Some extras about professional tricks Automatic translate

The topic of this chapter has actually been discussed in previous chapters. There were many examples of how, in order to reduce the number of errors in the image, the artists shifted them (of course, intuitively) to those plans that were not subject to the image. This approach is the basis of such a wonderful (practically error-free) perspective system, as axonometry, and can be quite successfully applied in isolated transmission of not only the closest plan. If the image is subject to a sufficiently deep plan and, it would seem, errors are inevitable, you can try not to reproduce them. This was also already discussed in the analysis of the Provencal Farm near Gardan landscape, where Cezanne’s chosen version of the perspective system shifts errors to the transmission of verticals, which are almost absent in the landscape, and therefore errors remain potentially possible, but unrealized (except for slightly distorted proportions of houses).

The two examples cited show that sometimes artists manage to avoid portraying elements that carry particularly undesirable distortions. But there are times when it is almost impossible. Then the experience of art suggests the methods we have already described for concealing the inevitable distortions - cutting off and masking the most odious elements of the image.

To clarify this idea, we turn to the canvas by Y. Pimenov "Before entering the stage." Here the artist shows the interior, trying as accurately as possible to convey the geometry of horizontal planes: floor, table, dressing table, sofa. To a certain extent, this is understandable, because it is the named horizontal planes with objects located on them that attract the attention of those who look at the picture. The inexpressive distant wall and mirror appear to be completely secondary objects. The fact that it is precisely the horizontal planes that are important for the artist can be seen, in particular, from the fact that the floor boards are transmitted to them by a series of almost parallel lines (we will not forget that a person sees close space almost axonometrically). As you know, in the artist’s version of the perceptual system of perspective, the inevitable errors are shifted to the vertical image. Therefore, the first thing that Y. Pimenov does is refuse to show the upper part of the walls and the ceiling. In fig. a diagram of the image of the same room with the ceiling and the upper part of the walls is given (the construction technique can be found in my previous book [Rauschenbach, 1986]. The above diagram shows that only by cropping the image from above, Yu. Pimenov managed to show the actress’s restroom, and not some well to the bottom of which it is lowered.

The fight against distorted vertical transmission does not come down to cutting off the unnecessary. It’s enough to pay attention to how fuzzy the vertical objects are written and hidden in deep shadows, concerning the floor (this is the bottom of the dressing table and the bottom of the sofa), and the front wall of the dressing table, which would turn out to be unnaturally high, is masked by the image of the chair and the surface of the round table with a cup on it. The whole set of techniques that the artist tactfully resorts to has, from the perspective of the perceptual-system of perspective, one goal - not to depict, hide, mask the inevitable errors of the transmission of the geometry of natural visual perception that are characteristic of the used version of the perspective system.

I.E. Repin also resorted to similar measures in the famous painting “They Didn’t Wait”. It is written almost strictly according to the laws of the Renaissance system of perspective. Here the artist was faced with the task of hiding a very significant drawback of this promising system - the inconsistency of the scale of the image of different plans; the task of overcoming the unnatural increase in foreground objects was especially urgent for him. To clarify the above, in fig. The perspective design of this painting is shown. If the artist thoughtlessly transferred the walls of the room completely, to that plan, which is determined by the lower edge of the picture, then a completely unnatural image would have arisen. In a real room of such a size, a person almost does not feel the apparent narrowing of the room, and this is well conveyed in the painting by Yu. Pimenov, where the floor boards are shown with almost parallel straight lines. The perspective design diagram in the figure clearly shows the absurdity of the simultaneous image of the width of the AA room at the far wall and the explosive at the bottom of the picture. To hide the image of this absurdity, I. E. Repin cuts off the areas adjacent to the unnatural angles B and makes the width of the picture almost equal to the width of the image of the far wall. As a result, the viewer has a feeling that the width of the room, which has a rectangular shape in plan, almost does not change in the picture.

In addition to cutting off those parts of the image that would interfere with the artist, I.E. Repin makes extensive use of techniques to mask inevitable distortions. It should be noted that the reduction required by the Renaissance system of perspectives of the depicted sizes of objects as they move deeper into space can be completely differently perceived by those who look at the picture. This can be illustrated by the following example. A perspective diagram shows the narrowing of one of the floor boards - at the far wall it has a width a, and at the bottom edge of the picture width b. The width of the board changes in the figure by about 2.5 times, which is almost imperceptible due to the narrowness of the board. The exact same change in the width of the room, that is, 2.5 times - from the size of AA to the size of an explosive - seems huge and contradicts the natural visual perception. This is backward: the narrowing angle of the board is only about 10 °, so it seems to be bounded by almost parallel lines, that is, almost the same as it is visible to a person, while the angle between the image of the iol borders exceeds 100 °! But in the natural visual perception, these boundaries are almost parallel. Therefore, Repin calmly, without masking at all, writes out floor boards, but is very concerned about hiding the unnatural change in the distance between the side walls in the picture.

The left corner of the room lies in the shade, and the wall is interrupted by the image of the door, and all this together makes it impossible to feel the unnatural expansion of this part of the room, especially since the light doorway and the wall brighter than the floor create the feeling that the floor border in this corner almost rectangular. The right border of the floor is completely covered by furnishings and therefore completely invisible. Such a masking of the configuration helps to enhance the visual illusion that arose as a result of cutting off the unnatural angles shown in the diagram of the perspective design of the picture. Repin compensates for the unwanted increase in the size of foreground objects, which is characteristic of the Renaissance system of perspective, by the fact that the mother of the person entering the room is shown not standing, but standing up. Her bent figure is justified not only psychologically, but also compositionally: her unnaturally large (compared to the growth of the incoming) growth, which she would have if she was depicted standing straight, was softened. This is also facilitated by the fact that the lower part of the mother’s dark figure is almost completely obscured by a light armchair.

The desire to mask the configuration of the floor in the picture is also visible on the canvas already discussed by A.P. Ryabushkin, “They Expect the Tsar to Come Out”. The group of boyars standing on the left excludes the opportunity to see the floor configuration of the room. And it is quite justified. If the borders of the room were shown to be consistent with the narrowing of the scarlet path, then its far wall would be so small that no more than one or two boyars would be placed near it, as shown in the picture. Therefore, having depicted the floorboards with a series of almost parallel lines, which is in good agreement with visual perception, and almost without reducing the growth of the boyars as they are removed, the artist achieves the feeling that the far wall is wide enough. Moreover, the apparent parallelism makes one feel that the left side wall behind the boyars is as if parallel to the track axis and therefore a large group of boyars can stand there. This perception is connected with the habit of people to see in a certain way. From everyday experience, we know that in such a small and close room, its narrowing is not felt.

A discussion of the features of the three paintings shows that in the 19th century. artists do their best to mitigate the inevitable errors of perspective systems, especially when conveying close spaces (shallow interiors). This applies equally to the Renaissance system of perspective (Repin), and to the perceptual (Pimenov). Artists clearly feel that a strictly perspective construction too often gives the wrong visual effect that the artist seeks. What a contrast with the old European masters of the XV century! Those literally reveled in the opportunities that the newly acquired doctrine of perspective opened up for them in the image of long corridors, streets, deep halls.

They were little confused by the violation of natural visual perception (although many understood this and even described it). More important to them was the opportunity to create a spatial illusion, new for that time, to show the vast and deep space in the picture as a whole, and not as the geometry of isolated objects that were not connected by the geometry of space, as was done before.

If we summarize the methods used by artists to hide errors, especially relevant when transferring close space, then we can name the following four main tricks.

1. The use of a variant of the perceptual perspective system in which the inevitable errors are shifted to a plan that cannot be depicted. (A classic example is axonometry.)

2. The use of a variant of the perceptual perspective system in which errors are shifted to elements that are not in the plot to be depicted. (For example, errors are concentrated on the transmission of verticals, but on the picture of verticals - columns, etc. - no.)

3. If these options cannot be used, then it is possible to cut off distorted objects. (For example, when errors are shifted vertically and the image of the walls of the room is necessary - rejection of the image of the ceiling and the upper part of the walls, cutting off what will distort the visual perception by the upper edge of the picture.)

4. Disguise - obscuring various objects of the image of the most odious, visually unnatural, promising constructions.

The discussion in this chapter, strictly speaking, does not apply to the theory of perspective, but rather to the professional techniques used by artists. In this regard, it seems appropriate to mention another well-known professional technique, which now receives scientific justification.

Artists have long noticed that often a certain object, depicted by all the rules of the Renaissance perspective system, does not look natural enough. Its image can be significantly improved if you write according to the same rules, but moving away from it a greater distance, at the same time pretending that when transmitting the geometry of the visual perception of the object, no removal from it occurred. This highly effective technique can be explained as follows.

Let the artist not go anywhere to transfer the object. Having made sure that the renaissance version of the perspective system does not suit him, he can now take advantage of a whole set of variants of the perceptual perspective system, the existence of which he simply did not know before. A distinctive feature of the renaissance version of the scientific perspective is the preservation of similarities - that is, for example, the relationship between height and width in the visual perception of the object is preserved in the picture. Sometimes this seems to the artist important, and then from the whole set of options for the perceptual system of perspective, he should select those in which there is no similarity error. Among them, he can always find one that suits him and gives a reasonable image without changing the distance to the object.

The natural question that immediately arises is the question of specific methods of image according to the new rules of the perceptual perspective system. And here a very convenient circumstance is revealed. There is a class of perspective systems preserving similarity. Approximately they can be built according to the rules of the Renaissance system, but shifting the point of view - increasing the distance to the transmitted object. Therefore, preferring today a remote image to a close one, the artist, without knowing it, prefers some version of the perceptual system of perspective to the Renaissance. As shown by mathematical calculations, the required removal must sometimes be 2–2 times greater than the natural distance from which the picture is written.

Not only artists, but also photographers faced the discussed problem and the way to solve it. They also long ago discovered that a portrait taken from close range was much worse than the same portrait taken from unnaturally long range. That is why special portrait lenses appeared in photography. As you know, ordinary lenses, including portrait ones, provide images built according to the laws of the Renaissance perspective system. Consequently, there are no similarity errors in the photographs. According to the law of conservation of the total value of errors, their sum will be composed of scale transmission errors (violation of the natural size ratio in the close and deep plan; for a portrait, violation of the ratio of nose and ears in the photograph taken in full view) and depth transmission errors. To illustrate the foregoing, we give one numerical example for a specific object.

A photograph of a close object from a natural distance gives it a transmission error of scale equal to 64% of the sum of errors, and the remaining 36% is the error of transmitting depth. When doubling the distance, the ratio between errors changes - now the transmission error of the scale is 42%, and the depth is 58%. With a triple distance - 20% and 80%, respectively. The quadrupled distance gives in this particular example the complete disappearance of the transmission error of the scale. The whole amount of errors turned out to be concentrated on the transfer of depth. So, deletion leads to the transition of transmission errors of scale to errors of transmission of depth.

Which of these combinations of errors is less significant? What to prefer? The answer to this question does not lie in the field of perspective theory or mathematics. This is a matter of aesthetics. Experience shows that people prefer to reduce the transmission errors of the scale (relatively speaking, to improve the transmission of the ratio between the sizes of the nose and ears in the full-face photograph) and put up with the increase in the errors of transmission of depth (the image becomes flatter). It is important to note that a person looking at such photographs will always think that they are all taken from the same natural distance. If you put two photographs of the same size in front of him, on which the face of the person being portrayed is also the same size: one taken from a natural distance, and the other from a triple, he will see not that two images were taken from different distances, but that on the first the poor consistency of the size of the nose and ears is striking, and on the second this error is almost imperceptible (although due to the fact that the image is now perceived as more flat).

The foregoing applies not only to photography, but also to paintings. The somewhat vague art criticism characteristic of a close visual image, built according to the Renaissance rules, as violating the "usual visual proportions of an object" (compared to the remote one), now acquires clarity and quantitative certainty.