By Olivier Chauvignat - 2025, dec 12
Conceptual foundations followed by application in Lightroom Classic
Definition refers to the total number of pixels that make up a digital image.
It is expressed in dimensions (for example, 6000 × 4000 pixels) or in megapixels.
Definition indicates how many pixels are available, but on its own it says nothing about the image’s ability to describe fine detail.
Today, a 24-megapixel camera is commonly considered a “baseline definition.”
In reality, it is more accurate to describe it as a standard definition.
A camera between 40 and 60 megapixels is considered high definition.
A camera with 100 megapixels or more is considered very high definition.
Resolution refers to the number of points available to resolve an image over a given surface area.
In film photography, this corresponds to the number of silver grains per unit area.
In digital photography, it corresponds to the number of pixels per unit area (per square centimeter, per square inch, etc.).
In other words, resolution describes the spatial sampling grid available to represent the image.
The more points there are per unit area, the more closely spaced details can be resolved.
Resolution primarily concerns output devices: printers, screens, EVFs, and so on.
Using the term “resolution” to refer to the number of pixels is incorrect.
One should speak of definition, not resolution.
The resolving power of a lens refers to its ability to exploit the available resolution, in other words its capacity to separate and reproduce closely spaced details.
An image may have high resolution without the lens being capable of fully exploiting it.
In this text, the French term piqué is used.
It is a French term with no direct equivalent in English.
Pronunciation: piqué is pronounced [pee-kay], with a final “é” sound.
Piqué refers to the resolving power of a lens, that is, its ability to exploit the available resolution and to reproduce a large amount of information, including in out-of-focus areas.
Piqué must not be confused with Netteté [neh-teh-tay] (focus point and rendering at this point), nor with what English broadly groups under the term sharpness.
Much of today’s confusion surrounding lenses and image quality comes from a mixing of concepts that actually describe very different phenomena. This confusion is particularly visible in English-language content, but it also exists in French. It is therefore essential to establish a clear framework.
Piqué corresponds to the amount of information a lens is capable of reproducing, and to the way this information is organized within the image.
It is not about where focus is placed, but about what information the lens reproduces, and how that information persists throughout the image.
Fundamental point: piqué is not limited to the plane of focus.
With a genuinely high-piqué lens, out-of-focus areas remain readable, structured, and coherent.
Even though they are optically blurred, they retain such a high density of information that they may appear sharp, when in fact they are simply out of focus but highly resolved.
Conversely, a lens with poor piqué produces out-of-focus areas that are flat, structureless, and devoid of usable information.
Piqué therefore answers a precise question:
How much information is the lens capable of reproducing, including where the image is not focused?
A sharp photograph is one in which focus is placed exactly where it was intended.
Sharpness does not describe the overall quality of an image, but the accuracy of focus relative to intent.
In French, netteté encompasses two inseparable elements:
– where focus is placed, and
– the way the lens renders sharpness on the plane of focus, either naturally or in an exaggerated manner, relative to what a human observer with 10/10 vision in each eye would see.
Sharpness depends neither on depth of field nor on focusing distance.
It depends on focus accuracy and on the lens’s ability to reproduce that focus.
The correct question is therefore:
Is focus placed where it should be, and is that area rendered naturally or exaggerated in terms of sharpness?
An image is over-sharpened when the resulting sharpness exceeds what a human observer with 10/10 vision in each eye would naturally perceive.
This represents an exceedance of a physiological reference, generally produced by processing rather than by any real gain in information.
A photograph is sharp in only one place: the exact plane where focus is set.
This property follows directly from the principle of focusing:
light rays originating from a point in the scene can converge at one point in image space, and only one.
It is physically impossible for these rays to converge perfectly at multiple distances at the same time.
The plane of focus lies within the depth-of-field zone.
It is the only location where sharpness is absolute.
👉 Depth of field is not a zone of sharpness, but a zone of acceptable sharpness.
Zone focusing relies precisely on this notion of acceptable sharpness.
Even then, the subject is strictly sharp only in the plane of focus.
In English, the term sharpness indiscriminately refers to focus, sharpness on the plane of focus, the way that sharpness is rendered, software sharpening, and sometimes even resolving power.
The problem is not a general lack of vocabulary, but the absence of the concept of piqué in English.
Lacking a distinct term, sharpness is used for everything, leading to structural confusion.
Sharpness—and even more so over-sharpening—is frequently presented as a criterion of optical quality.
This idea is false.
A lens can be sharp or over-sharp and still be of poor quality in terms of information reproduction, particularly if it has very weak piqué.
Sharpness has become a marketing argument used to mask a lack of resolving power.
Being sharp does not mean being good.
The analysis of bokeh balls is probably one of the greatest contemporary intellectual scams concerning lenses. It is omnipresent on YouTube and social networks, endlessly commented on, compared, dissected—and yet entirely irrelevant for real lens evaluation.
The principle is flawed at its core. Bokeh balls correspond to areas that lie entirely beyond the resolving power of the lens. There is no longer any image, no reproduction, no detail—only luminous patches without usable information. In other words, one claims to analyze a lens precisely where it resolves nothing.
This extreme out-of-focus zone therefore provides no insight into piqué, fineness, texture retention, or quality of reproduction. Blur here is no longer resolving; it contains no structured detail and is reduced to a simplified geometric shape. Analyzing a lens at this point is like judging a reproduction system based on the total absence of information.
Added to this is a rarely questioned artificial constraint: to observe bokeh balls, one needs isolated, point-like light sources in the background, sufficiently contrasted. Such situations are neither common nor representative of real photographic use. Who seriously chooses lenses based on blurred fairy lights or specular highlights filling the background?
This fascination with bokeh balls distracts attention from what truly matters. A lens should not be judged by the shape of an empty luminous disk, but by how it reproduces information, including out of focus: foliage, fabrics, hair, complex backgrounds. It is in these blurred yet still readable areas that true optical behavior is revealed—not in detail-less patches.
Bokeh balls are therefore neither a quality criterion nor a relevant decision tool. They belong more to visual fetishism—easy to show and comment on—than to any grounded photographic analysis. Their current popularity rests on their spectacular appearance, not on real usefulness.
Lens tests on flat charts rest on a real scientific basis. They measure what happens on the exact plane of focus, on a perfectly flat subject, perpendicular to the optical axis. In this sense, these tests are not wrong. The issue is not scientific—it is practical.
A portrait photographer does not photograph flat surfaces. They photograph faces, which are volumes. A face is never contained within a single plane: eyes, nose, cheekbones, and ears lie at slightly different distances. Yet everything that makes a lens interesting for portrait photography happens precisely around the plane of focus, in slightly out-of-focus zones.
That is exactly what charts do not measure. They say nothing about the sharp-to-blur transition, volume rendering, texture readability out of focus, or the overall rendering of a face. They describe a situation that practically never exists in real portrait photography.
Moreover, many analyses emphasize differences in definition in the corners of the image. In practice, this information is irrelevant: a portrait photographer does not place their subject’s head in the lower-right corner to judge lens quality there.
We therefore encounter a situation comparable to that of bokeh balls. In both cases, analysis is based on a real phenomenon, but outside any meaningful usage context. The lens is observed where it has no impact on real photography, and general conclusions are then drawn.
Charts can help detect a defective or grossly flawed lens. However, choosing a portrait lens based on flat charts amounts to confusing instrumental measurement with real photographic quality.
In the daily practice of a portrait photographer, this type of analysis is therefore, quite simply, useless.
Grain and noise are fundamentally different concepts and must not be confused.
Grain is a structure. Historically, it corresponds to the size and distribution of silver grains in a photographic emulsion. It has organization, visual coherence, and an identifiable aesthetic. Grain fully participates in photographic rendering: it structures the image, influences the perception of sharpness, resolution, and overall character. It also represents the plastic dimension of photography, bringing a natural smoothing—in the noble sense of the term—that softens transitions, tempers excessive digital harshness, and restores visual materiality.
Noise, by contrast, is an artifact. In digital photography, it results from electronic phenomena related to the sensor and signal amplification. It is random, unstructured, and unrelated to any photographic language inherited from film.
This distinction is now blurred by frequent misuse of language, particularly in video content. Many creators describe images or videos as “grainy” when they are in fact noisy. The correct term would be “noisy.” Using “grainy” to describe noise amounts to confusing a technical defect with an aesthetic choice, perpetuating lasting confusion.
The Grain module in Lightroom Classic does not work on noise. It creates and modulates a granular structure inspired by real photographic processes and must be understood as such.
The Grain module in Lightroom Classic is one of the most intelligent modules in the software—and paradoxically one of the most underestimated.
It is based on a coherent modeling of real physical phenomena derived from film photography, without any use of artificial intelligence.
This module allows one to act on the very structure of the image, on how resolution is exploited, and therefore on the perception of sharpness and piqué.
Amount controls the overall presence of granular structure in the image.
It determines the proportion of visible grain relative to fine information.
It is the “strength” of the effect.
Size corresponds to the size of silver grains and directly relates to ISO sensitivity in film photography.
On low-sensitivity film, grains are small and numerous: resolution is high.
On high-sensitivity film, grains are larger and fewer: resolution drops.
As grain size increases, there are fewer grains over the same surface area, and therefore fewer points available to resolve the image. This results in a reduction in resolution, which mechanically leads to a loss of fineness and sharpness.
This loss of sharpness produces what is deliberately described as a “film blur”:
it is not a focus blur, but a loss of resolving power linked to the structure of the medium.
It is precisely this behavior that allows excessive, harsh, or artificial sharpness to be broken down.
Roughness controls the shape, roundness, and spatial behavior of the grains.
This slider does not change their size, but their regularity and variability.
At low Roughness, grains are round and homogeneous.
At high Roughness, grains are deconstructed: the “grain matter” becomes more irregular and more chaotic.
A credible simulation of film rendering relies on the combination of Size and Roughness.
Fine, regular grain typically corresponds to slide film or low-sensitivity film.
More pronounced grain evokes medium- to high-sensitivity color or black-and-white negatives.
Very large and highly rough grain approaches processes such as Polaroid or heavily pushed films.
The most direct use of the Grain module is to reduce resolution while accepting visible grain. It soften the rendering and attenuate excessive sharpness.
The logic is strictly film-based: by increasing grain size and irregularity, the number of points available to resolve the image is reduced. This results in a drop in resolution, accompanied by a clearly perceptible loss of fineness and sharpness.
In this configuration, Amount is increased to make the grain visible, Size is increased to simulate larger grains, and Roughness is adjusted to define the irregular character of the structure.
This behavior produces what is simplistically referred to as “film blur”: a loss of sharpness linked to the structure of the medium, not to focus.
Here, grain is an assumed aesthetic element and directly contributes to the final rendering.
It is by contrast with this obvious use that the following method becomes particularly interesting: reducing resolution without making grain truly visible.
The Grain module in Lightroom Classic can be used to lower the perceived resolution of an image while keeping grain almost invisible. The goal here is not to create an obvious grain effect, but to induce a sufficient loss of resolving power to soften the rendering and attenuate excessive sharpness.
The adjustment logic relies on a precise balance between the three sliders. Amount is kept low, around 20%, so that the granular structure does not impose itself visually. Roughness is also set to low values, generally between 7 and 10%, in order to maintain a regular, discreet structure. The main role is then assigned to the Size slider, adjusted according to the desired effect and the file’s definition. Values between 40 and 60% most often allow a perceptible drop in resolution without the grain becoming obvious.
It is essential to evaluate the result on the exported JPEG, not only within Lightroom’s interface, because the final output (resizing, compression, distribution chain) strongly influences the real perception of fineness and sharpness.
Adding grain is sometimes presented as a loss of quality, especially when applied to files from high-definition or medium-format sensors. This interpretation is incorrect.
This is in no way a loss of quality, but a rendering choice. Grain, like softness, reduced sharpness, or vintage rendering, is an integral part of photographic history and culture. These renderings are neither obsolete nor devalued: they are fully embraced and still used today, including in very high-level publications.
Major magazines such as Vogue and other international titles continue to publish images with visible, sometimes pronounced grain—not due to technical constraints, but as an aesthetic choice grounded in photographic culture. Grain is not a defect to be corrected; it is a visual language.
Speaking of a “loss of quality” amounts to confusing maximum technical performance with artistic intent. An image can be deliberately less sharp, less resolved, or more grainy, while still being of very high photographic quality.