Here is the full readme text file included:
Features of This San Giorgio Tipo B Gun Sight Mod
-------------------------------------------------
1) More correctly sized.
2) More detailed and accurate.
3) 6DOF friendly, with polygons extending well into the rear sides.
4) Utilizes the height-settable head feature of the real sight (NOT adjustable on the fly in-game!). Variants with heads set at heights of approx. 0, 1, 2 and 3 inches are provided.
5) All through-the-sight lines of sight are higher than were set for stock, particularly for the MC-200. This provides a better overall view, especially during maneuver shooting.
6) The iron sight strap is separated from the reflector.
7) The iron sight strap flips down for use when the sight is damaged (lamp/electrics failure).
A printed-on-glass reticle is included for use with the iron sight, as was incorporated on the real thing.
9) The optical reticle is correctly dimensioned.
10) The SHF-F1 POV is set near enough to the sight for full optical reticle visibility.
11) A 'floating' lens reflection is incorporated, which can be set to one of three brightness levels, or disabled.
12) I rendered invisible the useless and ugly outside-the-cockpit iron ring sight texture for the MC-200 and MC-202.
13) Incorporates my canopy glass treatment, which can be adjusted in intensity by editing a single value for opacity in the .mat file.
Compatibility and Conflicts
---------------------------
As always, when such fundamental files as
hier.him
body.msh
are altered there is the possibility of conflicts with other mods that have altered these files as well. The first copy of any file found by the game is the one read; in an alphebetized MOD folder listing, the topmost is the first. If you have any other mods for these stock planes, check for the presence of these two files. If present, you should copy over the changes I've made in my versions to your existing files, or vice versa--whichever is easiest. Then make sure the version to be used appears first, or rename the unwanted one with a preceding minus sign to <-hier.him>
In hier.him, my changes typically involve
[Z_Z_RETICLE] (optical reticle)
[Z_Z_MASK] (optical reticle mask)
[pricel] (undamaged sight)
[pricel_d1] (damaged sight)
There are a few other alterations to sight-related items, with "pricel" in their names. This should be obvious from the liberal use of "//" and other appended comments.
If your other mod's hier.him has not been 'expanded' into a more readable format, you can add hard returns at the appropriate places; use mine as a guide--after making a backup of your hier.him to be altered.
Body.msh contains the view hooks for setting the POVs. This is critical for non-6DOF equipped players. No other changes were made here, except...
...For the Cr.42 I had to alter body.msh so as to remove the old sight's lamp house (another example of sight parts being scattered about among a number of files.) I left the old data in place, but REM'ed them out with "//" characters. Of course the number of vertices and faces changed, as reflected in the [FaceGroups] section.
For the G-50 I altered the cockpit glass in front.msh, but only to reverse the polygon surface normals--more on this treatment below. Other than this small change, the file can be thought of as essentially unchanged, and still stock for the purposes of compatibility. If this file is deleted, the cockpit glass reverts to the usual lighting behavior for the surface normals.
Optically Generated Reticle
---------------------------
This Italian reticle is among the largest in angle which appeared in WWII-era optical gun sights. It is dimensioned as follows:
- Adjacent dot center-to-center separation: 28 mils
- Outside dot center-to-center separation: 56 mils
- Inner circle radius/diameter: 60/120 mils
- Outer circle radius/diameter: 118/236 mils.
As designed originally, the dimensioning was determined as follows. The dot separation of 28 mils corresponds to a 30's era 7m fighter length as seen from a distance of 250m. The 60 mil radius inner circle corresponds to the distance traveled for a target transverse velocity of 50m/s (roughly 100 kt) at a distance of 250m during the time of flight of projectiles traveling at 850m/s. The 118 mil radius outer circle corresponds to the same conditions, but with a target transverse velocity of 100m/s (roughly 200 kt). Note that the seemingly discrepant 118 mil radius, instead of 120 mils as might naively be expected, derives from the departure from linearity imposed by projection effects of perspective.
I've dimensioned the reticle pretty closely to that of the real thing, and expect it's accurate to within a couple or few per cent. (My old CRT monitor is a source of uncertainty.)
For the ranging of 10m wingspan fighters and 18m wingspan twin-engine bombers:
A 10m wingspan will extend to adjacent dot centers (28 mils) when 357m distant.
A 10m wingspan will extend to outside dot centers (56 mils) when 179m distant.
An 18m wingspan will extend to adjacent dot centers (28 mils) when 643m distant.
An 18m wingspan will extend to outside dot centers (56 mils) when 321m distant.
The formula:
Distance = wingspan / (# mils / 1,000)
For example, the distance at which an 18m wingspan just fills the inner 120 mil diameter circle:
Distance = 18m / (120 / 1,000)
Distance = 18m / 0.12
Distance = 150m
Printed-on Reticle
------------------
The Tipo B sight incorporated a unique feature; a reticle pattern printed on the lower reflector plate to be used in the event of lamp or electrics failure. A wrap-around strap, closely surrounding the reflector when stowed, was flipped downward until snapping into a slightly less-than-horizontal orientation. This placed the bead on the iron sight post at the correct height for sighting through the printed reticle.
In my damaged version of the sight (which is otherwise identical to the undamaged one), this strap instantaneously drops down into its position for operational use. (I don't have the skill nor tools to animate this item, unfortunately.) If you use head tracking, it's simple enough to lower your head the small amount required in order to use this sight. And if your tracker supports for-aft head movement (along the Z axis), you can place your virtual eye nearer to the sight's bead as well.
The small reticle pattern printed on the glass would subtend the proper angle of 236 mils for the large circle when viewed from the correct distance. From crude measurements of photos and diagrams, I estimate this printed pattern to have a diameter of 20mm for the outer circle. And so it would subtend 236 mils when viewed from a distance of 20 / 0.236 = 84.7mm. This is barely greater than the separation between the sight bead and the reticle, which I find strange. One would hardly like to place his eye in near contact with a sight bead, and in a moving airplane no less!
That notwithstanding, the game induces clipping of objects when nearer than some minimum viewing distance. In my 4.08 game version, a good part of the sight is clipped away when my virtual eye is at the correct place. And so I've defined POVs which do not clip the sight, and which make the reticle appear to be at 1/2 scale. That is, the outer circle is sized in apparent angle so as to subtend about the same diameter as the optically generated inner circle. Even then, at larger FOVs of about 85 degrees and larger, when one's head is turned the nearby sight post/bead becomes clipped when near the screen edge.
I've designed this part of the sight to be in alignment when the top of the bead is placed so as to appear to 'touch' the underside of the central dot in the reticle pattern. POV changes as small as 0.1mm have a visible effect on alignment.
The worked-out POVs are included in the plane's body.msh file, where the various view hooks are defined. I include this in case your game version has a 'slot' available for defining a selectable viewpoint. (In my 4.08 version, I have only two; the 'normal' sitting position and the 'SHF-F1' gun sight view.) This is covered in the next section.
If you wish to use this 'strap down' version of the sight, either full time or just for testing out, do this. In hier.him, locate the section [pricel] ( or [pricel_D0] for the G-50); this defines the mesh used for the undamaged sight. Change the mesh name from "Pricel" to "Pricel_D1", the latter being the 'damaged' sight model. Here's the relevant section from hier.him for the Cr.42 (a bit compacted here to save space):
[pricel]
Mesh pricel
Parent _ROOT_
Attaching 0 1 0 -1 0 0 0 0 1 -1.17514 0.00016 0.73736 //F1 U2.8
[pricel_d1]
Mesh pricel_d1
Parent _ROOT_
Hidden
Attaching 0 1 0 -1 0 0 0 0 1 -1.17514 0.00016 0.73736
You can change the section [pricel] to read:
[pricel]
Mesh pricel_D1 //pricel
Parent _ROOT_
Attaching 0 1 0 -1 0 0 0 0 1 -1.17514 0.00016 0.73736 //F1 U2.8
[pricel_d1]
Mesh pricel_d1
Parent _ROOT_
Hidden
Attaching 0 1 0 -1 0 0 0 0 1 -1.17514 0.00016 0.73736
I've 'REM'ed out' the original "pricel", preceding it with the mesh name "pricel_D1". It's OK to use the same model for both the undamaged and damaged sight; the game doesn't care. You could even show both simultaneously, by removing the "Hidden" status for the damaged one (which normally becomes un-hidden, and replaces the undamaged model, when the game flags the change of state.)
As you can see, hier.him is a powerful 'control center' that builds the cockpit from its various components. Just as for the view hooks (covered next), one can fine tune object placement in XYZ by altering one or more of the last three numbers of the attachment coordinates; I've necessarily had to do this for all my sights.
View Hooks and POVs (points of view)
------------------------------------
By way of example, here are the view hooks sections from body.msh for the G-50:
[Hooks]
CAMERA <BASE> <-- 'normal' view
CAMERAAIM <BASE> <-- SHF-F1 gun sight view
[HookLoc]
//0 0 1 1 0 0 0 1 0 0 0.9552 -2.31474 L U F ?
//0 0 1 1 0 0 0 1 0 0 0.905 -2.20006
0 0 1 1 0 0 0 1 0 0 0.955 -2.24 //U5 with new sight
0 0 1 1 0 0 0 1 0 0 0.948 -2.0 //U4
//0 0 1 1 0 0 0 1 0 -0.00058 0.914 -1.875 //for iron sight when 'damaged'
The lines preceded with '//' are considered as remarks, and not read by the game. CAMERA defines the 'normal' sitting position, and CAMERAAIM defines the SHF-F1 gun sight view. Later game versions may include others. I've 'REM'ed out' the original two lines of data--to retain the original values-- and added my own.
For each line of positional data, the first 9 numbers define the rotation matrix and are of no interest to us for these purposes. The last 3 numbers define the XYZ position of the viewpoint, in meters, with respect to some datum as defined by the modeler.
As determined experimentally, I note what the XYZ values control, and the sense in which they increase in value. This can and does vary among planes built by different authors; in one plane increasing X might move the POV left, but in another move the POV rearward.
Here, the last three numbers control L-R, up-down and fore-aft viewpoint placement, respectively. Moreover, these values move the POVleft, up and foreward, respectively, when increased in value. Hence my appending of "L U F". The question mark is there because I've not verified the sense of motion on the L-R axis, not having had to make a change here.
For your convenience, I've included the POV setting for the iron sight when 'damaged', and identified as such with a remark. As noted, this distance renders the printed reticle as subtending an apparent angle at 1/2 scale.
If you wish to use this 'iron sight POV', it must be assigned to an available slot in the [Hooks] list, and in the correct order with other POVs. It's OK to have 'REM'ed out' lines between active ones; all that matters is that the order of active lines is observed. If there are excess 'non-REM'ed out' lines, only the first active ones equalling the number of defined hooks are read; subsequent ones are ignored.
Reflector Plate Transparency
----------------------------
A couple of diagrams which I've seen online, and which appear to originate with the manufacturer, suggest that the reflector plate was laminated from two sheets of glass. Moreover, a translated passage of text from the manufacturer's manual for the sight's predecessor, the Tipo A, states that a lamination and partial silvering was employed for it. This partial silvering was intended to increase reflectance of the optically generated reticle. I assume the same was done for the Tipo B's reflector.
Such partial silvering to increase reflectance necessarily decreases transmission, this being the principle of the plate beam splitter, and the two-way mirror. Here the ratio of reflectance to transmittance would have been kept fairly large, especially as the manufacturer claimed no notable dimming of the scene. But some darkening in transmission would be unavoidable, as the conservation of energy demands.
And so, as compared to a simple glass plate as employed in many gun sights, I've given the reflector texture a bit more opacity to suggest the slightly darker view imposed by the deposited metallic film.
Reflector Plate Reflection
--------------------------
A consequence of higher reflectance for the reticle is a higher reflectance of any other illuminated bits--such as the collimating lens when sunlit. I gather that in real planes spurious reflections, from the sight's reflector and windscreen, were a perennial problem. For this sight, because of the partial silvering I've decided to add a reflection of the sight's collimating lens.
Some of our game's planes have incorporated lens reflections as part of the reflector's texture. This technique necessarily places the reflection in the plane of the reflector, which is not realistic. A real reflection is located at a distance beyond the reflector equalling the distance at which the lens lies below the reflector. Not a big deal for sights whose reflectors closely crowd the collimator. But when the reflector is some distance above the lens, the latter's reflection must of course lie a considerable distance beyond the reflector. This is certainly the case for the MC-200, where I've raised the reflector quite high.
A notion I've pondered for quite some time is the placement of the reflection at the correct distance, which requires the use of a mask if it's not to be seen outside the reflector when the POV gets far enough off axis. One argument against this derives from the fact that the same reticle mask must be used for both the reticle and the lens reflection (unless a way exists to use *two* masks at different distances?).
With just the one mask available (without so far having tried experiments to see if this is ultimately the case), it requires to add the extra lens reflection element to the ZZ_Reticle.msh which, just like other meshes is not limited to one element only. As is usually done, the mask is placed at or near the plane of the reflector so as to ensure that the reticle cannot be seen outside the reflector when the POV is moved far off axis, as is so easily done with 6DOF head tracking.
When a circular mask is used, so as to correctly mirror the collimator's circular aperture, corners and other regions of the reflector are unused for the transmission of the reticle. The same goes for the reflection of the lens when placed beyond the mask. In reality a reflection could be seen anywhere on the entire reflector.
But with the Tipo B the reflector's shape partially obviates the 'corner problem' by having them truncated, thus reducing the area through which the incongruous masking of the reflection would occur. And so I've elected to proceed with correct reflection placement in distance, even though with larger POV offsets the mask clipping which does occur is unavoidably unrealistic.
The texture lighting (via its surface normal) is defined so as to be brightest when the sun is shining straight down onto the collimator. For example, this condition occurs in level flight when the sun is directly overhead, at the zenith. With the sun shining straight into the lens (and focusing an image of the sun upon the reticle itself), reflections are expected to be most intense. The texture brightness decreases as the angle of incident sunlight decreases. The degree of ambience I've set in the .mat file does let the overall scene lighting contribute somewhat, but the sun is the very dominant source.
This graphical touch is quite effective with head tracking in use, as it adds another dynamic element to the gun sight picture. With no head tracking, it might be considered a boringly static annoyance. And so I let the player choose. I supply two versions of ZZ_Reticle.msh, in dedicated sub-folders for each plane. One version has just the reticle, the other has both the reticle and lens reflection. Simply copy/paste (don't drag/drop, if this *removes* the file from its folder) the desired file to the plane's parent folder; it's OK to overwrite, as the two sub-folders are the 'archive', should you change your mind later. And carefully note: do not put one plane's ZZ_Reticle.msh into another plane's folder! They are not interchangeable.
I supply three versions of the lens reflection texture, in dedicated folders, designed to give a bright, less bright and dim reflection. Copy/paste the one you wish to use to the plane folders. My 'default' is the middle brightness variant. You could copy a bright texture to one plane and the dim texture to another plane. In this way you can quickly switch planes in a QMB flight so as to compare, then settle on the version you prefer.
If for some reason you wish to make a version differing in brightness from these three, increase or decrease the brightnes of the grey circle in the .tga's alpha channel. You should be careful to leave the black surrounds as black, else in the game you might faintly see the square frame surrounding the circular reflection. You can make this alteration using any photo editing software that can handle--and save--alpha channels.
A Note on Reticle Mask Realism
------------------------------
This is of importance when considering ultimate realism. The reflection of the collimator aperture IS ITSELF THE RETICLE MASK! The lens is the window through which is seen the collimated image of the reticle, the latter lying or near optical infinity. The farther from your eye lies the lens, the angularly smaller the window, and hence the more restricted the extent of reticle seen. The interposed reflector plate does not alter this in any way.
If one were to go for maximum realism, the reticle mask should be sized and located exactly with the lens reflection. For instance, a 60mm aperture collimating lens located 80mm below the reflector (as measured on the optical axis) will have its reflected virtual image located 80mm beyond the reflector. And so the reticle mask should be made the same 60mm diameter, placed also at 80mm beyond the reflector. The reticle mask and lens reflection would be *exactly* coincident in space.
To really appreciate just how restricted the reticle can be in reality, hop in the MC-202 for a ride. Observe the lens reflection, and how very much smaller than the reticle it is when your POV is well back. If I were to impose full realism here, the reticle would only be seen inside the circle of the lens reflection; restricted indeed!
If you have 6DOF head tracking, observe what happens as you move your POV fore-aft. The reflection becomes angularly larger as you get closer to the reflector, which would permit more of the reticle to be seen within the reflected lens image if the mask were to be realistically made coincident with that reflection.
Reflector Plate Edging
----------------------
Among aircraft optical gun sights, some reflector plates were left bare, in various degrees of surface finish. This could range from polished or semi-polished to finely ground. Others were coated with a paint-like application, most often black, in order to reduce one or both of 'glare' and internal reflections.
The Tipo B reflector edge seems to have been coated black, perhaps in part because of the lamination, as a means of protecting the interface where the sheets joined. A collection of color photos online, in which it appears a single-sheet, non-laminated, partly damaged glass is installed, also shows some flecks of black 'paint' remaining. This particular plate might well be a substitute, fabricated some time ago to fill in where the original was too damaged or missing. If so, the seeming blackening for this is presumed to emulate the original, supporting the likelihood of the original plates having been blackened. I've decided to offer a blackened edge only.
Sight Head Height
-----------------
Another rather unique feature of the Tipo B sight was its height-adjustable head. This accommodated for differing panel mounting heights with respect to the front windscreen, and, I suppose, pilot height variation or personal preference. I estimate the range of travel afforded by the vertical posts to be about 80mm.
For the Cr. 42 and the G-50 I've set the sight head to its minimum, finding no real need to go higher. Even then the line of sight is a bit higher than for the stock sights, mainly due to my higher positioning of the sight body, added to by the somewhat higher placement of the sight head on the body.
The MC-200 has its sight head raised by 75mm, affording a significantly improved view over the cockpit coaming when sighting, notably aiding in a turn fight.
The MC-202's sight head is raised a moderate 23mm, this being quite sufficient for the more confined head room.
I've laid out the sight's mesh file so that all sight head elements which can be adjusted in height are grouped together. The dividing partition is indicated with a remark, stating that all vertices below it can be uniformly adjusted in the Z (up-down) direction, whose values reside in column 3. I also indicate the useful range of values for the line immediately following, so that one knows at a glance by how much movement can be effected before exceeding the permitted range.
Here's the relevant lines in the vertices section from the pricel.msh for the MC-200. The partition is roughly halfway down the section, [Vertices_Frame0]:
*** code ***
[Vertices_Frame0]
0 0 0.0059 0 0 1 lens center
0 -0.0315 0 0 0 1 outer ring 63mm
.
.
.
// ALL BELOW CAN BE MOVED UP/DOWN IN Z (column 3).
// Range for rod top: -0.005 to 0.075 (currently 0.07)
-0.0261716 -0.0368284 0.07 0.7071 -0.7071 0 rod R top
-0.025 -0.034 0.07 1 0 0
.
.
.
-0.01 -0.036 0.0846 0 1 0
-0.01 -0.016 0.1046 0 1 0
[MaterialMapping]
0.49609375 0.49609375 lens center
0.49609375 0.00390625 outer ring 63mm
*** \code ***
You adjust values in column 3 between the partition and the bottom of the section. DO NOT do anything in the [MaterialMapping] nor [Faces] sections!
Here's how to set your own custom sight head height.
- Determine by how much you'd like to raise or lower the head.
- Open pricel.msh, or pricel_D0.msh, with any text editor. (Afterward, do the same for the damaged sight, pricel_D1.msh.)
- Note the Z value (column 3) for the first row following the aforementioned partition. This is in meters (multiply by 1,000 to get millimeters.) Add or subtract your desired increment. Ensure that the new height does not fall outside the permitted range of -0.005 to 0.075.
- Copy all data in the [Vertices] section below the partition to a spreadsheet. When pasting, choose the <TAB> character as the data separator.
-Add your desired vertical increment (in meters!) to all values in column 3. I do this by copying column 3 to another column to the right, such as column i, for instance (you need to avoid columns containing comments.) Then in column 3 itself make a formula adding to column i the desired increment. Such a formula for field c1 might read, "=i1 + 0.0125", which adds 12.5mm, or 1/2 inch. After hitting ENTER, click on the field containing the formula you just made, then click/drag the selection anchor at the lower right corner all the way to the last field in column 3. This will apply the formula to all fields. Then select all data fields (but NOT the new column containing the copied column 3 values!) and paste back into the mesh file.
- Naturally, you first made a backup copy of the mesh, right?! ;->
The existing sight head heights:
- Cr.42 and G-50: 0mm, with the noted element 'datum' of the rod top at -0.005
- MC-202: 23mm, with the datum at 0.018 (0.018 - -0.005 = 0.023)
- MC-200: 75mm, with the datum at 0.07 (0.07 - -0.005 = 0.075)
After setting the sight head height, it will be necessary to raise or lower the height of the reticle mask similarly, as defined in the section [ZZ_mask] in hier.him. Furthermore, it might be wise to adjust the reticle height in section [ZZ_Reticle] as well if your height change is larger than a few centimeters. But only if you can discern the bullet stream is high or low with respect to the 'pipper'. Lastly, the POV height in Body.msh should be adjusted too, certainly if one does not use head tracking gear.
As we thus far have three sight variations where the head is set roughly to 0 inch (0mm), 1 inch (23mm) and 3 inches (75mm), I offer another with head height at 2 inches (50mm). This set of four could well avoid the process of altering the mesh files, if such operation seems onerous. This extra pair of meshes--for the undamaged and damaged configuration--reside in a dedicated folder, obviously named. The mesh files must of course be renamed as appropriate so as to match the names in hier.him.
Old Iron Sights Removed
-----------------------
The MC-200 and MC-202 used to have a small ring sight on a post, located just outside the cockpit. It did nothing, as it was at the wrong place to be effectively used. And it was an eyesore. And the new sight's incorporated iron sight obviates it. The texture resided in the file, gs_lthr.tga, outlined with a transparency in the alpha channel. I simply made the whole sight thingie transparent, and so it's still generated but is now invisible. You could restore it by deleting the gs_lthr.tga file, but the texture could end up being placed higher and thereby reveal no attachment to the cowling.
Canopy Glass
------------
My latest canopy glass 'dirt' texture work has involved reversing the surface normals for all glass surface polygons, by reversing the signs of the values. Why do this? So as to emulate the greater efficiency of forward scattering. A transparent surface having dust, smudges or sleeks/scratches/pits will scatter light more strongly when looking through it toward the light source than when looking in the opposite direction.
Formerly, cockpit glazing had the surface normals defined so as to make the 'dirt' texturing more prominent when opposite the sun. Now this scattering effect is more correctly stronger in the up-sun direction. In the opposite direction, down sun, the texturing now becomes very subtle.
I created a new 'dirt' texture by significantly reducing the contrast between the light and dark parts, so that a gentler, more general hazing--still with brighter bits--is generated. This does not impair visibility to any meaningful degree, but very nicely suggests a less-than-pristine sheet through which one is looking. Moreover, it's hardly to be expected that glazing used in war is going to be pristine--even cleaning by itself can induce fine sleeks. And with altitude changes condensation was not uncommon, with even a small amount bringing out the visibility of the innumerable condensation nuclei of dust. And finally, the presence of a hazing on the glass nicely locates one in the cockpit, enhances the sense of head movement, and nicely differentiates between glazing and open air for canopies partly covered or which are openable.
If you wish to reduce the intensity of the scatter, lower the opacity assigned to the texture in glass.mat or GlassCanopy.mat. The relevant value is the last of 4 defined by the ColorScale parameter:
ColorScale 1.0 1.0 1.0 0.5
The first 4 numbers set the red, green and blue levels, respectively. The last number sets opacity, which (like the color values) can be between 0.0 and 1.0. This is in addition to any transparency already set in the .tga's alpha channel (as is the case for my glass.tga). I've set 0.5 here; you could lower it to 0.3, which will be notably subtler. Of course, setting to 0.0 will make the glass completely clear, which is unrealistic.
If you wish to go back to stock canopy glass treatment, delete from the plane folders:
Glass.msh, Front.msh
Glass.mat, GlassCanopy.mat
Glass.tga, GlassCanopy.tga
------------
I give permission for the free use of this work, in whole or in part, in any other work by any other person or entity.
WxTech
Aug 30, 2015
Author
Topic: San Giorgio Tipo B gun sight for stock Cr.42, G-50, MC-200 and MC-202, Ver. 2 (Read 12229 times)
