Ranging and fire control problem

[Charos]

I would like to invite others to test and see if they conclude as I have that the Attack periscope as well as the TBT field of view and graticule markers are not correct.

This also affects the TDC as its using the wrong parameters.


The following screenshots are of a stationary Iowa class Battleship broadside With a
S-Class submarine also stationary.

The attack periscope should have a FOV of 8 Deg divided by 32 graticule markers of 0.25 Deg each. (In high magnification 6X setting)

The Iowa is 270M long and has a mast height of 46M.

Now for an 8 Deg FOV to see the Iowa fill the entire screen would require a range of
135M/Tan 4 Deg = 1930M .


Now for a mast height of 46M and a Range of 1406M (This range is from game stadiometer) we see the mast fills 5.5 Graticule markers in the Attack periscope (see attached screen shot)


http://members.optusnet.com.au/%7Etrentelliott/SH4/IowaStadiometer.jpg



http://members.optusnet.com.au/%7Etrentelliott/SH4/IowaGameMap.jpg


Each marker SHOULD be 0.25 Deg (IE 8 Deg/32 = 0.25) as per game manual.

Mast /Range = Tan (Theta) in this case Theta = 0.25*5.5 = 1.375 Deg.

Solving for Range = 1916M (Which is very close to the previous 1930M and within error tolerance reading from stadiometer)


Note the Iowa in SH4 fills the 6X attack scope at approx 1680M (see attached screen shot)

Also note that the TDC via the stadiometer provides us a range of 1406M (see attached screen shot)

Something is clearly wrong…Yes?


Now lets dig a little deeper into the problem………………


To cut a long story short the Graticule markers are not 0.25 Deg but approx 0.29 Deg.

135M = Half Iowa Length.
4.64 Deg = Half Attack Periscope FOV.

IE: 135M/4.64 Deg = Range
Range = 1663M (Which is correct in our example).


So that attack periscope markers on 6X and its associated FOV work off 0.29 Deg per graticule marker for its 9.3 Deg of view.

But what happened to our TDC its showing 1406M ????

Well its working off approx 0.34 Deg per marker for Range finding (IE: off the vertical graticule scale) which makes things even more strange.

Hope this all makes sense.

[joea]

Bug!!! ..can it be fixed or modded? My head hurts. :88) Nice work though.

[Joe S]

My math skills are limited to basic multiplication , division, addittion and subtraction. Based on your research, what is the actual magnification of the scope in high and low power? Once we know the magnification, how do we use the recticle to get the range of a target? thanks! Joe S

[Charos]

My math skills are limited to basic multiplication , division, addittion and subtraction. Based on your research, what is the actual magnification of the scope in high and low power? Once we know the magnification, how do we use the recticle to get the range of a target? thanks! Joe S

Joe going by the real life values of the following:


Type II
Magnification high power 6.0X
Magnification low power 1.5X
Maximum elevation of line of
sight (above horizontal) 74.5 degrees
Maximum depression of
line of sight (below horizontal) 10 degrees
True field high power 8 degrees
True field low power 32 degrees
Ranging Device Stadimeter
Telemeter
Scale

Outer diameter reduced section 1.414 in
Optical length 40 ft


Type IV
Magnification high power 6.0X
Magnification low power 1.5X
Maximum elevation of line of
sight (above horizontal) 45 degrees
Maximum depression of
line of sight (below horizontal) 10 degrees
True field high power 8 degrees
True field low power 32 degrees
Ranging Device Radar
Telemeter
Scale

Outer diameter reduced section 3.75 in
Optical length 36 ft


You can see that both the Observation and Attack scopes have 6X and 1.5X
magnification at 8 Deg and 32 Deg Field of view respectively.

You can see that the pattern:

8 Deg = 6X
32 Deg = 1.5X

Therefore 16 Deg = 3X
and 48 Deg = 1.0X

So the attack periscope on high magnification in SH4 provides approx 9.28 Deg FOV
or approx 5.17X magnification.


This whole thing to my mind is the most URGENT Bug in SH4 as it takes the whole TDC measurements out from what they should be.

[Charos]

Im bumping this.

To put this problem in simple terms it means the Games TDC is providing a 17% Ranging error on all firing solutions.:o

I hope the magnitude of this problem has not been lost in the we need a working chronometer white noise. :damn:

In my example above the TDC has clearly provided a range solution of 1,406M while the sub the TDC is sitting in is STATIONARY and approx 1,680M from the target.

[jmr]

Great investigative work here. Hopefully more experienced skippers than myself can give their input on your results. All I can add is that I've been practicing with Kim Ronof's Mark 3B range calculator and my results don't jibe with the auto TDC.

[SteamWake]

Very well done.

Lets hope someone can get to the root of this.

[joea]

This must go on the bug list. :yep:

[Charos]

Although im confident there is a major problem here I still require at least another person to verify this so as we can be 100% shore im not barking up the wrong tree.

Anybody?


Just fire up the mission editor and place one stationary sub and any stationary ship and check stadiometer distance with ingame map distance as a starter.

For confirmation they should be out by 17% from each other.

[akdavis]

Hmm...I've certainly noticed that the stadimeter does not measure range consistently, but it seems to vary from ship to ship.

[Charos]

Hmm...I've certainly noticed that the stadimeter does not measure range consistently, but it seems to vary from ship to ship.


I have tested this myself on the Iowa BB and the Kongo BB as well as the small tanker (which has the 20M mast) or 19.7M depending on which reference.

In any case my results so far in my testing have shown consistancy.

There may be some slight variance in boyancy of the vessels (if this is modeled) but the error could not be that high. and the horizontal graticule scale is the same as the vertical scale IE: 0.25 Deg - and as I have shown that is also off by the same amount as the vertical scale.

[Charos]

Hmm...I've certainly noticed that the stadimeter does not measure range consistently, but it seems to vary from ship to ship.

To be more specific the reason you most likely see a variance in distance error between TDC calculated range and actual range on game map is because the difference is a percentile error.

The error is approx 17% - the further you are from the target and or the smaller the mast height is the greater the difference in distance will be - but the difference is still 17%.

[SteamWake]

The comment that "it seems to vary from ship to ship" made a light bulb go off

Its not the Scope thats in error its the models ?

[akdavis]

Hmm...I've certainly noticed that the stadimeter does not measure range consistently, but it seems to vary from ship to ship.


I have tested this myself on the Iowa BB and the Kongo BB as well as the small tanker (which has the 20M mast) or 19.7M depending on which reference.

In any case my results so far in my testing have shown consistancy.

There may be some slight variance in boyancy of the vessels (if this is modeled) but the error could not be that high. and the horizontal graticule scale is the same as the vertical scale IE: 0.25 Deg - and as I have shown that is also off by the same amount as the vertical scale.

A simple test can be conducted in the training missions. In the artillery mission, you will get a correct range (as compared with map contact updates on) by measuring from the tip of the merchant's masts, as the documentation instructs you too. In the torpedo mission, you must lower the stadimeter image down further on the Mogami to get an accurate range. More like the first crosstrees rather than the top of the masts.

[Sailor Steve]

Picky, picky, picky!


But only because I probably never would have noticed, and certainly couldn't have figured it out mathematically. Hope you get it solved.:up:

[GSpector]

To make matters worse, I do believe UBI stated that there was an error in Torp speed and suggested only firing at Slow since the Fast setting calculates for a faster speed than Torp's actual speed causing Torp to hit Aft section of moving Targets :nope:

[anthrax]

The standimeter is pretty accurate. Its just they you have to be careful exactly on where you measure from.
1.) Measure from the higher point horizontal point on on the mast. (i.e, it highest "cross" shape.)
2.) Always exclude the flags.

Here are some examples. Here an Iowa class at 1400 yards. (~0.7 NM)

http://usera.imagecave.com/evilcommie/SH4/SH4Img17-4-2007_1.26.41_610.JPG

Map view

http://usera.imagecave.com/evilcommie/SH4/SH4Img17-4-2007_1.26.54_904.JPG

[anthrax]

Here is a second example. A T3 tanker.

http://usera.imagecave.com/evilcommie/SH4/SH4Img17-4-2007_1.36.8_59.JPG


http://usera.imagecave.com/evilcommie/SH4/SH4Img17-4-2007_1.36.43_687.JPG

[Charos]

Anthrax

I welcome all testing but firstly the most glaring observation with your conclusion I can make is that the T3 example is not ranged properly on the stadiometer.

Therefore repeatability of your conclusion is wanting.

IE: On your T3 Tanker your Stadiometer is under the waterline by 0.5 Graticule markers (which scews your result) instead of the TDC reading of 2.75 markers it should only be 2.25 markers (Useing your mast cross member theory)

In the T3 screen shot moving the Periscope X axis up to the waterline 0.5 markers would put the stadiometer at the very top of the T3 mast, not on the mast cross member.

Since the Stadiometer is a visual tool at great range error's will be lost - testing should be conducted on closer targets where it is easier to read the stadiometer markers off manually.

We need this issue resolved - It affects the entire games ability to connect torpedoes with a target which is what the general idea is in the first instance.

[Torpex752]

I dont recall the game manual telling us how its to be used. In navy Contact Coordinator School I was taught that its waterline to mast top, that was the same method used during WWII, I am quite sure of this.

Frank
:cool:

[Charos]

I dont recall the game manual telling us how its to be used. In navy Contact Coordinator School I was taught that its waterline to mast top, that was the same method used during WWII, I am quite sure of this.

Frank
:cool:


Thats the rational assumption Frank and the one we have to work with. :up:

As soon as we start picking arbitrary points on a ship - the results can be fudged to fit whatever we like.

[BlackSpot]

Anthrax

IE: On your T3 Tanker your Stadiometer is under the waterline by 0.5 Graticule markers (which scews your result) instead of the TDC reading of 2.75 markers it should only be 2.25 markers (Useing your mast cross member theory)

In the T3 screen shot moving the Periscope X axis up to the waterline 0.5 markers would put the stadiometer at the very top of the T3 mast, not on the mast cross member.

Hmmm...I've just tested this. It doesn't seem to matter whether it aligns with the waterline or not. Try it.

[Charos]

Anthrax

IE: On your T3 Tanker your Stadiometer is under the waterline by 0.5 Graticule markers (which scews your result) instead of the TDC reading of 2.75 markers it should only be 2.25 markers (Useing your mast cross member theory)

In the T3 screen shot moving the Periscope X axis up to the waterline 0.5 markers would put the stadiometer at the very top of the T3 mast, not on the mast cross member.
Hmmm...I've just tested this. It doesn't seem to matter whether it aligns with the waterline or not. Try it.
I assure you it does matter because thats your REFERENCE Point from which the stadiometer is measuring from - Your measuring the ship not the water in front of it.


I just tested a T3 with a S-Class broadside - BOTH stationary.

T3 - Had a loadout of Fuel.

Attack scope at high magnification used.

When T3 fills FOV broadside my sub is at 1,170M from target-which gives again a FOV of 9.3 Deg.

Stadiometer reads 1,040M range or an error of approx 12%. TDC Graticule markers read off 5 of them. Which calculates out to approx 0.3 Deg per marker (They should actualy be 0.25 Deg per marker)

T3 is 190M long with an 8Deg FOV I should see it fill my field of view at 1,360M.

This test represent much the same as before - my other tests reveal approx 17% deviation and this one provided 12% - still ballparking the same.


EDIT: A Torpedo at 36K released with this TDC setting would take approx 7 Seconds longer to reach target as the real distance is 130M further away than the TDC calculation.

[akdavis]

No, it doesn't matter. You can't bring the mirror image down below the original, no matter the orientation of the periscope. This:

http://img251.imageshack.us/img251/3938/stadimeterfc4.jpg

Returns the same, correct measurement as having the waterline centered. There is no variance. For the Mogami, the correct measurement is returned when using the first crosstree from the top. This is not the case with all vessels.

For the Medium Modern Composite in the artillery training mission, the correct range is obtained using a point above the very tips of the masts:
http://img251.imageshack.us/img251/1867/stadimeter2do6.jpg

Also, you are wasting your time using the ruler on the navmap to compare stadimeter range to actual range. Simply turn the TDC on and compare the TDC plot on the attack map with the ship's actual position.

[Charos]

No, it doesn't matter. You can't bring the mirror image down below the original, no matter the orientation of the periscope.

Also, you are wasting your time using the ruler on the navmap to compare stadimeter range to actual range. Simply turn the TDC on and compare the TDC plot on the attack map with the ship's actual position.

Firstly appologies to Blackspot because he and Akdavis are correct as far as the operation of the splt image stadimeter.

The split image utilises whatever elevation/depression is currently on the scope and then via the split image of known height moves a given angular movement to obtain range.

The scope takes the reading from its current declination as the reference point.

Ok good I have my head around that now.


Point 2: Useing the attack map provides the exact same distance error as the navmap (I just tried it with the T3 - no change) Why should it be any different?

EDIT: It is easier though when trying to manually convert graticule markers to range via math to use the X-Axis of the scope as it starts the measurement on a full marker - which eases
reading them off.

[Charos]

Just did some testing with a Mogami this time.

It seems to me we are dealing with a Hydra.

Krupp's mod covers some of this: http://www.subsim.com/radioroom/showthread.php?t=109953

I see the following:

#1 Models proprtion's do not seem consistant - as Krupp has found.

#2 Field of view of attack periscope and most likely TBT are incorrect (I have not focused on the TBT as yet)

#3 For those attempting to go full manual (without TDC) the graticule markers are also incorrect at something like 0.3 Deg instead of 0.25 Deg.

#4 The magnification factor of attack periscope seems more like 5X not 6X which is obtained by the field of view in the game world.


Depending on what ship you range on depends on what result you end up with as the results vary with target model as well as the viewing instrument utilsed.

The combinations of which all throw out measurements as few combination's show consistancy.

[akdavis]

No, it doesn't matter. You can't bring the mirror image down below the original, no matter the orientation of the periscope.

Also, you are wasting your time using the ruler on the navmap to compare stadimeter range to actual range. Simply turn the TDC on and compare the TDC plot on the attack map with the ship's actual position.

Point 2: Useing the attack map provides the exact same distance error as the navmap (I just tried it with the T3 - no change) Why should it be any different?


You misunderstand me. I'm saying it is easier to see if the measured distance and the ship's actual position coincide by observing the TDC solution marker on the map in comparison to the ship's actual location. This is more accurate than trying to determine an error on the order of 100s of yards with a tool that measures in nautical miles.

[Charos]

No, it doesn't matter. You can't bring the mirror image down below the original, no matter the orientation of the periscope.

Also, you are wasting your time using the ruler on the navmap to compare stadimeter range to actual range. Simply turn the TDC on and compare the TDC plot on the attack map with the ship's actual position.

Point 2: Useing the attack map provides the exact same distance error as the navmap (I just tried it with the T3 - no change) Why should it be any different?


You misunderstand me. I'm saying it is easier to see if the measured distance and the ship's actual position coincide by observing the TDC solution marker on the map in comparison to the ship's actual location. This is more accurate than trying to determine an error on the order of 100s of yards with a tool that measures in nautical miles.


I see where your coming from :up: - this is now I do it.


To start with im a Metric guy so im working in Meters.

On the Nav map you can drill down to 100's of Meters IE: 1.1KM, 1.2KM, 1.3KM etc

You just mark off JUST before it rolls over to the next unit length.

then its not to bad to at least cut that into 50% or 25% intervals IE: down to approx 25M chunks.

Which is close enough when in general 25M is only 1/40th- 1/60th or so of the total Range.


Things seem to be out in general by 12% to 18% so 2.5% to 1.5% error isnt that bad compared to what we are seeing elsewhere.

I only really started to take notice of all this when I shelved the TDC to go full slide rule in the S-Boats then things really went wrong when I tried to get a solution on anything.

EDIT: Now with Patch 1.2 we will have to test again to see if they didnt SNEAK any changes past us.

[Charos]

Well they certainly changed the high range periscope FOV BUT they went too far the other way this time. :-?

The Iowa is 270M long and has a mast height of 46M.
Real 8 Deg FOV scope to see the Iowa fill the entire screen would require a range of
135M/Tan 4 Deg = 1,930M .

___________________________________

The data from the Patch 1.01 Iowa test:

Note the Iowa in SH4 fills the 6X attack scope at approx 1,680M.(see attached screen shot)

Also note that the TDC via the stadiometer provides us a range of 1,406M.(see attached screen shot)

So that attack periscope markers on 6X and its associated FOV work off 0.29 Deg per graticule marker for its 9.3 Deg of view.
__________________________________

Now the same test in 1.02 provides:

Note the Iowa in SH4 fills the 6X attack scope at approx 2,850M

Also note that the TDC via the stadiometer provides us a range of 2,450M

So that attack periscope markers on 6X and its associated FOV work off 0.17 Deg per graticule marker for its 5.5 Deg of view.
__________________________________


1.01 shows a TDC to real range variance of 19% less and 1.02 shows 16% less.

Still ballparking around the same 17% mark as before now its just blown past in the other direction.

Problem hasnt changed its just been moved in the opposite direction.


EDIT : We now have approx a 9.7X Magnification attack scope. (Should be 6X)

Hopefully everything I typed makes coherent sense as Im still awake at midday after doing night shift.

[Charos]

If anyone is still following

I turned my attention to the Observation Periscope in patch 1.02 .

Low and behold the Observation Scope - seems PERFECT. :up:


The observation scope seems to be exactly 8 Deg FOV in high range and each
graticule marker is spot on at 0.25 Deg per marker.

This coupled with Krupp's excellent work on fixing the mast sizes and models now gives us an accurate fire control system. :()1:


On my S-Class the outermost tube diameter and head seems the same on both attack and obsrvation scopes :-? .

The Observation scope I guess may be easier to spot as fas as the AI is concerned though, in any case there is now a working realistic solution useing the observation scope for the time being.

[Jace11]

Great News

[jmr]

So the observation scope has a max zoom of 6x?

[Mav87th]

Snip from the Fleet Type Submarine Periscope Manual page 137.

"4. The stadimeter scale dials are graduated
for use with the periscope in high power.
When necessary to range on an object more
than 130 feet high, the stadimeter may be
used with the periscope in low power, and
the object set up on the height scale dial at
one-fourth its actual height. The range
reading is then correct."

The actual Stadimeter was installed with a "Course Angle Finder" as well as the Range Finder. What you did was, similar to the range finder, to adjust an image in length so it would fit with the actual image. Thus you could read out a Course Angle at the Stadimeter on the periscope....neet:arrgh!: WE WANT THAT UBI:rock:

[Charos]

So the observation scope has a max zoom of 6x?


From what I have read and understand both the Attack and Observation periscope are both 1.5X and 6X magnification in low and high range respectively.

You may ask then why both scopes?

Well the following comes to mind:

#1 Obs scope has less optics- In reality I dont think it is kitted with a stadimeter.

#2 Its a backup should the optics go in the main scope due to pressure via normal operations or depth charges.

#3 It has larger ojective lense and optics - just like a telescope it will gather more light and resolve smaller/fainter objects better IE: Stars and planes.

#4 It has a greater vertical range of movement.

#5 Downside is that for all its advantages the Obsevation scope is easier to spot when in use than the attack periscope.

[jmr]

Charos, by chance do you have Kim Ronof's Range Calculator Mark 3B whiz wheel handy? It was written in the SH 1 days and is capable of giving range in magnifications 1x, 2x, 4x, and 8x. Since the periscopes don't give us 2x, 4x, and 8x, we're limited to just 1x but even then I'm getting inconsistent results even with Krupp's ship dimension fix.

[Charos]

Charos, by chance do you have Kim Ronof's Range Calculator Mark 3B whiz wheel handy? It was written in the SH 1 days and is capable of giving range in magnifications 1x, 2x, 4x, and 8x. Since the periscopes don't give us 2x, 4x, and 8x, we're limited to just 1x but even then I'm getting inconsistent results even with Krupp's ship dimension fix.
Jmr

I intend on doing some extensive range tests with Krupp's figures .
I dont have the Whiz wheel you speek of - I do all my Trig with an Electronic slide-rule (calculator).

Could you tell me what ship you were attacking. There may be some out there that still need mast height adjustments - dont forget Krupp's fixes were only for Japanese
shipping the allied stuff is not corrected as yet.

[jmr]

Ahhhhh only fixes for Japanese ships so I guess that explains why I wasn't getting the correct range on American ships.

I just used the sub school training missions and I was able to get accurate periscope/whiz wheel ranges on the Japanese ships. However, on the navigation training exercise two US vessels pass off the port side, a Victory cargo and a T3 tanker, neither of which I could get accurate ranged with the periscope and whiz wheel. On the torpedo and artillery training missions I had no trouble getting accurate ranges with the same tools on the Mogami cruiser and a medium modern composite (don't know the Japanese name for it).

Wish I knew how to change the periscope mag so that it matches one of the whiz wheel's mag so I could get accurate readings at long range. 1x just doesn't cut it for those far away ships.