[WIP] Optical Targeting Assist, with Omnimeter

[CapnScurvy]

This modification is intended to correct the optical view of targets through the various scopes, enabling a correct reading of height/length dimensions when using the Telemeter markings (the vertical and horizontal hash marks in the scope). This will aid in correct range and angle on bow determinations when using these Telemeter scope marks. An in-game tool called an Omnimeter will be used to help facilitate these findings. Corrected mast heights and added ship length dimensions will be placed in the recognition manual to aid in this process.

The basis of information which I have used for the majority of my reference comes from a once confidential instructional pamphlet entitled Submarine Torpedo Fire Control Manual found HERE. It was published in 1950 at a time when the WWII fleet boats were still used as the main ship of the line submarine. Although the whole manual is interesting, Chapter 5 holds the most information on the specific periscope characteristics and the process of turning a targets height/length through Telemeter readings into range and AoB findings. I do admit some of the measurements do not add up when correctly figuring the formulas. For instance on page 5-3 it states We know that at a range of 1000 yards, 17 ½ yards, or 52.5 feet will subtend an angle of 1 degree. To be exact, 1 degree of angle at 1000 yards distance is 52.356 feet. Also, the correct range formula for the Hi Power magnification should read 76.4 not 76.2 (4 x's 19.1= 76.4). Small issues but whoever proof read these manuals should have done a bit better in what they cleared for print. The basic process is correctly laid out, so the information is accurate when using the corrected math formulas.


The overall process of making the modification is to have a working Omnimeter for in-game use, then correcting the optics (scope game views) so an accurate measurement can be obtained with the tool. Next, each ship will have a corrected lead mast head height and ship length dimension added to the Recognition Manual to aid in the two equations of finding target range and angle on bow. The goal is to give a Manual Targeting player the ability to use tools that were authentic and that will help in these determinations.


Omnimeter

A while back I posted images of a working circular Omnimeter (my first attempt) which did indeed work, but at a slight disruption to the in-game functions of the other dials it was linked to. I wasn't entirely happy with the fact that movement of one of the two dials of the Omnimeter also moved the torpedo spread dial. This wasn't too terrible since a good captain would check this dial when setting up the shot and reset it to what he intended. However, I wanted to try and make a tool that resembled the original and would cause a bit less trouble when manipulated. Since modded in-game devices need to be linked in some way to existing game devices I decided a less intrusive process is to have it linked to a little used game tool (the opacity slider of the message box). This allows for a more authentic Omnimeter with fewer disturbances to other game entities.

Below is an image of an Omnimeter that was constructed for the crew of the U.S.S. Cod. You will find a similar device (but not exactly alike) in the Torpedo Fire Control Manual. There was no manufacture that made them, rather the crew usually came up with their own version.






This is my in-game version.





It is moved about the screen using the message box's capture/drag device. Each slider is independent of the other for correct use. There are a couple of issues with its use. When at the Command Room screen the device seems to want to lock up the mouse when dragging the sliders after a bit of use. The correction is to jump to another station (the Navigation Map or one of the other Periscope stations) and the mouse will return to normal. This will only occur when at the Command Room screen!! Since the only time one will use the Omnimeter is when you're at one of the scope stations (Observation, Attack, and TBT) this really isn't much of a concern. Just keep in mind the correction if you drag up the tool and manipulate the sliders when you're in the Command Room. The second problem is to have the tool lost at the bottom of the screen if you drag it and the message box completely out of sight. The bottom of the screen will not allow you to retrieve the device if the message box is completely gone. The only fix for this is to restart the game from a windows execution. This really isn't too big a problem either since the stock game allows this behavior. Try it, the stock message box will not return if you drag it out of sight at the bottom of the screen and release the mouse capture. Just another game quirk we all live with.


Game Optics

So, the in-game Omnimeter is completed and works well. The next issue is to correct the optics. You may ask, what's wrong with the optics? Well for starters the "magnification" and field of view of the stock game are incorrect. The Submarine Torpedo Fire Control Manual states on page 5-1, the magnification for the Type II (Observation scope) should be:

High Power 6.0x
Low Power 1.5x

The Type IV (Attack scope) has the same power.


The stock game optics gives us a magnification of:

High Power 4.0x
Low Power 1.0x

Its little wonder the game seemed to have the ships farther away than they should be. I made a modification (Ship Centered, Accuracy Fix or SCAF for short) to help allow a player to use a more visible part of a ship (funnels, flag tops) to measure with the stadimeter because the mast tops were not visible (least what I thought) at normal range finding distances. Having the magnification corrected will allow for a clearer view of the mast top. It's my plan to have all corrected heights return to the main or tallest mast head with this mod. If we can see them better, then lets use them!

To correct the field of view we need to look at the stock view and what it shows. The following image is taken with several stationary ships set in front of a stationary sub. The Akita (at the relative bearing of zero) is 1000 yards away. The range isn't really important, (neither are the other ship distances) but what is important is the field of view between the scopes left and right outer edges. The first image shows the Akita at a relative bearing of 0, using the Low Power magnification. I've drawn a red line on the right hand side to point out the area on the ship we will use to mark the outer right hand field of view position. This point on the stationary adjacent ship has to be used, open water doesn't leave you with much of a reference point. It's just about centered on the super structure of the freighter





The next image has us moving the scope to the right, placing the center cross hair on the point we referenced.





Reading the relative bearing of the stock game, the outer FoV is approximately 20 degrees from center. When you add the left side, the total FoV is approximately 40 degrees. On page 5-2 of the Submarine Torpedo Fire Control Manual you will see the true field of view should be 32 degrees. Were off by a total of 8 degrees at Low Power. So, how does this translate to feet? One degree at 1000 yards is equal to 52.356 feet. Times 8=418.8 feet off!! You move out to a range of 2000 yards and yes, the error doubles.

While we're in the manual, a High Power magnification true field of view should be 8 degrees. So, let's see what we have.

The following image has the same grouping of ships, this time with the center reference point to the left bow of the Akita at the 355 relative bearing. Notice the right hand edge of the view is marked at a point on the Akita behind the super structure of the ship.





The next image below shows this same reference point centered with the cross hair and now the relative bearing reads approximately 0 degree, a 5 degree difference. This means at High Power the stock game has a total FoV of 10 degrees from left edge to right edge, where it should read 8.





Without doing a lot of head scratching we can see the stock game needs to be corrected if we want to compare apples to apples and determine what really are we looking at through the games inaccurate eye. Count the Telemeter divisions of the scope, either horizontally or vertically there are 32 divisions. The game dev's planned to have a correct 32 degree FoV but they failed to input the correct "camera" parameters to make the correct view size. It's understandable why there was so little amount of reference in the game manual about the Telemeter marks. They really don't work when trying to use it as described. Not only is the magnification off but the FoV is too great for what the scope should be. I realize the stock relative bearing compass is very forgiving considering its small size, it only as a mark every 5 degrees. Lot's of room to have a couple of degree error, and no one to be the wiser. So, the mod will be using a much larger and more accurate compass in the future. It's made by Captain Cox (an oldie but a goodie). In my next post I'll be showing you what a corrected view of the world will look like (surprisingly, it's not too far off from what my MaxOptics IV modifications would give you, but I'm not here to talk about that. We've got bigger fish to fry).

More later.

[Hydra]

Fantastic work Looking forward to your followup.

[Dignan]

I'll definitely be watching this one. Nice job. Any estimation of when you will have a working version?

[CapnScurvy]

Originally Posted by Dignan

I'll definitely be watching this one. Nice job. Any estimation of when you will have a working version?

It will take a while, maybe several weeks. That's why I started a "Work in Progress" thread.

Although the Omnimeter is done and working well. I want to test the optical settings for the various game resolutions. The "3000 Yard Bearing Tool" proved the need to have a separate version for whatever game resolutions a player may use. I hope I won't have to do that with this modification, but I may.

Also, there's over 100 ship classes to change in the stock game alone. That's for mast height correction and determining ship length. I've been around this game enough to know you can't take for granted what the game states the heights and lengths should be. Each ship will have to be checked through an edited mission placing it at a specific distance from the sub, at exactly a 90 degree AoB. It's time consuming just to make the test missions, then the real work comes in to measure the heights and lengths.

I'll admit, the testing part is much easier compared to the weeks I've spent on the Omnimeter. So I hope to have it soon, in the mean time I'll use this thread to keep you posted on my progress.

Thanks for asking.

[Sailor Steve]

A thing of beauty!

[clayp]

CAP you are the best!!!!!!

[Hitman]

In my experience with doing optics for SH3, I have found that the best method to determine true degrees dimensions in the game is actually already built-in, and that is the hardcoded german reticle (Which is in degrees). I recommend you to fire up the game in the highest X=*/768 resolution (4:3) your monitor can handle, and then choose a german uboat and look through the periscope. Take a screenshot and count the pixels between each degree mark, and you will have a very good indicator of that dimension in the game. Then you can adjust the periscope mask and reticle markings in size to ensure it has the correct FOV in terms of linear horizontal pixels. It will work directly for the US counterpart, as the cameras.dat file is shared

Also, don't take that 1x/4x vs. 1.5X/6x thing to seriously. In game terms, augmentation is only equivalent to field of view, and hence you can have a 1x periscope that actually produces the correct linear FOV for a 1.5x. This is because you have in the game two values to play with: Augmentation and Camera aperture (FOV), whereas in real life you don't (They operate in different areas). But the end result in the game is the same, because true augmentation is not possible to be scaled accurately, as you have different sizes of monitors for all players, and you look at them from different distances. The important thing is that between both zooms the difference is 4x, because specially in US optics the small/large marks come in series of 4, as it was primarily designed with 6x in mind, where each big mark is 1 degree and each small one 1/4 degree.

Hope that can give you some ideas, and excuse if you had thought of this all already

[tomoose]

Excellent work. I'm cringing thinking about my pitiful attempt to do the slider in html format, LOL.

Looking forward to this one. Thanks.

[CapnScurvy]

Thanks Hitman for your insight!! I'll take a look at the German scope for what you're talking about. Yes, I do realize the ratio between the two magnifications is the important factor (4) not the actual numbers used. Again, thanks for the pointers.

I've got some bad news and some good news.

The bad news first. While running through the various resolutions I found one discrepancy with the stock game optics. The 1280x1024 resolution created a 2 degree field of view difference (low power) compared to all others!! The total FoV for the 1280x1024 resolution was 38 degrees wide! The total FoV for all other resolutions is 36 degrees wide (the figure for having a correct FoV is 32 degrees wide). Admittedly, I grew tired of guessing what relative degree I was on using the stock scope compass so I put in the Big Bearing compass of Captain Cox's and proceeded to check with clearer results. This 2 degree wider FoV only showed on this one resolution, no others. And wouldn't you know it, this is the resolution I usually use for my modding. I plan to check this difference on my other monitor. If past testing holds true, this quirk will require a specific optical setting just for it. The others seem to be exactly the same, even the 1280x768, 1280x960, and the 1280x800. Why, I don't know. I just know it's there.

So you may say, "What's the big deal about a couple of degrees wider with the field of view"?

A target in a wider FoV will appear smaller due to the increased size of its view width. This occurs even though the target is at a specific range away from your sight point. It will appear smaller to you due to the way the game defines its parameters. The trouble comes when you have the game set to "Manual Targeting". As a player you will either use the Stadimeter, scope Telemeter divisions, or Sonar to determine range to target (you can use the map to plot a targets position, measuring with the ruler if the "map contacts update" is enabled. This gets into a bit of "cheating" if you want to really play using manual targeting). Further, both Sonar and the map measurement method does not rely on the FoV for getting their range. Doesn't make a tinkers damn what the view is, the game plots the objects through its game engine and knows precisely where everything is. That's why using the auto targeting feature is accurate when displaying range. It does it through the games built in "GPS" (sort of speak), it knows exactly where it puts things. The trouble is with the Stadimeter, or using the scopes Telemeter divisions needing to have a correct view of a target to calculate range. Without a correct view neither will accurately give correct range. The Stadimeter relies on the targets second water line image to be placed on the screen to determine height. If the height is lower than it should be (due to the smaller size of the target) the range finding will be inaccurate; leaning to a greater distance in range than it should be. The scopes Telemeter divisions are marked for a 32 degree view (check it out, count the divisions from left to right, or top to bottom). A 36 or 38 degree camera view will not give accurate measurements if the the division marks are made for 32 degree FoV!!

The good news is I've corrected the other resolutions to give a 32 degree FoV. It would seem I'll have to make just two different mods to cover the possible resolutions a player may use. I do plan on testing more to make sure but this is where I'm heading.

I told you I would provide an example of what a corrected periscope view would look like. The following images are of a game resolution of 1280x1024 using the "Optical Targeting Correction" modification at low power. We start out at the compass bearing of zero, the plan is to check the bearing at the left edge of the screen.




We move to the point of reference onto the adjacent ship and find....





Ok, we're on the money. 16 degrees = half the field of view, x's 2 = 32 degree FoV.

How about a High Power magnification? We start with the scope centered on the 355 degree relative bearing. This time we will check the right side of the scope, centering on the freighters superstructure, just behind a porthole.




We're looking for 4 degrees of change from the center to the edge. This will give us the correct 8 degree total FoV. Here's what we have.....





So the optical corrections have been implemented for the 1280x1024 resolution. I have the others as well. I'm going to test more but for now I'm thinking there are just two resolution settings that need to be addressed. I just need to check more.

If all goes as planned now, I'll be testing mast heights and ship lengths soon. I'll keep you up to date.

Again, thanks for your input. Every little bit helps!!

[Hitman]

I grew tired of guessing what relative degree I was on using the stock scope compass so I put in the Big Bearing compass of Captain Cox's and proceeded to check.

If I might make another input ... here it goes :

It is customary to see in many optical mods scope lenses that cover the whole screen. This is usually justified because the monitor resolution is poor compared with the MKI eyeball and because when you look through a scope/binocular you just see a view circle that covers more or less your whole field of view -even if it is actually way more reduced than when looking with naked eye.

This solution is of course realistic in that aspect.

However, I took a very different approach in my own optics mod, and I think this is a good oportunity to explain why, so you can consider it for the best possible solution.

1.5x zoom was chosen for low power specifically for one reason: At that magnification, objects appear to observer at their natural 1x size (i.e. as if looking at them with naked eye), and that was a much desired effect since naked eye range estimations were common practice in WW1 torpedo firing and carried on through prewar to WW2. The german navy also used that magnification, and according to Norman Friedman (US submarines through 1945) it was because of the same reason. Also, situational awareness is much improved because the observer has a real sense of how far the object is, and of the chances of maneuvering the submarine in the space available.

Hence, enlarging the scope size in the screen as shown in those pictures produces one realistic effect, and also one irrealistic one: You lose any sense of proportions between what you see through the scope and what you see from the bridge. Furthermore, it subtedly makes you want to submerge, because you actually have a better sight of the enemy from periscope view ... which is exactly the opposite to what historically happened. US periscopes were well known for poor light transmission and limited capabilities to use the stadimeter at ranges above 5000 yards with any kind of accurancy (Again same book from Norman Friedman).

So, to sum up, I think that in a overall picture of the game, a periscope that appears smaller in the centre of the screen would be more effective to fully simulate the conditions under it was operated in WW2. But of course, this is your mod, adn anyway you solve the matter I will agree (And download and enjoy it ).

[CapnScurvy]

As a follow up on the resolution differences I pointed out earlier; comparing the 1280x1024 setting to others. There is indeed a difference in the stock game when looking at the scopes field of view of the 1280x1024 resolution compared to all others (1024x768 up to 1920x1080). The 1280x1024 gives you a wider FoV by two degrees over the other resolutions. I mentioned I used a larger compass bearing to achieve greater accuracy and found the FoV of the stock 1280x1024 to be 38 degrees (using the stock compass, I had said the FoV was approximately 40 degrees). Further, I found the remaining stock resolutions to be a consistent 36 degree FoV, so there's a 2 degree difference between the two. These numbers are not as bad as I first originally thought but they still are not at a 32 degree field of view where they should be. I confirmed this finding by running my test missions on another computer using a native 1920x1080 monitor.

In doing a bit of checking, I found the stock game was first released with a 1.5 and 6.0 magnification in the periscopes with a 57.265 Angular Angle. It would appear the developers had some notion of having the game optics as they should be. However, after the first patch (released here in the U.S. only weeks after the game) and up through the last 1.5 (U-boat Missions addon/Gold Edition), the magnification has been 1.0 and 4.0 with an Angular Angle of 62.4814. What does this mean to the average gamer? Short answer, you aren't going to get a proper view of the world. Also, you don't stand a chance of getting mast heights or ship lengths correct if you expect to use the Telemeter scope marks to help in counting. Actually, I believe the plan all along was to bastardize the U.S. game for the long held intention of putting the "Nazi's in the Pacific". AND charge us extra to do it. Thanks UBI, you're a real peach!!

Sorry, I couldn't resist telling you what I think. I'm better now.

So, here's a couple of comparison images between the stock and the Optical Targeting Correction mod. First in low power magnification. The Akita is at 1000 yards range.



Now in High Power.




Is this the same view of the real world through the periscope. No, not at all (least I wouldn't know, I've never looked through a periscope). But it does accurately give us a relative perspective of the game world with a consistency that should enhance game play.

Look at the monitor you're reading this on. It's probably the same one you use to play the game. If not, it's very close to being the same. The point is, it's 2 dimensional!! Like it or not, there is no depth perception to it. None! Close one eye and the flat screen looks the same. Now, go to a window and take a look about. Then close one eye. You see the difference in your depth perception? This difference (or lack of) is what we see playing the game. We have no depth perception in the game because of this simple fact, the image is reproduced on a flat, 2 dimensional monitor.

Did the real life captains enjoy using the naked eye to judge target assessments? Heck yes, using the scopes single eyepiece gave them the same "lack of depth perception" one gets when covering one's eye. Of course, it's much better to have complete binocular vision than monocular. But, in the game we get monocular vision in the periscope and on the bridge. Like it or not. In real life the Telemeter marks were put into the scopes lens to give the captain a basis for accurately determining targets heights and lengths without the aid of depth perception. Its my goal to have these marks corrected to the image the game displays to aid in the same way.

What the captains did enjoy about the scope views were the magnifications. They could "bring in" the targets to get a better view than looking at it through the naked eye. There aren't many photos of periscope images but here's a few.










You can almost see the boys waving on the deck can't you!?!

If this is max magnification, how close do you suppose they are? The pictures don't tell you but I'm of the opinion they were darn close!! Which makes me wonder why in the world do we think the game image of a target should be so small we have to squint to see the mast head? I have no intention of changing the optics just to see the target ships with better clarity. My plan is to provide the same optical parameters as the real life fleet boat, and do it so it is accurate within the game world that's provided. In doing this I believe the target will appear as close to what it should, with the added benefit of appearing closer than what the stock game provides. In other words, having a 50 foot dimension really read 50 feet in the game. Whether it truly is 50 feet in real world dimensions isn't important as long as the game thinks so. This is something I believe has been a long standing problem of modders using real world dimensions in the game. Using the ONI ship recognition reference to change ship dimensions and thinking this will solve any calculations the game makes. In my opinion the game is flawed in its view of the world. Using real world dimensions will be inaccurate. I'm not holding out hope for this, but just maybe, correcting the optical view will allow a closer relationship of what the game portrays and what the real world view truly is.

[Hitman]

Of course you have a point there, and it actually isn't incompatible at all with what I wanted to highlight: Do the optics as you see better, and then update the other angluar angle views of the open bridge, etc, to match. What I want to stress is the importance of keeping the proportions, and not being used to see the enemy ship at a certain distance when still surfaced, and then suddenly it appears much closer when submerged. This can easily screw your approach because you lose a sense of how far or close the enemy is, and you miscalculate the space needed to maneuver.

1.5x should allow keeping the tactical view, 6x all details and getting accurate masts measuring, etc. as you rightfully point out.

[CapnScurvy]

Hitman, your point is well taken.

I had not thought of looking at the bridge view and have it relate to the new scope views until you brought it up. The only views that give you the tools for making assessments for measurements were the scope views, so I have concentrated my efforts on them. I'll be able to make the adjustment to the bridge view, making it a bit smaller than the scopes. It's just that I'll have to "eye ball" the relationship (yep, pun intended).

Here's the difference between the two views as they stand now. As Hitman pointed out the difference between the two is too great to be considered accurately portrayed.





The above (and below) pictured test mission is having the Akita at a 1000 yard distance (1001 to be exact).





If you wish to check target ships with the same type of mission test I'm using you can download it HERE. It's simply called "Test 21" and will be found in the "Quick Mission" menu of the main game screen. It has several target ships positioned around a stationary sub. The down loadable "Optical Test Mission Pack" (JSGME compatible) has two mods. The above mentioned test mission and Captain Cox's "Big Bearing" compass for the attack periscope. The compass should help as you check the field of view degree findings. Extract each mod separately to the JSGME "MODS" folder to activate.

Using both mods, you can check the stock field of view yourself, and come up with your own conclusions. Try the 1280x1024 resolution for a different size than the rest!! Depending on the monitor you use, some of the various resolutions will of course appear distorted. But, as you compare the relative compass bearing to the outer views edge you will see a consistent degree figure used (except for the1280x1024), no matter if the view is distorted or not.

[Hitman]

I will give it a try, but if you use 1.5x in the scope as basic zoom, then it's all a matter of setting for the bridge view the same aperture angle as for the scope

[CapnScurvy]

Hitman I'm wondering if you know (or anyone else), what is the Field of View for the TBT in the real life fleet boats? For that matter, the UZO for the U-boat. I've tried the Internet and have come up empty regarding the TBT. Little information all together.

The stock game has the magnification at 7x. When reading the FoV within the TBT it covers 3 1/4 degrees to the edge (from center) meaning it's 6 1/2 degrees total FoV. Just wondering what was it in real life?