Do Scanline Filters Authentically Replicate a CRT Look or Are They a Gimmick

Ask the wrong question and you get the wrong answer. The wrong question is “do scanline filters look like a CRT?” Every flat panel fails that test under direct scrutiny. The right question is “what does a scanline filter CRT authenticity test measure, and which of those measurements matter to how you play?” The answer separates a gimmick from a tool.

A scanline filter is not a CRT. It is a simulated optical phenomenon that replicates one specific behaviour of a cathode ray tube: the visible dark lines between scan lines. They appear because the electron beam retraces during the blanking interval and the phosphor decays before the next pass. On a real CRT, those lines are not a filter. They are a physical artefact of how the screen draws an image line by line. On a flat panel, the filter recreates that artefact in the framebuffer. The measurable differences between the real thing and the simulation are lag, motion clarity, and light-gun support. Everything else is taste.

FirebrandX Profiles And The Reference Monitor

The reference look for accuracy comes from FirebrandX's CRT and HDR profiles used in RetroTINK and MiSTer. The benchmark against which every filter is measured is the Sony PVM-20L5. The standard tool for verifying geometry and convergence on both a real CRT and a filtered flat panel is the 240p test suite. That is the contract. What follows walks through what the filter does, what it cannot do, and where the line between accuracy and preference sits.

Mike Chi's RetroTINK 5X-Pro, released in 2021, is the reference lag-free scaler for retro consoles. It accepts composite, S-Video, component, and RGB SCART. It applies scanline filters in a line buffer with zero added input lag. That last point is the reason the 5X became the measuring stick.

Preset Names And The Parameters That Count

The scanline filter names in the 5X menu tell you exactly what the designer intended: PVM-600, PVM-600 Sharp, BVM, Consumer, Slot Mask, and Aperture Grille, with RGB variants of the PVM and BVM presets. The intensity range runs 0 to 255 per preset. The sub-options are Scanline Strength, Scanline Gamma, and Scanline Colour. The settings that matter are the mask type and the gamma. The aperture grille preset emulates a Trinitron's sub-pixel structure. The slot mask preset emulates a shadow mask CRT. The gamma setting changes how the scanline darkens the line below it, which is not a linear effect on a real CRT.

Set The Order And Verify

Set the mask type first, then the strength, then the gamma. Verify the result against the 240p test suite. The test suite's grid pattern and alternating horizontal lines reveal whether the filter is converging with the panel's native pixel grid. If the lines crawl or shimmer when you scroll, the scale is not integer and the mask is fighting the panel.

MiSTer FPGA uses the Adaptive Scanlines (ASL) framework, a set of filter coefficient tables stored as .txt files in the /Filters/ folder. Load the filter per input resolution via the MiSTer.ini [video] section. The preset names are explicit: GS_Sharpness_065, GS_Sharpness_080, GS_Sharpness_100, PVM, BVM, Consumer, Composite, S-Video, and VGA. The single most important parameter is Scanline Depth, on a 0-100 scale.

Mask Choice Depends On Your Panel

The scanline mask comparison on MiSTer is not about which filter is “better.” It is about which mask type matches the screen you are using. A PVM mask on a 4K OLED looks different from the same mask on a 1440p IPS monitor because the sub-pixel structure of the panel is different. The ASL filters are resolution-dependent. A filter that looks right at 1080p will alias at 4K.

The Practical Test

Load the test suite, scroll a horizontally-scrolling game, and watch the scanlines. If they shimmer or roll, the mask is not integer-aligned. The DE10-Nano FPGA board that runs MiSTer adds no lag from scanline generation. The filter is applied in the scaler pipeline with no frame buffer. That is the same zero-lag property as the OSSC, and it is why MiSTer remains a reference for accuracy.

The person who defined the target look for accurate CRT emulation is FirebrandX, a calibrator who produces the CRT and HDR profiles used in RetroTINK and MiSTer. These profiles are not arbitrary sliders. They are derived from measurements of reference monitors. They set the colour temperature, gamma, and scanline behaviour that the hardware should reproduce. FirebrandX's profile accuracy is the difference between a plausible approximation and a calibrated match.

The PVM-20L5 Baseline

The profiles are built around the characteristics of the Sony PVM-20L5, which supports 240p to 1080i. It is the benchmark against which scaler CRT filters are measured. The 20L5 has a specific TVL count, and the visibility of its scanlines is a function of that count: higher TVL means a sharper gap between lines. FirebrandX's profiles encode that behaviour.

RetroTINK 4K CRT Simulation Engine

The CRT Simulation engine uses FirebrandX-derived profiles with parameters for Mask Type, Mask Strength, Scanline Strength, Scanline Width, Phosphor Decay, Bloom, Halation, and Sub-pixel Layout. The sub-pixel layout options are RGB Stripe, RGB Triad, Delta, and PVM-style. That level of control is only meaningful if you understand what each parameter does. The profile is the starting point, not the end.

Artemio Urbina's 240p test suite, first released in 2015 and updated in 2023, is the standard measurement tool for lag, geometry, and colour accuracy. It runs on SNES, Genesis/MD, PC Engine, Dreamcast, Wii, GameCube, Neo Geo, and PlayStation. The scanline test pattern is a grid of alternating horizontal lines that reveals exactly what the scaler is doing to the image.

How To Read The Grid

For scanline calibration on a flat panel, load the grid pattern and look at the edges. If the lines are perfectly even and stable, the scanline filter is applied at the correct phase. If the lines crawl, shimmer, or pulse, the filter is not aligned to the panel's pixel grid.

Lag Numbers That Matter

The test suite also includes a lag test that measures the difference between the original signal and the scaled output. On the OSSC and RetroTINK 5X-Pro, scanline filters add zero milliseconds of lag. They are applied in the line buffer with no frame delay. The RetroTINK 4K adds less than 0.25 milliseconds, which is sub-frame processing and imperceptible. The MiSTer ASL filter also adds zero milliseconds. A scaler with overly aggressive sharpening, by contrast, adds ringing halos around sprites that were not present on a CRT. The test suite does not lie.

A scanline filter only affects the appearance of the image. It does nothing for motion clarity. It does nothing for input lag. Those are separate problems with separate solutions.

Motion Clarity And Black Frame Insertion

Motion clarity on a flat panel is limited by the panel's sample-and-hold behaviour. A scanline filter does not change that. Black frame insertion (BFI) is the screen-side solution that addresses motion blur. The RetroTINK 4K offers BFI modes at Off, 60Hz, 50Hz, and Variable. BFI works by inserting a black frame between game frames, which emulates the phosphor decay of a CRT and reduces motion blur. It is a different mechanism from a scanline filter, and it has a visible side effect: flicker. On a 60Hz BFI mode, the flicker is acceptable. On a 120Hz BFI mode, it is much less so.

Light Guns Need A CRT

Light-gun support is the third issue. A scanline filter does not help a light gun work on a flat panel. The gun reads the blanking interval of the screen, and a modern panel does not present the same signal timing as a CRT. The RetroTINK 5X-Pro and the RetroTINK 4K do not fix this. The only reliable light-gun solutions are a real CRT, a dedicated light-gun screen converter, or a solution like the Sinden Light Gun that uses a camera. If light-gun support is a requirement, skip the scanline filter conversation entirely.

The choice of scaler determines what scanline filter options you have and how they behave.

OSSC And OSSC Pro

The Open Source Scan Converter (OSSC) is a line multiplier with no framebuffer. Its lag is measured in microseconds. It produces razor-sharp pixels but has no composite or S-Video input. Its scanline alignment modes are Even, Odd, and Alternating, and the strength is configurable in 25% steps. The OSSC Pro adds a framebuffer, adaptive line multiplication, and HDMI input for modern consoles.

RetroTINK 5X-Pro And 4K

The RetroTINK 5X-Pro is a framebuffer scaler with composite and S-Video support. The RetroTINK 4K is the ceiling of the market, outputting 4K with HDR. The key difference between line doubling and framebuffer scaling is that line doubling repeats each line exactly with zero lag. Framebuffer scaling captures a full frame and resamples it, which enables rotation, zoom, and HDR filters but costs a small amount of lag.

Why Resolution Headroom Matters

Integer scaling mathematics matter here. A 240p image integer-scaled to 1080p uses a 4x scale to 960p with black borders. The RetroTINK 5X cannot output 4K, so its scanline masks are limited by the output resolution. The RetroTINK 4K has the headroom for finer masks and HDR brightness headroom, which changes how the scanlines interact with the panel's tone mapping.

Not all CRTs look the same. A scanline filter cannot emulate all of them at once. The difference between a Sony PVM-20L5 and a consumer Sony Trinitron KV-27FV310 is the TVL count. The PVM has a higher TVL count. The gap between scanlines is sharper and the image is more detailed. The consumer set has a softer gap because its 600 TVL count spreads the electron beam wider.

PVM Preset Versus Consumer Preset

The FirebrandX profiles target the PVM look because the PVM is the reference monitor in studios. The Consumer preset on the RetroTINK 5X-Pro targets the softer look of a consumer set. The difference is visible in the scanline test pattern: the PVM preset shows a dark, crisp line. The Consumer preset shows a lighter, wider line. You cannot have both.

Aperture Grille And Slot Mask

The mask presets on the 5X-Pro also differ. Aperture grille, used in Trinitrons, has vertical stripes and no horizontal shadow mask. That produces a brighter image with different scanline geometry. Slot mask, used in most other CRTs, has a grid of holes that let the beam through. The mask type changes how the scanlines look at the edges of the image, not just in the centre. The 240p test suite cannot tell you which mask to choose, but it will show you the difference.

Scanline filters handle the look. Black frame insertion handles the motion. On a CRT, the phosphor decays over time, so the image is gone before the next frame is drawn. On a flat panel, the image is held until the next frame. That causes motion blur and persistence. BFI solves this by inserting a black frame between game frames, emulating the dark period of phosphor decay.

Brightness Trade-Off

The RetroTINK 4K's BFI modes operate at 60Hz, 50Hz, and Variable. The effect is most dramatic on fast-scrolling games like Sonic or Castlevania. The trade-off is brightness. BFI cuts the light output roughly in half. If the screen does not have enough HDR brightness headroom, the image becomes too dark. The RetroTINK 4K's HDR CRT emulation compensates by raising the highlight brightness.

Two Systems, One Feel

The scanline filter and BFI work together, but they are not the same thing. A scanline filter without BFI looks like a static overlay. BFI without scanlines looks like a flickering image. The combination produces the CRT motion feel. The 240p test suite's response time test helps you verify that BFI is not introducing double images or artifacts.

The most common failure case is the mClassic. Marseille claims the mClassic outputs 1440p. What it actually outputs is a 1440p frame with a 1080p active image area that has been upscaled, not a native 1440p render. The mClassic does not generate scanlines at all. It is not a scanline filter device. If you bought an mClassic expecting a RetroTINK, you are in the wrong product line.

Analogue Pocket Dock And GBS-Control

The second failure case is the Analogue Pocket Dock. The Dock outputs 1080p, but the default scaling is not integer, which causes uneven pixel sizes on some cores. Enable integer scaling in the menu. It adds black borders but fixes the pixel geometry. The third failure case is the GBS-Control. Its motion adaptive deinterlacing mode works but exhibits chroma artifacts on fast-moving red objects. The firmware is open-source and updated frequently, so behaviour changes.

Capture Cards And The OSSC Input Gap

The fourth failure case is capture cards misreporting the EDID. The RetroTINK 5X-Pro outputs a clean 1080p signal. Some capture cards report the EDID incorrectly and record a 1080i file. If your recording looks interlaced, check the capture card settings before blaming the scaler. The last failure case is the OSSC's composite input gap. The OSSC accepts no composite or S-Video. If you have an NES or a Master System without modded RGB or component, the OSSC cannot help you. The RetroTINK 5X-Pro can.

The sub-pixel structure of a flat panel is not the same as that of a CRT. The scanline mask must account for the difference. An LCD panel uses an RGB Stripe layout by default. Some panels use RGB Triad or Delta. The RetroTINK 4K's Sub-pixel Layout options let you match the mask to the panel. The PVM-style option emulates the aperture grille of a Sony PVM, which has vertical stripes. Choose the wrong layout and the mask will show Moiré patterns on the test screen. The 240p test suite's grid pattern is the only way to see this.

Resolution And Soft Edges

The MiSTer ASL filters also have resolution-dependent behaviour. A filter designed for 1080p will alias at 4K. The practical effect: the scanline mask at 4K has more headroom to approximate the soft edge of the CRT's phosphor decay. At 1080p, the mask is a hard line, which looks harsher. The RetroTINK 4K is the only scaler that does this convincingly.

Analogue Pocket's Different Target

The Analogue Pocket's Original Display Mode takes a different approach. It simulates the sub-pixel structure of original handheld LCDs (Analogue GB, GBC, GBA) rather than CRTs, using third-party .json display mode files for OpenFPGA cores. That is a different target and a different aesthetic.

A scanline filter is a simulation of one physical phenomenon: the dark space between scan lines on a CRT. It is not a simulation of everything a CRT does. It does not change the panel's response time. It does not fix the sample-and-hold blur. It does not make a light gun work. What it does is change the distribution of bright and dark pixels on the screen. That affects how the eye tracks motion and how the image is perceived.

The measurable differences between a real CRT and a filtered flat panel are lag (the flat panel adds at least one frame for HDMI processing), motion clarity (BFI helps but does not match phosphor decay), and light-gun support (not possible without extra hardware). Everything else, the width of the scanline, the mask type, the gamma curve, is taste. The reference for taste is FirebrandX. The reference for measurement is the 240p test suite. If you cannot pass the test suite's grid pattern, the filter is not calibrated.

Do scanline filters add input lag? No. The OSSC and RetroTINK 5X-Pro apply scanlines in the line buffer with zero added lag. The RetroTINK 4K adds less than 0.25 ms, which is sub-frame. MiSTer ASL adds zero.

Can I use a scanline filter on a 4K OLED? Yes, but the mask must be integer-scaled to the panel's resolution. Verify with the 240p test suite. A 1080p mask on a 4K panel will not align.

What is the difference between PVM and Consumer scanline presets? PVM presets have a sharper, darker gap between lines due to higher TVL count. Consumer presets have a softer, wider gap. The RetroTINK 5X-Pro has both.

Does the mClassic generate scanlines? No. The mClassic upscales to 1440p but does not have a scanline generator. It is not a substitute for a RetroTINK or OSSC.

How do I know if my scanline filter is set correctly? Load the 240p test suite's grid pattern. If the lines are even and stable, it is correct. If they shimmer or crawl, the mask is not aligned.

Will a scanline filter make my light gun work? No. Light guns require the screen to output a blanking interval signal that a flat panel does not provide. BFI does not fix this.

What is the best scanline filter for a RetroTINK 5X-Pro? Start with the FirebrandX PVM-600 (RGB) profile. Adjust the gamma to your screen. Verify with the 240p test suite.

The scanline filter CRT authenticity test is not a single measurement but a suite of them: zero added lag on the OSSC and RetroTINK 5X-Pro, less than 0.25 ms on the RetroTINK 4K, and the hard fact that the Analogue Pocket Dock defaults to non-integer scaling that you must correct in the menu before any mask will align.