Shipbucket

Ok, I admit, that the highlight/darker shading on the bow of "my sub" could be thinner at the forward extreme (sorry for that), but the drawing was meant to show the general concept and I didn't considered the curvature of the bow to be particularly relevant here, because Dekabrist class does not have a teardrop-shaped hull anyway.
(That said, while Your representation might be slightly better when it comes to upper part, I'd say that IMHO the darker part of the bow is too thin - for simplicity, this "my sub" is meant to have a perfectly circular cross-section of the hull, so the "highlight" and "darker" shaded stripes would have equal widths all the way - approx. 1/7th of the diameter for a perfect circle, only, due to the fact that source of light is in front of the vessel, the highlighted stripe would extend further than the darkened one - as can be seen on the sample of Pr.877's bow).
Also, if we are going into minutiae details of SB-style shading of spherical-ish forms in such spatial relation to the source of light as we have in SB style, then in the front-view it would look approximately like this (without counting pixels, as it's just a simple hand-made sketch, but also while taking into account, that we can't depict sub-pixel-sized values, in the sense that any given pixel certain color, or not, but we can't draw a 1/2-pixel-wide line or 1/5-pixel-wide line etc):


And in this case that orange lens-shaped part in the middle, in my interpretation of drawings, is simply not visible when viewed at 90-degrees angle.

And as for the bottom fin - there are IMHO no clear rules about it - that's something open for interpretation, but even if we were to adapt Your way of shading it, then I'd say that because the hull gets wider toward the bow, so the shade (red stripe along the hull) on the bottom fin should be getting wider towards the bow too.

There is no lens shaped part, it is just that the red and yellow areas.
[
become so narrow when seen from the side, they are less then 1 pixel thick. So the opposite is true: the orange part remains visible because when seen from the side, the red and yellow become too thin to be visible. In a top view, the yellow would be the only one to remain visible as the orange would no longer be there at the extreme bow.

And yeah I know it was meant as the general concept but I know I commonly re-use images like that as they have the habit of becoming relevant at some other point so thought it important to discuss the first time I noticed it
I agree that top and bottom should be symmetrical but I was too lazy to do that there 20 second edit is to blame.

As for the fin: the way you shaded it follows practice as is common in FD scale. I usually try to avoid mixing shading styles, so large shaded areas show shape (as what we are discussing here) while overhangs are represented the way they are done in superstructures as well, matching the above the water shading with the shading below the waterline.

As for the width of the shading under the fin, the exact dimensions and shape of that shading could differ somewhat but again, 20 second edit. I honestly did not put that much thought into that apart from showing the general idea, that is reserved for proper drawings and/or tutorials

_________________
Drawings are credited with J.Scholtens
I ask of you to prove me wrong. Not say I am wrong, but prove it, because then I will have learned something new.
Shipbucket Wiki admin

I would say, that by the logic of SB style shading there has to be lens-shaped part on an object of spherical-ish shape, because (assuming we're not talking about object simply too small to fit 3 shades into it) immediate "switch" from highlight to dark could happen only when said switch is abrupt, and here it's not the case. Also, the source of light is in front of the bow, so the "not darkened" (highlight and basic) area goes past ("below") the "central point" of the cross section - meaning, that the centre point of the circle of the drawing is also "shine upon".
The way You shaded it (without the lens) IMHO could be relevant if the source of light was above the ship, not above and in front of.

Well, if You're replying with "20 second edit is to blame", then my original drawing was also "20 second...", and yet it didn't deterred You from starting this whole discussion about it's supposedly horrible errors.

I am back at school again, to study interior design!!

and this is how we look at shadow and highlight:

Well, to be fair, it was not my intention to make this a whole discussion. My point about the 20 second edit was that we should not find new issues with our 20 second edits every time we try to quickly explain something to each other The important factor is me pointing out a single point and you saying you always do it this way so we should discuss it. Then my example not being symmetrical is a different subject that is in my opinion not worth it being discussed, as that is something I try to avoid doing in any real drawings.

That said. Could you please explain to me where this lens-shaped thing comes from? Was that what you got from my explanation because then I must have explained things wrong or at least not clearly.

My thought is this: if the cross section remains constant (let's use a perfectly circular shape in this example) the shaded area relative to the total cross section height remains constant. In this case, it is a circle, so the 45 degree shading rules state that the middle 70.7% is the base shade and the bottom and top 14.7% are highlighted and shaded. This means between the black line on your submarine, at the straight section, 49 pixels should be base shade and the bottom and top 10 pixels should be shaded. This amount decreases as the cross section does, similar to how this happens in the stern of your example. (I know the numbers don't add up to 70 but here you run into trouble with half pixels, I calculated this in excel and just let it round to the nearest whole pixel)

Using this method, the cross sections at the very front are 4,14,18,22,24 etc on your submarines bow; so the shaded area at the bottom would be 1,2,3,3,4 pixels. And using that method, we see that even at the very front, an base shaded area of 2 (technically even 2.8, as the shaded areas are technically 0.6 pixels) pixels wide remains. Even if the cross section would only be a single pixel, that pixel would be 70.7% base shade.

Do I think something like the above excel sheet results should be made for every drawing? Certainly not. We work as artists and there is room for interpretation. But using the 45 degree shading rule, on a circular or near circular cross section, even with an big bulbous front, the dark shade and the highlight can never meet as there is always an area of base shade in between that is more then twice as big as the shaded and highlighted areas combined. Yes, in reality, those shades would meet as the cross section reaches zero, but we can only draw whole pixels

And a very small note, on hull shading it is common to ignore the light coming from the front as that would make it impossible to accurately represent any shape. The one exception is the very clear case of transom sterns which are shaded, and similarly some flat bows can get a lighter colour to match the superstructure shading...... but the gentle curving shape of a ships hull is extremely hard to shade with light coming from the front, made even more complicated by round shapes..... as heuhen's image proves. I think it would be beyond the spirit of shipbucket to go that way

_________________
Drawings are credited with J.Scholtens
I ask of you to prove me wrong. Not say I am wrong, but prove it, because then I will have learned something new.
Shipbucket Wiki admin

No, actually I did not said "I always do it this way so we should discuss it. If You construed my "I'm not sure" as a genuine question, then my apologies - it was sarcasm. My intention was exactly to the contrary: I'm doing it like this for 10 years, and for that 10 years you had opportunity to write I'm doing it wrong like nearly 3500 times - and you didn't, so... It was you, who insisted on continuing this discussion. :>

As for "where this lens-shaped thing comes from"...
Yes, ignoring of the light coming from the front tends to be ignored on hulls of surface ship - and might be even largely ignored on "old-style submarine hulls", but the hull in question (on "my submarine") is teardrop-shaped (yeah, sort of, crudely... ), therefore IMHO demand consideration of certain aspects that might be otherwise ignored on most other kinds of hulls.
So, the "lens" is there because (disclaimer: THE WAY I SEE IT ):
1) on "typical" hull the 45* rule tends to be applied only along the X-axis (or whichever else you want to substitute here - I'm calling it here "x" because I'm mentioning it as first) or the "cut" perpendicular to ship's length. But for a bow of teardrop-shaped hull (or nose of 747, or aircraft propeller cone...) there is also place for Y-axis and Z-axis, which also have their own 45*s, which makes the "highlight/dark" wider at the tip than it would result from your "x-axis-only" interpretation.
2) in the absence of sharp breaks, in SB style, "hightlight shade" and "shadow shade" can't meet - there has to be an intermediate ("basic") shade between them. In my interpretation, although it might look like they meet along certain length, in fact they merely "approach each other by corners" and the fact that it looks like they meet is because we can draw only full pixels, while at the same time while watched at 90* from the side the "lens" simply would not be visible due to being - from such perspective - less than 1 pixel thick (because - due to common practice of "rounding" fractions - >0,5px becomes 1px and <0,5px becomes 0px). And the "lens" is below the axis of symmetry (the centre of circle that makes the cross-section) is because light shines from top and front, therefore it shines on that "centre of circle" and only "somewhere below" that light shines insufficiently strongly to keep it "highlighted" and therefore warrants transition from "highlight" to "basic" shade (before transferring to "darker"). That's the way I see it, and that's the interpretation of shading that stood (in relevant parts) behind the way I was making - reasonably well received within the community - drawings for the last 10 years, during which years, as mentioned, you expressed no concerns about it at all.
(oh, I'm never, ever making any exact calculations regarding the 45* of Y-axis and Z-axis of the bow/nose - in fact, I think I developed this manner of shading before the "45* rule" was even formulated - it's down to "this is more above and that is more below, at a glance ")

This is an extremely crude, done in a careless manner and without concern for detail or even too much symmetry illustration of the basic concept of my understanding of the issue.

I'm sorry for not mentioning this 10 years ago, but I honestly did not spot it. Which might be because I mostly work in the SB section and not in the FD section, and also because it is single pixels on a big drawing, which I never properly looked at until the above discussion drew my attention to it. It is probably fine to say that a few pixels off is no issue on the existing works but it might be better to "fix it" in future works, IMHO. But in the end these are of course your drawings and you decide if anything needs changing (or even everything)

That said, I understand your "lens thing" now. It is what happens when the fore and aft shading overlaps with the top and bottom shading. And I have to admit, I am not too sure what to say about it.
- On the one hand, it is clearly wrong. Reality does not have light from the top and light from the front, the light comes at an angle. In reality, the dark section would just be lower down the ships hull as more of the front and sides is lighted up.
- However, such a realistic way of drawing shading on things would complicate things massively, requiring shipbucket artists to make 3D models or draw light models. The simplicity of the 45 degree rule or other shading rules would disappear for the sake of realism.

I personally choose the cowards way out to do shape shading (for example on ships hull) based on the 45 degree rule which assumes a top down light source, even if the superstructure and parts shading gets some fore-and-aft lights. In hindsight, I think I did so too on the few FD drawings I did, shading that entirely with top down lighting and no forward lighting.

In the end, there is merit in every way of drawing things as long as we remain consistent within the drawing itself, I guess.

A very small note I like to add, is that this lens shape solution does not explain the examples you showed of the sub and Boeing noses with fully symmetrical top and bottom of the highlight and shade.

_________________
Drawings are credited with J.Scholtens
I ask of you to prove me wrong. Not say I am wrong, but prove it, because then I will have learned something new.
Shipbucket Wiki admin

1) I am not planning to "fix it" in future works, because I consider my shading style to be entirely consistent with the spirit of SB/FD shading.

2) Reality has "light from the top and light from the front" - it's called multiple sources of light, but that is NOT what I mean here. I mean that "your" interpretation of 45* shading "flattenes" it to just one dimension, while mine takes into account also other two dimensions - still with a single source of light, and while keeping in mind that SB shading is, by it's very nature and use of extremely limited amount of arbitrarily chosen (and tuned by their relative values) shades, a stylized representation of shading, whose purpose is not as much represent the "realistic depiction of shades as such", but rather to more clearly inform the viewer about shapes of object in question. (How effectively my style of shading conveys that information is another matter, but also completely another discussion that would demand entirely different set of arguments)

3) I don't see this way of drawing shading as particularly realistic, in fact, I consider it (as mentioned above) highly stylized, and am not doing any calculations nor 3D models. It's all "rule of thumb" and arbitrary "looks ok to me". I also don't see it as a "threat" to simplicity of 45* rule (unless - perhaps, just perhaps - in front-view depictions, which aren't exactly abundant so far), because in SB-scale context it's practically limited to teardrop-shaped submarine hulls and bulbous bows, and in both cases, when viewed at 90* angle (standard elevation view) that "lens" would be too small to be visible and effectively would be "implied" rather than "drawn".

4) Shading Pr.877's bow and B747's nose shading are not fully symmetrical - in both cases "highlight" in the rightmost column of pixels extends further down past the middle point of said column (like on the drawing from my previous post). The "lens" is not visible there - it's implied.

I think we are essentially agreeing here in some way. No shading method will be perfect. I don't think your method is accurate (the 'lens' makes no sense to me even though I can see where it is coming from) and I think the way you have chosen to represent it is confusing ("implied" in a way that can be confused to just having a different cross section shape) hence why I said I will keep avoiding this issue by indeed keeping to work in one dimension for shape shading.

As said, I didn't notice it for 10 years so it is not as if it matters too much if it is right or wrong. I do know one thing though, and that is that I get the feeling you feel assaulted in some way by this discussion and that is not my intention (I just want to help everyone to create the best work possible) so even when no perfect conclusion is reached yet, I am ending my replies on this subject here.

_________________
Drawings are credited with J.Scholtens
I ask of you to prove me wrong. Not say I am wrong, but prove it, because then I will have learned something new.
Shipbucket Wiki admin

Well, indeed, there could be not "perfect" shading method and it's all fundamentally a matter of interpretation. I think we can indeed agree on this one thing.
I also think we can agree that this discussion lasted long enough (especially given the fact that shading of teardrop-shaped bows has pretty much no relevance to the shape of bow of Dekabrist-class submarine ), so I wish You a good night and good rest of the weekend.