SL Phototools: View Angle versus Field of View. HELP!

[Scylla Rhiadra]

So, I'm writing this piece of depth of field in Second Life photography, and I'm having a difficult time trying to differentiate between two sets of sliders in the Phototools in Firestorm, which I think are also both available in Black Dragon and the LL Viewer. These are "View Angle" and "FOV" or "Field of View."

My understanding has always been that these are essentially the same thing, but measured somewhat differently: they both reflect the amount of a view that are recorded by the camera, which is itself determined by the focal length of the lens.

So far as I've been able to determine, the main difference between these two is in how they are measured: View Angle is measured in radians, while FOV seems generally to be a measurement of the distance across the focal or object plane, measured in linear units such as metres. So, View Angle measures, as the name implies, the angle of sensor's capture, while FOV should be the distance of visible field at a right angle to a line drawn between the camera sensor and the photo subject.

If this is correct, adjusting the one should automatically change the other, as increasing or decreasing the View Angle should have the effect of increasing or decreasing the size of the Field of View, a wider angle meaning a larger field of view.

This is, of course, not what happens in the SL viewers: adjusting the one does nothing to change the values of the other.

As confusingly, each slider does something different. View Angle does indeed change the angle of view, and the size of the field captured in the photo, as one would expect, and the values do seem to correspond pretty closely to what one would get by "zooming" or changing lenses in an RL camera.

Field of View, on the other hand, seems only to increase or decrease the Depth of Field, and so far as I can see, in no way affects the shot if Depth of Field is disabled. Also confusing is the fact that the FOV slider runs from 0.1 to 180, which suggests that what it is changing is view angle, measured in degrees rather than radians. Except of course, it is doing no such thing

Finally, unless I am misunderstanding, increasing the field of view should actually widen the depth of field, rather than narrow it, as seems to happen if you adjust this slider in the viewer. Bumping up the FOV, in other words, should decrease rather than increase the amount of blur.

So, basically, I don't understand the function of the FOV slider here. On one hand, it seems to simply split the effect of view angle between two sliders, one ("View Angle") zooming in and out and effectively changing the field of view, and the other ("FOV") impacting on depth of field -- except that, as I've suggested, it seems to do the opposite of what it should do.

Help?

I'll tag a few people here who I know have an interest and/or expertise in cameras or SL photography, but I'm sure there are others who have suggestions to make! @taisiyakarpenko @Nalates Urriah@Orwar@Myra Wildmist

Many thanks,

Yours in confusion,

Scylla

[Orwar]

   So, yes. Kind of.

   Your view angle is the camera's zoom (or, 'how much width goes into the space that is your screen'), whilst FOV specifically tells your viewer what depth of field to simulate (since, well, it's all simulated). For most cases, you want the two to be aligned, which, since they use different metrics, requires you to do a little bit of maths when you set up a shoot. 

   If you're taking a pic in, say, 0.300 rads, you'll have to calculate how many degrees that is to set your FOV to match it, thusly:

   0.300×180°/π = 17.188733854° - i.e. you want 17 on your FOV.

[Scylla Rhiadra]

9 hours ago, Orwar said:

   So, yes. Kind of.

   Your view angle is the camera's zoom (or, 'how much width goes into the space that is your screen'), whilst FOV specifically tells your viewer what depth of field to simulate (since, well, it's all simulated). For most cases, you want the two to be aligned, which, since they use different metrics, requires you to do a little bit of maths when you set up a shoot. 

   If you're taking a pic in, say, 0.300 rads, you'll have to calculate how many degrees that is to set your FOV to match it, thusly:

   0.300×180°/π = 17.188733854° - i.e. you want 17 on your FOV.

Right. So they split two of the effects of view angle into two separate sliders, and then, by employing two different units of measurement for essentially the same thing, made it necessary to use a calculator to harmonize them so that the corresponding values make sense.

Except that, as I noted -- and correct me if I am wrong -- the slider for FOV has the opposite effect that it is supposed to, actually decreasing the depth of field (and therefore increasing blur) as the angle (the "width" that "goes into the space that is your screen," as you say) increases.

On top of which, of course, view angle and field of view are actually dependent upon focal length, which is yet another separate variable that has its own slider that, so far as I can see, does exactly the same thing as the FOV slider. So that's another value that you have to calculate to bring into conformity with view angle. At least this one operates as it is supposed to, decreasing depth of field as the focal length increases.

How does any of this make sense?

Edited May 29, 2022 by Scylla Rhiadra

[Orwar]

4 hours ago, Scylla Rhiadra said:

How does any of this make sense?

   In short, it doesn't, because there are yet more aspects of it that isn't actually taken into consideration - primarily your camera position and the size of your subject. So you shouldn't blindly trust that equalling the FOV with the View Angle = 'realistic' or even 'good'.

   Furthermore, viewer-generated DOF suffers another teeny tiny issue of being scaled to your native resolution, just the same as the shadows are. If you've got a 1920 x 1080 resolution but take a picture in, say, 4K (3840 × 2160), the effects of the DOF will be a lot less prominent (1/4th in magnitude) in the rendered snapshot, even if it looks good in the preview. So just as with shadows, that do the exact same thing (primarily as far as shadow smoothing is concerned), you might have to start tickling around in the debug settings to set things way up (lest the shadows that look perfectly fine in your viewer end up like jagged lines in your saved picture). 

   And then there's the whole alphas = solid as far as generating DOF is concerned, so stray hairs and the like tend to make it so that nothing behind the mesh for the strands won't get any DOF on it (I think BD handles this better than FS though). 

   I've tried to work with the in-world DOF many times, but between the inconsistencies and the issues, I've found that I can usually make a better DOF effect by just taking a depth shot of my setup, or in the case of a non-arranged pic make a hand-drawn blur mask - I'd rather spend those extra 2 minutes in post than trying to get all the settings to play ball (and then you have to go back and set them to something more sensible again, because going to, say, a shopping event with your shadow smoothing cranked up to 11, your computer is likely going to decide it no longer likes you). I'll occasionally pop it on when I'm taking a quick snap of something I'm doing or seeing when out exploring or such, but those pics I always take in native resolution anyway (as I tend to share them directly on Discord rather than uploading them to Flickr or ImgBB). 

[Nalates Urriah]

Orwar has explained the terms well. I think the real answer to your questions and challenges to understanding what the viewer is doing was too quickly skipped over. Saying it is all simulated is the key. And that is probably the most important facet of this thread.

Focal length is a distance from the lens' focal point to the film or focal plane. Think of the effect of looking out a window. The closer you are to the window, the more you can see outside the window, the larger your field of view. The farther you stand from the window, the less you see.

The Depth of Field, what is well focused, is only changed by the camera's f-stop in RL cameras. The actual opening size in the lens, aperture, controls the depth of field. There is no aperture for the render engine. So the simulation is from an equation. And not that good simulation. Plus all the other render problems of a 3D world... think the transparency problems.

The render engine has none of the features or constraints of a RL camera. The Lab created controls on the render engine that sort of relate to a RL camera. Supposedly that was to make it easier for people to understand the render engine settings by using an analogy many people already understood. But the analogy falls short and has inconsistencies. We can do things with the render engine that cannot be done with a RL camera. So there are some controls that don't accurately reflect what happens in a RL camera. The physics of the render engine is simply a set of equations. For RL cameras it is the properties of the materials used to build the camera and light. One simply cannot think of a render engine as one does a real camera.

So you have found flaws in the simulation. Don't over think it.

[Jenna Huntsman]

To add on to this, as someone with some industry experience and have spent a fair amount of time researching the subject:

As the others have said, the key thing to remember is that the camera is simulated - so the camera, again, is not constrained by the laws of optics and optical science.

Another thing to keep in mind is that the default settings are.. just.. bad. They actually represent a lens which couldn't really exist, so these are best changed to represent a more accurate sample. To use computer science terms, GIGO. (Garbage In, Garbage Out).

The Debug settings "CameraAngle" and "CameraFieldOfView" are unrelated - namely, as they use different units and are measuring a different thing, see the wiki page linked below (Under "Debug Setting Glossary").

In terms of screenshots and SL photography specifically: Don't use "overscan" if you can avoid it. The SL rendering engine handles resolution scaling pretty poorly, and not all postprocessing effects will be scaled the same. (That's not to say you can't use it).

Take a read of my wiki page, it contains a lot of useful info relating to DoF and lighting within Second Life:

https://wiki.secondlife.com/wiki/User:Jenna_Huntsman#Lens_Settings

[Scylla Rhiadra]

42 minutes ago, Jenna Huntsman said:

To add on to this, as someone with some industry experience and have spent a fair amount of time researching the subject:

As the others have said, the key thing to remember is that the camera is simulated - so the camera, again, is not constrained by the laws of optics and optical science.

Another thing to keep in mind is that the default settings are.. just.. bad. They actually represent a lens which couldn't really exist, so these are best changed to represent a more accurate sample. To use computer science terms, GIGO. (Garbage In, Garbage Out).

The Debug settings "CameraAngle" and "CameraFieldOfView" are unrelated - namely, as they use different units and are measuring a different thing, see the wiki page linked below (Under "Debug Setting Glossary").

In terms of screenshots and SL photography specifically: Don't use "overscan" if you can avoid it. The SL rendering engine handles resolution scaling pretty poorly, and not all postprocessing effects will be scaled the same. (That's not to say you can't use it).

Take a read of my wiki page, it contains a lot of useful info relating to DoF and lighting within Second Life:

https://wiki.secondlife.com/wiki/User:Jenna_Huntsman#Lens_Settings

Thank you! The lens settings are really useful. I've been using Myra Wildmist's for a long time, and they've been useful, but these are more comprehensive.

There's a great deal of useful material here, which I plan to study! Thank you!

[NiranV Dean]

I think there is some confusing in this topic.

View Angle and Field of View are not the same, not at all. The confusion comes from Field of View wrongly being labeled View Angle in Second Life (whereas you would expect it to be correctly Field of View). Field of View in SL however refers to a Depth of Field specific setting which determines the Field of View of Depth of Field whereas View Angle controls the Field of View of your camera.

[Scylla Rhiadra]

7 hours ago, NiranV Dean said:

I think there is some confusing in this topic.

View Angle and Field of View are not the same, not at all. The confusion comes from Field of View wrongly being labeled View Angle in Second Life (whereas you would expect it to be correctly Field of View). Field of View in SL however refers to a Depth of Field specific setting which determines the Field of View of Depth of Field whereas View Angle controls the Field of View of your camera.

Thanks Niran, but your response actually leaves me even more confused.

Yes, I get that "View Angle" in SL = Field of View.

I'm not getting what you mean by the "field of view of depth of field," however. Field of view is field of view, whether there is a shallow or very large depth of field: there aren't two different "kinds," with one that especially applies to depth of field. So you can't really talk about "field of view" where depth of field doesn't apply: all RL cameras actually have depth of field, always: all cameras (other than ones designed for tilt-shift pics) have only one focal point. It's just that it can be adjusted so that it is so wide that, for practical purposes, everything in the scene is "acceptably sharp."

Can you explain what you mean by "field of view of depth of field"?

[Jackson Redstar]

since it is all simulated anyways, maybe just a re write to simulate using it on a RL camera. the 2 main things on a RL camera that affect DOF would be amount of zoom and Appaure and the focus point but we do that by alt flick on the subject or having the focus follow the pointer. We can control amount of zoom essentuialy using the ctrl+0 , so if there was a slider that tied into that function and labeled like a RL camera, like 50mm, 85mm, 180mm 2000mm etc moving the slider would be the same as ctrl+o to zoom in then an appature slider label like a RL (so like 1, 3.5, 4, 5.5, 6, 7.8 etc etc ) camera seems would clear up a alot of confusion. I am not really sure since this is all simulated if there is even the need for things like CoC but maybe, but in essence re-write the UI sliders to match a RL camera -  i think alot of people would understand that alot better. And since unlike a RL camera, our appature options don't need to be compensated for with shutter speed or ISO settings for the correct exposure, users who dont even know what DOF is in the first place can watch any tutorial to get a understanding of DOF on RL cameras and how it works

 

Edited June 5, 2022 by Jackson Redstar

[NiranV Dean]

On 6/5/2022 at 5:51 AM, Scylla Rhiadra said:

Thanks Niran, but your response actually leaves me even more confused.

Yes, I get that "View Angle" in SL = Field of View.

I'm not getting what you mean by the "field of view of depth of field," however. Field of view is field of view, whether there is a shallow or very large depth of field: there aren't two different "kinds," with one that especially applies to depth of field. So you can't really talk about "field of view" where depth of field doesn't apply: all RL cameras actually have depth of field, always: all cameras (other than ones designed for tilt-shift pics) have only one focal point. It's just that it can be adjusted so that it is so wide that, for practical purposes, everything in the scene is "acceptably sharp."

Can you explain what you mean by "field of view of depth of field"?

As far as i can tell the field of view of depth of field is essentially what field of view in RL photography would be. Not sure what that means, you'd have to go read on wikipedia or some professional depth of field explanations on google to get a proper explanation what "field of view" really is. I'd describe it as view angle but separate but not really. Both work together and they add on top of each other. Turning down view angle will exaggerate the effects of field of view. From a user standpoint however field of view is essentially just "how strong do you want your depth of field to be according to the rest of your settings"

[Jenna Huntsman]

1 hour ago, NiranV Dean said:

As far as i can tell the field of view of depth of field is essentially what field of view in RL photography would be. Not sure what that means, you'd have to go read on wikipedia or some professional depth of field explanations on google to get a proper explanation what "field of view" really is. I'd describe it as view angle but separate but not really. Both work together and they add on top of each other. Turning down view angle will exaggerate the effects of field of view. From a user standpoint however field of view is essentially just "how strong do you want your depth of field to be according to the rest of your settings"

So, to clear this up, this graph shows you how DoF works from an optical science perspective:

(Source: https://www.strollswithmydog.com/equivalence-focal-length-fnumber-diffraction/ )

CameraAngle is the physical angle of the camera's lens - i.e. what you actually see. Changing this is equivalent to swapping a lens, or zooming in or out. It directly affects the camera's Field of View. This actually represents the width of the visible frame. (Thus, the overall size of the image circle).

CameraFieldOfView is purely data which is used for the Depth-of-Field calculations, and nothing else. This represents the height of the frame.

In an ideal world, we'd actually specify the FoV diagonally, as then it can be decoupled from the aspect ratio of the 'sensor' (the user's monitor), but that's just how it is.

[Scylla Rhiadra]

22 hours ago, Jenna Huntsman said:

CameraFieldOfView is purely data which is used for the Depth-of-Field calculations, and nothing else. This represents the height of the frame.

Thanks, Jenna!

Still grappling with this, as I'm finding a lot of confusing and often contradictory information about Field of View. Which, I gather, can be measured as an angle (angular field of view) or a distance (linear field of view)?

Given that CameraFieldOfView ("FOV" in Firestorm, and "Field of View(°)” in Black Dragon) is measured in viewer as an angle, I'm assuming that what it represents is AFOV (where A = "angular" rather than "actual" or "apparent").

So, as  you say, FOV is "data," produced by a calculation that includes angle of view (itself dependent on focal length), sensor size and type, and distance from subject. It can be adjusted in-camera only by changing lenses (focal length/angle of view) or distance to subject.

But I am confused with how this works in-viewer, as increasing the value of CameraFieldOfView narrows the depth of field. Should not the opposite happen, just as increasing the view angle (which, again, I know is not quite the same thing) widens the depth of field?

And an ancillary question: why, in your excellent set of specifications for different lenses and focal lengths, do you leave the CameraMaxCoF unchanged at 29.0 for all?

(I might also ask why the 8m field distance and the native sensor size dimensions are left at defaults in the calculator you link to, but I don't want to put you to too much trouble. You've been more than helpful!)

[Jenna Huntsman]

1 hour ago, Scylla Rhiadra said:

But I am confused with how this works in-viewer, as increasing the value of CameraFieldOfView narrows the depth of field. Should not the opposite happen, just as increasing the view angle (which, again, I know is not quite the same thing) widens the depth of field?

I may be mistaken, but I believe that increasing the CameraFieldOfView value will shift the Depth of Focus forward (or backward, if decreased) of the simulated image plane ('sensor'), thus increasing bokeh.

(Source: https://www.opticsforhire.com/blog/understanding-depth-of-field )

1 hour ago, Scylla Rhiadra said:

And an ancillary question: why, in your excellent set of specifications for different lenses and focal lengths, do you leave the CameraMaxCoF unchanged at 29.0 for all?

Essentially it's a guide value. It's assumed to be ~0.029 mm for full-frame cameras, but in reality CoC is a function of a sensor's pixel spacing (the gap between pixels on a camera sensor - this means that every camera has a slightly different (actual) CoC!)

Given that my specs always assume that we are using a full-frame sensor, using the guide value does a good job. You *can* recalculate the entire table based off a different sensor size (thus, different guide CoC value), but for the sake of keeping the length of the table under control I have only given the values for full-frame.

1 hour ago, Scylla Rhiadra said:

(I might also ask why the 8m field distance and the native sensor size dimensions are left at defaults in the calculator you link to, but I don't want to put you to too much trouble. You've been more than helpful!)

The field distance is used to calculate magnification, which we don't need in our case.

[Scylla Rhiadra]

5 minutes ago, Jenna Huntsman said:

I may be mistaken, but I believe that increasing the CameraFieldOfView value will shift the Depth of Focus forward (or backward, if decreased) of the simulated image plane ('sensor'), thus increasing bokeh.

(Source: https://www.opticsforhire.com/blog/understanding-depth-of-field )

Essentially it's a guide value. It's assumed to be ~0.029 mm for full-frame cameras, but in reality CoC is a function of a sensor's pixel spacing (the gap between pixels on a camera sensor - this means that every camera has a slightly different (actual) CoC!)

Given that my specs always assume that we are using a full-frame sensor, using the guide value does a good job. You *can* recalculate the entire table based off a different sensor size (thus, different guide CoC value), but for the sake of keeping the length of the table under control I have only given the values for full-frame.

The field distance is used to calculate magnification, which we don't need in our case.

Brilliant. Thank you so much, Jenna!

[Myra Wildmist]

I thought I'd try to add to this a little. FoV does get confusing. Here's a helpful article on FoV: https://www.studiobinder.com/blog/what-is-fov-definition/

FoV is suppose to be about how much* of the view you can see. I think they reversed FoV and View Angle in the settings. You can see this by playing with the sliders.

When I take photos in SL, I try to simulate working with RL lenses. This is my preferred style. So I set my FoV  the same as my View Angle. One note: View Angle is in radians in SL, which makes things even more confusing since in RL the Angle of View (proper rl terminology) of a prime lens is measured in degrees.

So if I'm shooting with a portrait lens, such as an 85mm lens, my View Angle will be .497 (radians), my FoV is 28.5, and my focal length is 85. I do not adjust my FoV to change my DoF. I use my f-numbers. This is what I do in RL. I know SL is different but I'm applying my RL methods to SL.

Here are the specs for a full-frame 85mm lens. You can see the Angle of View (FoV in SL) in the specs: https://www.bhphotovideo.com/c/product/1497159-REG/nikon_nikkor_z_85mm_f_1_8.html/specs

In my opinion, it would make more sense, from a RL perspective, if the FoV, View Angle, and Focal Lengths were linked, initially, and the user had the option* of decoupling them. As they're presented in FS (LL too? Haven't used that in a while.), it's overwhelming and confusing for new SL photographers to use these.

And just for fun, here's a handy tool for "making" your own lenses:

https://imaging.nikon.com/lineup/lens/simulator/

[Scylla Rhiadra]

7 hours ago, Myra Wildmist said:

I thought I'd try to add to this a little. FoV does get confusing. Here's a helpful article on FoV: https://www.studiobinder.com/blog/what-is-fov-definition/

FoV is suppose to be about how much* of the view you can see. I think they reversed FoV and View Angle in the settings. You can see this by playing with the sliders.

When I take photos in SL, I try to simulate working with RL lenses. This is my preferred style. So I set my FoV  the same as my View Angle. One note: View Angle is in radians in SL, which makes things even more confusing since in RL the Angle of View (proper rl terminology) of a prime lens is measured in degrees.

So if I'm shooting with a portrait lens, such as an 85mm lens, my View Angle will be .497 (radians), my FoV is 28.5, and my focal length is 85. I do not adjust my FoV to change my DoF. I use my f-numbers. This is what I do in RL. I know SL is different but I'm applying my RL methods to SL.

Here are the specs for a full-frame 85mm lens. You can see the Angle of View (FoV in SL) in the specs: https://www.bhphotovideo.com/c/product/1497159-REG/nikon_nikkor_z_85mm_f_1_8.html/specs

In my opinion, it would make more sense, from a RL perspective, if the FoV, View Angle, and Focal Lengths were linked, initially, and the user had the option* of decoupling them. As they're presented in FS (LL too? Haven't used that in a while.), it's overwhelming and confusing for new SL photographers to use these.

And just for fun, here's a handy tool for "making" your own lenses:

https://imaging.nikon.com/lineup/lens/simulator/

Thanks Myra! I have of course seen your really useful pieces on replicating lens types in Kultivate, and have been using your settings there for, well, years.

I have a reasonable grasp, I think, of how DoF works in RL cameras: what I've been trying to grapple with is the relationship between the Phototools settings for DoF and their RL analogues -- and come to the conclusion that it's a highly imperfect match. As you say, they may have mixed up FoV and View Angle. Certainly, FoV seems to have the exact opposite effect of what it should: increasing it should effectively widen the DoF, but it seems instead to narrow it.

So I'm not at all sure that setting FoV, View Angle, and Focal length to the "correct" RL numbers is actually replicating what you'd get in an RL camera from that particular setup -- particularly, as I said, because FoV seems to work backwards. But given the fact that DoF is turned "off" by default most of the time anyway, which is of course also not really possible in an RL camera, it may be expecting too much.

I agree entirely that it would have been great if they had linked FoV and View Angle to Focal length, so that adjusting one automatically changes the others (as would happen with an RL camera if you changed Focal Length), but I guess that's too much to ask. Also, wouldn't it have been nice if Aperture (f-number) also affected the lightness/darkness of the pic? And ISO and shutter speed would have been nice . . .

(The LL viewer has the same settings as FS, by the way, but hidden in the Debug settings, and with slightly different names. Black Dragon is much like FS, but with View Angle listed separately from the others.)

Thanks for this! And I'm having fun with the lens simulator!

[Jenna Huntsman]

18 hours ago, Myra Wildmist said:

FoV is suppose to be about how much* of the view you can see. I think they reversed FoV and View Angle in the settings. You can see this by playing with the sliders.

When I take photos in SL, I try to simulate working with RL lenses. This is my preferred style. So I set my FoV  the same as my View Angle. One note: View Angle is in radians in SL, which makes things even more confusing since in RL the Angle of View (proper rl terminology) of a prime lens is measured in degrees.

I actually disagree with the view that the FoV and View Angle are mixed up.

I did some digging in the backend to see how the DoF shader works in SL, and came up with the following process:

1. Convert given FoV value to the lens' Radius Of Curvature
2. (Unknown (to me) formula, see below)
3. (Unknown, final result) - POSSIBLY - calculate dioptre from index (Value calculated in step 2) div (/) radius of curvature

I believe steps 2 and 3 are used in calculating the power of an "equivalent" defocusing (concave) lens which is used by the shader; although not 100% sure. (My brain isn't working well enough right now to dig much deeper, may revisit another time).

As a side note, I found out why the level of bokeh changes as the CameraAngle value is changed -
When calculating the power of the *final* "defocusing" lens, the code actually retrieves the *actual* camera's FoV value, and this is mixed into the formula presumably as some sort of failsafe. Disconnecting the 2 would actually only involve swapping out the name of 1 variable.

 

[Myra Wildmist]

2 hours ago, Jenna Huntsman said:

I actually disagree with the view that the FoV and View Angle are mixed up.

 

That's fine, Jenna. I've posted the info and some useful links. The View Angle parameter behaves as a FoV adjustment: it increases the field of view of the scene. You can see this for yourself by using the sliders. If you accept that, as defined, RL field of view increased and decreases how much you can see, then SL View Angle behaves like RL FoV should. SL FoV, on the other hand, behaves as a RL Angle of View.

So I stand by what I said. To state if more clearly: SL View Angle is really RL field of view and SL FoV is really RL Angle of View.

[Jenna Huntsman]

14 minutes ago, Myra Wildmist said:

That's fine, Jenna. I've posted the info and some useful links. The View Angle parameter behaves as a FoV adjustment: it increases the field of view of the scene. You can see this for yourself by using the sliders. If you accept that, as defined, RL field of view increased and decreases how much you can see, then SL View Angle behaves like RL FoV should. SL FoV, on the other hand, behaves as a RL Angle of View.

So I stand by what I said. To state if more clearly: SL View Angle is really RL field of view and SL FoV is really RL Angle of View.

Oh, sorry, I read and interpreted wrong in that case. I do agree that the CameraFieldOfView value is roughly equivalent to AoV. I actually already had that as part of my "Calculating Lens Presets" section on the Wiki.

I did reach out to one of the members of the Graphics team - in their words -

Quote

it's all fakery and just trying to get a similar near and far focal plane as you'd get from a real lens of a given focal length and f-number

the actual size of the "defocus" isn't physically modelled or anything

So in the end it somewhat doesn't matter about the semantics, as the values aren't used in calculating a physically-accurate lens.

[Myra Wildmist]

I would substitute the word "fakery" with simulated. After all, isn't that what virtual reality is about? We're simulating real life. In this case, we're simulating lenses. Roughly. : )

Trying not to go down the rabbit hole, but SL has a long way to go on simulating RL. Please give me ray tracing. Fix light behavior. Fix shadows. hmmm it's a long list. But it's a direction we should be going. We strife to be more "real" in SL all the time. Our virtual cameras shouldn't be any different.

As for terms, they matter. People read about RL FoV and expect it to be similar/equivalent to SL FoV. Mixing the terminology confuses people. There are photography features built in. Those features would be more accessible to everyone if the jargon/terminology paralleled real life.

Photography in SL is huge and important to people, and it's just going to continue to grow as more people get turned on to virtual photography. I'd like to see LL focus more on photography (No pun intended.). Fixing these small issues helps. They really need to revisit all these settings, imo, and make them more accessible to everyone. Right now, it's a bit of an arcane mess.