , in other words, straight across your bow, a perfect setup. Assume the range is 1,000 yds. Figure out how long it will take your torpedo to hit the targets track. Then calculate how far the target will travel during that time based on the speed of the target. That figure tell you how much you must lead the target. If the target travels 300 yds during the run of the torpedo, you must lead the target by 600 yds to get a hit. Put a dot on the paper representing the point on the target traci, 300 yds in front of the target, where the target and torpedo will intersect. Now, on the same piece of paper, put another target with the same speed and relative bearing, out at 2,000 yds and do the same thing. Do it as many times as you want on the same piece of paper. You will notice that the track of the torpedo, representing the lead angle, remains the same no matter how far away the target is. This means that once you calculate the lead in degrees of angle, you can use that to aim the torpedoes no matter what the range. Say for example the lead is eight degrees. Your target is at 350 degrees. Aim the torpedo eight degrees in front of the target, say 358 degrees if the target is moving left to right, and fire!. I am at work and cannot take time to be more specific, but I could do it later. Try it and let me know what you come up with . As soon as you see it on paper you will get the idea. Joe S

[Hitman]

I don't see how range would be almost irrelevant. If a ship is doing 12 knots and he is 300 yards further away than the TDC thinks he is, I'm probably going to miss aft. I know that the constant bearing method eliminates some of the error, but the difference in target speed and torpedo speed still leaves some inaccuracy there.

Range is irrelevant in zero Gyro Angle shots, i.e. when your torpedo just follows a straight course form your bow. The reason is proportional triangles: If the target is 300 yards further away, so is also your torpedo proportionally from the estimated impact point, and hence the result is the same.

Things start to change however with angled shots, when convergence correction enters the equation ...

If you want to know more about the convergence, see this post I did some time ago: http://www.subsim.com/radioroom/showthread.php?t=121744

[Hitman]

When beyond 2000 yards, I commonly get stadimeter range errors of 100-500 yards. I can iterate the stadimeter measurement until it looks good on the plot, but it is frustrating to just replace one bad range with another until I get lucky.

Stadimeter ranging was inacurate in real life and even more in the game, where no current screen resolution no matter how high can match the density of pixels in our eyeballs. Tips of masts are constantly appearing/dissapearing and make distance reading very difficult. This is the reason why you must simply AVERAGE the results by drawing a line through the average centre of them. And that will give you something good enough to shoot.

There's a reason the Germans, with a much less sophisticated TDC, had twice the hit percentage of the Americans with their primitive U-Boats. They used the stadimeter much less. Capitalize the word much.:up:


Yup, they estimated by naked eye a lot, counting on the fact that they would attack on surface, with a good sight of the target,a nd not by peering through a monocular, thin & low light admiting (Hence dark) periscope. And they got close, VERY close and shot with almost no paralax (convergence). When you are under 1000 metres, at right angles to target and shooting with minimal parallax, estimating errors are not that important.

[jerm138]

Range is irrelevant in zero Gyro Angle shots, i.e. when your torpedo just follows a straight course form your bow. The reason is proportional triangles: If the target is 300 yards further away, so is also your torpedo proportionally from the estimated impact point, and hence the result is the same.

Things start to change however with angled shots, when convergence correction enters the equation ...

If you want to know more about the convergence, see this post I did some time ago: http://www.subsim.com/radioroom/showthread.php?t=121744

OH!!! I see now...
I didn't know we were only talking about zero Gyro angle shots. I see the difference now. Thanks for clearing it up.

[Joe S]

Hitman is obviously correct. However, it should be pointed out that hopefully, in the vast majority of your shooting, you will be amubushing the target from a nearly 90 degree angle. I would say the angle would have to be pretty bad before range gets to be an important factor. Do yourself a favor, save your torpedoes for the good setups, and pass on the poor setups. Just my opinion, fwiw. Sink em all! Joe S

[Rockin Robbins]

That's why the real boats didn't have a great hit percentage. They took the low probability shots that we pass on. They also tended to fire spreads for just about every target, and these were wide spreads, over 100% of the length of the target, guaranteeing multiple misses even when they hit the target. We tend to shoot so every torpedo should strike the target. We're playing a game and for them it was life and death.