The straightforward answer is: On the vast majority of modern digital cameras, shooting directly with an appropriately high ISO generally

delivers better image quality and less noise than shooting at a low ISO and forcibly brightening the image in post-processing. 01 How Is Noise Generated? In digital cameras, photographs are created when the sensor (CMOS/CCD) captures light, converts it into electrical signals, and processes these

signals to form a digital image. Light = Signal Noise mainly originates from: Thermal electronic noise (naturally generated during sensor operation) Read noise (introduced by the circuit when reading pixel data, the main focus of this article) Dark current noise (produced during long exposures) Quantum efficiency limitations (each pixel cannot capture 100% of incoming photons) Chroma Noise It appears as random red, green, blue, or other colored speckles across the frame, ruining color purity and consistency. Visual perception: It looks unsightly and unnatural, seriously distracting from the main subject and greatly reducing the clean quality of the image.

It is the most annoying type of noise to the human eye and the top priority for noise reduction. Key trait: Chroma noise contains almost no useful image details; it is purely erroneous information. 02 The Essence of ISO and Signal Flow What is ISO? ISO (sensitivity) refers to the degree of amplification applied to the photoelectric signals captured by the sensor. Raising the ISO

means amplifying the analog signal **before** it enters the Analog-to-Digital Converter (ADC). Signal Flow of the Two Methods Method A (Shoot at High ISO): Light signal → Analog amplification (High ISO) → Analog-to-digital conversion → Digital image file Method B (Shoot at Low ISO, Brighten Later): Light signal → Analog amplification (Low ISO) → Analog-to-digital conversion → Digital amplification (Post-processing brightening) 03 Why Is High ISO Usually Better? 1. Underexposure at low ISO means the sensor receives very little total light, resulting in extremely weak signals for each color channel.

Worse still, during analog-to-digital conversion, the faint useful signal may only occupy a minimal bit depth of the ADC, leading to irreversible

information loss. Shadow areas turn pure black and cannot be recovered, while read noise becomes highly prominent. 2. When you forcibly brighten the entire image in post-processing, you amplify not only the weak useful signal but also the noise in every color

channel. More importantly, color information is almost completely lost in the shadow areas where the signal is weakest. 3. Cameras and RAW processing software have to guess the correct color of each pixel (demosaicing) based on noisy, unreliable data. This

process causes massive errors and an explosive surge of chroma noise. Noise generated this way tends to look far worse, accompanied by severe

color casts and detail loss, and is much harder to fix. 4. Shooting directly at a high ISO allows the camera to apply analog amplification before analog-to-digital conversion. This records color signals

at a much higher signal-to-noise ratio, providing more reliable and accurate color data during demosaicing, and drastically reducing chroma

noise in the final photo. What remains afterward is mostly luminance noise proportional to the signal intensity, which is far easier to remove with

noise reduction software. In short, amplifying the signal early in the imaging chain by raising the ISO preserves more authentic image data and suppresses read noise that

is difficult to eliminate in post-production. 04 Bonus: Types of Noise Luminance Noise It manifests as random fluctuations in image brightness, resembling an even layer of black-and-white grain. It appears uniformly across all color

channels and does not cause color casts. Visual perception: Similar to traditional film grain; it is even sometimes regarded as adding texture and atmosphere to an image, especially in

black-and-white photography. The human eye is relatively insensitive to pure brightness fluctuations, making this type of noise more tolerable. Key trait: Luminance noise is mixed with abundant genuine image details. Excessive noise reduction erases fine textures such as hair and surface

grain, leaving photos looking artificial, plasticky or over-smoothed like an oil painting. 05 Key Exceptions & Boundary Conditions The conclusion that high ISO performs better comes with important prerequisites and exceptions: Native ISO Range: Camera sensors are engineered with a set of native ISO values (typically base ISO 100/200 and higher values via internal

analog amplification). The above conclusion holds within this range. However, some cameras’ extended ISO only relies on digital

processing, which has the same effect as brightening or darkening in post and results in degraded image quality. Overexposure & Dynamic Range: High ISO reduces a camera’s dynamic range. Accidental overexposure at high ISO makes highlight details far

harder to recover. Shooting at low ISO, despite heavy noise when lifting shadows, preserves full recoverable highlight detail. Advancements in Modern Sensors: Newer cameras equipped with BSI (Back-Side Illuminated), stacked sensors and advanced image processors

have extremely well-controlled read noise at low ISO. The gap between the two methods is narrowing, yet the advantage of shooting at high ISO

still remains. Long Exposures with a Tripod: If you use a tripod, shooting at low ISO with a slow shutter speed for proper exposure is always the ultimate

choice for maximum image quality in low-light scenarios. 06 Practical Shooting Tips Follow the decision logic below to make the optimal choice for low-light photography.

Chroma Noise It appears as random speckles of red, green, blue and other colors, undermining color purity and consistency. Visual effect: It looks messy and unnatural, seriously distracting attention from the main subject and greatly reducing image clarity. It is the most

unpleasant type of noise to the human eye and needs priority removal. Information feature: Chroma noise contains almost no useful image details; it is purely erroneous pixel information. 05 Important Exceptions and Boundary Conditions The conclusion that a higher ISO delivers better results comes with crucial prerequisites and exceptions: Native ISO range of the camera: Camera sensors are designed with a set of native ISO values (typically base ISO such as 100 and 200, plus higher

values achieved via internal analog amplification). The above conclusion is valid within this range. However, the extended ISO on some cameras

only applies digital processing, which has the same effect as brightening or darkening images in post-production and leads to degraded image

quality. Overexposure and dynamic range: A high ISO reduces the camera’s dynamic range. This means if you accidentally overexpose at a high

ISO, highlight details are much harder to recover than at a low ISO. Shooting at a low ISO produces heavy noise when lifting shadows, yet it fully

preserves recoverable highlight details. Advancements in new-generation sensors: In recent years, modern cameras equipped with BSI (Back-Side Illuminated) and stacked

sensors, along with high-performance image processors, have achieved extremely well-controlled read noise at low ISO. For such cameras, the

gap between the two shooting methods is narrowing, yet the advantage of shooting directly at high ISO still remains. Long exposure with a tripod: If a tripod is available, using a low ISO combined with a slow shutter speed for proper exposure is always the best

choice for maximum image quality in low-light conditions. 06 Practical Shooting Recommendations You can follow the decision process below to make the optimal choice:

Logically speaking, a lower ISO usually delivers better image quality. However, when you have to shoot handheld in low light, you face a choice

raise the ISO directly while shooting to brighten the scene, or keep the ISO low and lift the brightness later in post-processing. Which of these two methods yields better image results?